Training in progress, step 525, checkpoint
Browse files- checkpoint-525/README.md +202 -0
- checkpoint-525/adapter_config.json +34 -0
- checkpoint-525/adapter_model.safetensors +3 -0
- checkpoint-525/added_tokens.json +17 -0
- checkpoint-525/optimizer.pt +3 -0
- checkpoint-525/rng_state_0.pth +3 -0
- checkpoint-525/rng_state_1.pth +3 -0
- checkpoint-525/rng_state_2.pth +3 -0
- checkpoint-525/rng_state_3.pth +3 -0
- checkpoint-525/rng_state_4.pth +3 -0
- checkpoint-525/rng_state_5.pth +3 -0
- checkpoint-525/rng_state_6.pth +3 -0
- checkpoint-525/rng_state_7.pth +3 -0
- checkpoint-525/scheduler.pt +3 -0
- checkpoint-525/special_tokens_map.json +30 -0
- checkpoint-525/tokenizer.json +0 -0
- checkpoint-525/tokenizer.model +3 -0
- checkpoint-525/tokenizer_config.json +163 -0
- checkpoint-525/trainer_state.json +3820 -0
- checkpoint-525/training_args.bin +3 -0
checkpoint-525/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: microsoft/Phi-3-mini-4k-instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.11.1
|
checkpoint-525/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "microsoft/Phi-3-mini-4k-instruct",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": null,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 32,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": [
|
| 18 |
+
"embed_tokens",
|
| 19 |
+
"lm_head"
|
| 20 |
+
],
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 512,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"o_proj",
|
| 27 |
+
"qkv_proj",
|
| 28 |
+
"down_proj",
|
| 29 |
+
"gate_up_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
checkpoint-525/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8cf56c2e61a26295d78e2b0bb16737db863c683e8550003269dc7344fb4b3396
|
| 3 |
+
size 2004651224
|
checkpoint-525/added_tokens.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</plan>": 32014,
|
| 3 |
+
"</thinking>": 32012,
|
| 4 |
+
"<plan>": 32013,
|
| 5 |
+
"<thinking>": 32011,
|
| 6 |
+
"<|assistant|>": 32001,
|
| 7 |
+
"<|endoftext|>": 32000,
|
| 8 |
+
"<|end|>": 32007,
|
| 9 |
+
"<|placeholder1|>": 32002,
|
| 10 |
+
"<|placeholder2|>": 32003,
|
| 11 |
+
"<|placeholder3|>": 32004,
|
| 12 |
+
"<|placeholder4|>": 32005,
|
| 13 |
+
"<|placeholder5|>": 32008,
|
| 14 |
+
"<|placeholder6|>": 32009,
|
| 15 |
+
"<|system|>": 32006,
|
| 16 |
+
"<|user|>": 32010
|
| 17 |
+
}
|
checkpoint-525/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b07636d9fc0785dfcd0d01c600be5796324d76abd81e3071062a1fd2aa657f4d
|
| 3 |
+
size 4009448634
|
checkpoint-525/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11c96bca6128b3dacd7258db3cdab95978bdd0f07a323e97740f8edab7ac003c
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aee047011d5d66e6de4882f1c295e5cf184e11f66d730c9040f3c5cd5e2035d5
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c87b33322114c9c85efcfa061fffcc1288219de8c8fd736fdc13117a9a6ecef4
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eff9112a86c73ea29b4297d4535d6697a4a3e7cebb6dd21e444a52c76f343b84
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e01130834a6a01d224b4d7d1eae0ce77a7da0cd91eba9888f1cff32b168f6db
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e35710bdda61e1a48d4f6523ad37e357a7155270c452741208573ed996e728ea
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0dc447d3f896e2fb468daa43f2904140e72f13642e97ed9b4ea7f81062b277a
|
| 3 |
+
size 15984
|
checkpoint-525/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a60d7b390e009589f883c7b4ed50aec7ca4e1bcab1ce98a325906493ecb66eb
|
| 3 |
+
size 15984
|
checkpoint-525/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:553c3da6bc17b4563fe8deab838eccddb1337f6f4d65ba573e34dcf80cb8b71f
|
| 3 |
+
size 1064
|
checkpoint-525/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
checkpoint-525/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-525/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-525/tokenizer_config.json
ADDED
|
@@ -0,0 +1,163 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": true,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": false
|
| 29 |
+
},
|
| 30 |
+
"32000": {
|
| 31 |
+
"content": "<|endoftext|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"32001": {
|
| 39 |
+
"content": "<|assistant|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": true,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
},
|
| 46 |
+
"32002": {
|
| 47 |
+
"content": "<|placeholder1|>",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": true,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": true
|
| 53 |
+
},
|
| 54 |
+
"32003": {
|
| 55 |
+
"content": "<|placeholder2|>",
|
| 56 |
+
"lstrip": false,
|
| 57 |
+
"normalized": false,
|
| 58 |
+
"rstrip": true,
|
| 59 |
+
"single_word": false,
|
| 60 |
+
"special": true
|
| 61 |
+
},
|
| 62 |
+
"32004": {
|
| 63 |
+
"content": "<|placeholder3|>",
|
| 64 |
+
"lstrip": false,
|
| 65 |
+
"normalized": false,
|
| 66 |
+
"rstrip": true,
|
| 67 |
+
"single_word": false,
|
| 68 |
+
"special": true
|
| 69 |
+
},
|
| 70 |
+
"32005": {
|
| 71 |
+
"content": "<|placeholder4|>",
|
| 72 |
+
"lstrip": false,
|
| 73 |
+
"normalized": false,
|
| 74 |
+
"rstrip": true,
|
| 75 |
+
"single_word": false,
|
| 76 |
+
"special": true
|
| 77 |
+
},
|
| 78 |
+
"32006": {
|
| 79 |
+
"content": "<|system|>",
|
| 80 |
+
"lstrip": false,
|
| 81 |
+
"normalized": false,
|
| 82 |
+
"rstrip": true,
|
| 83 |
+
"single_word": false,
|
| 84 |
+
"special": true
|
| 85 |
+
},
|
| 86 |
+
"32007": {
|
| 87 |
+
"content": "<|end|>",
|
| 88 |
+
"lstrip": false,
|
| 89 |
+
"normalized": false,
|
| 90 |
+
"rstrip": true,
|
| 91 |
+
"single_word": false,
|
| 92 |
+
"special": true
|
| 93 |
+
},
|
| 94 |
+
"32008": {
|
| 95 |
+
"content": "<|placeholder5|>",
|
| 96 |
+
"lstrip": false,
|
| 97 |
+
"normalized": false,
|
| 98 |
+
"rstrip": true,
|
| 99 |
+
"single_word": false,
|
| 100 |
+
"special": true
|
| 101 |
+
},
|
| 102 |
+
"32009": {
|
| 103 |
+
"content": "<|placeholder6|>",
|
| 104 |
+
"lstrip": false,
|
| 105 |
+
"normalized": false,
|
| 106 |
+
"rstrip": true,
|
| 107 |
+
"single_word": false,
|
| 108 |
+
"special": true
|
| 109 |
+
},
|
| 110 |
+
"32010": {
|
| 111 |
+
"content": "<|user|>",
|
| 112 |
+
"lstrip": false,
|
| 113 |
+
"normalized": false,
|
| 114 |
+
"rstrip": true,
|
| 115 |
+
"single_word": false,
|
| 116 |
+
"special": true
|
| 117 |
+
},
|
| 118 |
+
"32011": {
|
| 119 |
+
"content": "<thinking>",
|
| 120 |
+
"lstrip": false,
|
| 121 |
+
"normalized": false,
|
| 122 |
+
"rstrip": false,
|
| 123 |
+
"single_word": false,
|
| 124 |
+
"special": false
|
| 125 |
+
},
|
| 126 |
+
"32012": {
|
| 127 |
+
"content": "</thinking>",
|
| 128 |
+
"lstrip": false,
|
| 129 |
+
"normalized": false,
|
| 130 |
+
"rstrip": false,
|
| 131 |
+
"single_word": false,
|
| 132 |
+
"special": false
|
| 133 |
+
},
|
| 134 |
+
"32013": {
|
| 135 |
+
"content": "<plan>",
|
| 136 |
+
"lstrip": false,
|
| 137 |
+
"normalized": false,
|
| 138 |
+
"rstrip": false,
|
| 139 |
+
"single_word": false,
|
| 140 |
+
"special": false
|
| 141 |
+
},
|
| 142 |
+
"32014": {
|
| 143 |
+
"content": "</plan>",
|
| 144 |
+
"lstrip": false,
|
| 145 |
+
"normalized": false,
|
| 146 |
+
"rstrip": false,
|
| 147 |
+
"single_word": false,
|
| 148 |
+
"special": false
|
| 149 |
+
}
|
| 150 |
+
},
|
| 151 |
+
"bos_token": "<s>",
|
| 152 |
+
"chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'system') %}{{'<|system|>' + '\n' + message['content'] + '<|end|>' + '\n'}}{% elif (message['role'] == 'user') %}{{'<|user|>' + '\n' + message['content'] + '<|end|>' + '\n' + '<|assistant|>' + '\n'}}{% elif message['role'] == 'assistant' %}{{message['content'] + '<|end|>' + '\n'}}{% endif %}{% endfor %}",
|
| 153 |
+
"clean_up_tokenization_spaces": false,
|
| 154 |
+
"eos_token": "<|endoftext|>",
|
| 155 |
+
"legacy": false,
|
| 156 |
+
"model_max_length": 4096,
|
| 157 |
+
"pad_token": "<|endoftext|>",
|
| 158 |
+
"padding_side": "left",
|
| 159 |
+
"sp_model_kwargs": {},
|
| 160 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 161 |
+
"unk_token": "<unk>",
|
| 162 |
+
"use_default_system_prompt": false
|
| 163 |
+
}
|
checkpoint-525/trainer_state.json
ADDED
|
@@ -0,0 +1,3820 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.7676767676767677,
|
| 5 |
+
"eval_steps": 38,
|
| 6 |
+
"global_step": 525,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.003367003367003367,
|
| 13 |
+
"grad_norm": 2.0,
|
| 14 |
+
"learning_rate": 4.5454545454545457e-07,
|
| 15 |
+
"loss": 1.0833,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003367003367003367,
|
| 20 |
+
"eval_loss": 1.032954454421997,
|
| 21 |
+
"eval_runtime": 8.7786,
|
| 22 |
+
"eval_samples_per_second": 56.957,
|
| 23 |
+
"eval_steps_per_second": 3.645,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.006734006734006734,
|
| 28 |
+
"grad_norm": 2.265625,
|
| 29 |
+
"learning_rate": 9.090909090909091e-07,
|
| 30 |
+
"loss": 0.996,
|
| 31 |
+
"step": 2
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.010101010101010102,
|
| 35 |
+
"grad_norm": 1.6484375,
|
| 36 |
+
"learning_rate": 1.3636363636363636e-06,
|
| 37 |
+
"loss": 1.0487,
|
| 38 |
+
"step": 3
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.013468013468013467,
|
| 42 |
+
"grad_norm": 2.046875,
|
| 43 |
+
"learning_rate": 1.8181818181818183e-06,
|
| 44 |
+
"loss": 1.0554,
|
| 45 |
+
"step": 4
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.016835016835016835,
|
| 49 |
+
"grad_norm": 2.4375,
|
| 50 |
+
"learning_rate": 2.2727272727272728e-06,
|
| 51 |
+
"loss": 1.0341,
|
| 52 |
+
"step": 5
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.020202020202020204,
|
| 56 |
+
"grad_norm": 1.9609375,
|
| 57 |
+
"learning_rate": 2.7272727272727272e-06,
|
| 58 |
+
"loss": 1.0132,
|
| 59 |
+
"step": 6
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.02356902356902357,
|
| 63 |
+
"grad_norm": 2.625,
|
| 64 |
+
"learning_rate": 3.181818181818182e-06,
|
| 65 |
+
"loss": 1.0298,
|
| 66 |
+
"step": 7
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.026936026936026935,
|
| 70 |
+
"grad_norm": 2.296875,
|
| 71 |
+
"learning_rate": 3.6363636363636366e-06,
|
| 72 |
+
"loss": 1.0216,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.030303030303030304,
|
| 77 |
+
"grad_norm": 2.234375,
|
| 78 |
+
"learning_rate": 4.0909090909090915e-06,
|
| 79 |
+
"loss": 1.0736,
|
| 80 |
+
"step": 9
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.03367003367003367,
|
| 84 |
+
"grad_norm": 2.1875,
|
| 85 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 86 |
+
"loss": 1.0216,
|
| 87 |
+
"step": 10
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.037037037037037035,
|
| 91 |
+
"grad_norm": 2.359375,
|
| 92 |
+
"learning_rate": 5e-06,
|
| 93 |
+
"loss": 1.0075,
|
| 94 |
+
"step": 11
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.04040404040404041,
|
| 98 |
+
"grad_norm": 1.890625,
|
| 99 |
+
"learning_rate": 5.4545454545454545e-06,
|
| 100 |
+
"loss": 1.051,
|
| 101 |
+
"step": 12
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.04377104377104377,
|
| 105 |
+
"grad_norm": 1.90625,
|
| 106 |
+
"learning_rate": 5.90909090909091e-06,
|
| 107 |
+
"loss": 1.0655,
|
| 108 |
+
"step": 13
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.04713804713804714,
|
| 112 |
+
"grad_norm": 1.6484375,
|
| 113 |
+
"learning_rate": 6.363636363636364e-06,
|
| 114 |
+
"loss": 1.0297,
|
| 115 |
+
"step": 14
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.050505050505050504,
|
| 119 |
+
"grad_norm": 1.6953125,
|
| 120 |
+
"learning_rate": 6.818181818181818e-06,
|
| 121 |
+
"loss": 1.0411,
|
| 122 |
+
"step": 15
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.05387205387205387,
|
| 126 |
+
"grad_norm": 1.9921875,
|
| 127 |
+
"learning_rate": 7.272727272727273e-06,
|
| 128 |
+
"loss": 1.0618,
|
| 129 |
+
"step": 16
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.05723905723905724,
|
| 133 |
+
"grad_norm": 1.4453125,
|
| 134 |
+
"learning_rate": 7.727272727272727e-06,
|
| 135 |
+
"loss": 1.0052,
|
| 136 |
+
"step": 17
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.06060606060606061,
|
| 140 |
+
"grad_norm": 1.390625,
|
| 141 |
+
"learning_rate": 8.181818181818183e-06,
|
| 142 |
+
"loss": 1.0053,
|
| 143 |
+
"step": 18
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.06397306397306397,
|
| 147 |
+
"grad_norm": 1.2734375,
|
| 148 |
+
"learning_rate": 8.636363636363637e-06,
|
| 149 |
+
"loss": 0.9786,
|
| 150 |
+
"step": 19
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.06734006734006734,
|
| 154 |
+
"grad_norm": 1.234375,
|
| 155 |
+
"learning_rate": 9.090909090909091e-06,
|
| 156 |
+
"loss": 1.0419,
|
| 157 |
+
"step": 20
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.0707070707070707,
|
| 161 |
+
"grad_norm": 1.1484375,
|
| 162 |
+
"learning_rate": 9.545454545454547e-06,
|
| 163 |
+
"loss": 1.0356,
|
| 164 |
+
"step": 21
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.07407407407407407,
|
| 168 |
+
"grad_norm": 1.28125,
|
| 169 |
+
"learning_rate": 1e-05,
|
| 170 |
+
"loss": 0.9778,
|
| 171 |
+
"step": 22
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.07744107744107744,
|
| 175 |
+
"grad_norm": 1.1875,
|
| 176 |
+
"learning_rate": 1.0454545454545455e-05,
|
| 177 |
+
"loss": 0.9691,
|
| 178 |
+
"step": 23
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.08080808080808081,
|
| 182 |
+
"grad_norm": 1.1953125,
|
| 183 |
+
"learning_rate": 1.0909090909090909e-05,
|
| 184 |
+
"loss": 1.0408,
|
| 185 |
+
"step": 24
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.08417508417508418,
|
| 189 |
+
"grad_norm": 1.078125,
|
| 190 |
+
"learning_rate": 1.1363636363636366e-05,
|
| 191 |
+
"loss": 0.9799,
|
| 192 |
+
"step": 25
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.08754208754208755,
|
| 196 |
+
"grad_norm": 1.0390625,
|
| 197 |
+
"learning_rate": 1.181818181818182e-05,
|
| 198 |
+
"loss": 0.9756,
|
| 199 |
+
"step": 26
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.09090909090909091,
|
| 203 |
+
"grad_norm": 1.1875,
|
| 204 |
+
"learning_rate": 1.2272727272727274e-05,
|
| 205 |
+
"loss": 1.0451,
|
| 206 |
+
"step": 27
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.09427609427609428,
|
| 210 |
+
"grad_norm": 1.0546875,
|
| 211 |
+
"learning_rate": 1.2727272727272728e-05,
|
| 212 |
+
"loss": 1.0093,
|
| 213 |
+
"step": 28
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.09764309764309764,
|
| 217 |
+
"grad_norm": 0.96484375,
|
| 218 |
+
"learning_rate": 1.3181818181818183e-05,
|
| 219 |
+
"loss": 0.99,
|
| 220 |
+
"step": 29
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.10101010101010101,
|
| 224 |
+
"grad_norm": 1.0703125,
|
| 225 |
+
"learning_rate": 1.3636363636363637e-05,
|
| 226 |
+
"loss": 1.0079,
|
| 227 |
+
"step": 30
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.10437710437710437,
|
| 231 |
+
"grad_norm": 0.89453125,
|
| 232 |
+
"learning_rate": 1.4090909090909092e-05,
|
| 233 |
+
"loss": 0.9653,
|
| 234 |
+
"step": 31
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.10774410774410774,
|
| 238 |
+
"grad_norm": 1.0546875,
|
| 239 |
+
"learning_rate": 1.4545454545454546e-05,
|
| 240 |
+
"loss": 0.9926,
|
| 241 |
+
"step": 32
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.1111111111111111,
|
| 245 |
+
"grad_norm": 0.96875,
|
| 246 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 247 |
+
"loss": 0.9857,
|
| 248 |
+
"step": 33
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.11447811447811448,
|
| 252 |
+
"grad_norm": 1.1484375,
|
| 253 |
+
"learning_rate": 1.5454545454545454e-05,
|
| 254 |
+
"loss": 0.9676,
|
| 255 |
+
"step": 34
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.11784511784511785,
|
| 259 |
+
"grad_norm": 1.078125,
|
| 260 |
+
"learning_rate": 1.590909090909091e-05,
|
| 261 |
+
"loss": 0.9686,
|
| 262 |
+
"step": 35
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.12121212121212122,
|
| 266 |
+
"grad_norm": 0.8203125,
|
| 267 |
+
"learning_rate": 1.6363636363636366e-05,
|
| 268 |
+
"loss": 0.9182,
|
| 269 |
+
"step": 36
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.12457912457912458,
|
| 273 |
+
"grad_norm": 0.8515625,
|
| 274 |
+
"learning_rate": 1.681818181818182e-05,
|
| 275 |
+
"loss": 0.9506,
|
| 276 |
+
"step": 37
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.12794612794612795,
|
| 280 |
+
"grad_norm": 0.86328125,
|
| 281 |
+
"learning_rate": 1.7272727272727274e-05,
|
| 282 |
+
"loss": 0.9533,
|
| 283 |
+
"step": 38
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.12794612794612795,
|
| 287 |
+
"eval_loss": 0.9329763054847717,
|
| 288 |
+
"eval_runtime": 8.7095,
|
| 289 |
+
"eval_samples_per_second": 57.409,
|
| 290 |
+
"eval_steps_per_second": 3.674,
|
| 291 |
+
"step": 38
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.13131313131313133,
|
| 295 |
+
"grad_norm": 0.7890625,
|
| 296 |
+
"learning_rate": 1.772727272727273e-05,
|
| 297 |
+
"loss": 0.9225,
|
| 298 |
+
"step": 39
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.13468013468013468,
|
| 302 |
+
"grad_norm": 0.890625,
|
| 303 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 304 |
+
"loss": 0.9378,
|
| 305 |
+
"step": 40
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.13804713804713806,
|
| 309 |
+
"grad_norm": 1.046875,
|
| 310 |
+
"learning_rate": 1.8636363636363638e-05,
|
| 311 |
+
"loss": 0.8982,
|
| 312 |
+
"step": 41
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.1414141414141414,
|
| 316 |
+
"grad_norm": 0.94921875,
|
| 317 |
+
"learning_rate": 1.9090909090909094e-05,
|
| 318 |
+
"loss": 0.8958,
|
| 319 |
+
"step": 42
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.1447811447811448,
|
| 323 |
+
"grad_norm": 0.9765625,
|
| 324 |
+
"learning_rate": 1.9545454545454546e-05,
|
| 325 |
+
"loss": 0.8796,
|
| 326 |
+
"step": 43
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.14814814814814814,
|
| 330 |
+
"grad_norm": 0.92578125,
|
| 331 |
+
"learning_rate": 2e-05,
|
| 332 |
+
"loss": 0.8808,
|
| 333 |
+
"step": 44
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.15151515151515152,
|
| 337 |
+
"grad_norm": 0.94921875,
|
| 338 |
+
"learning_rate": 1.9999931213623708e-05,
|
| 339 |
+
"loss": 0.8638,
|
| 340 |
+
"step": 45
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.15488215488215487,
|
| 344 |
+
"grad_norm": 1.0390625,
|
| 345 |
+
"learning_rate": 1.999972485544114e-05,
|
| 346 |
+
"loss": 0.9124,
|
| 347 |
+
"step": 46
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.15824915824915825,
|
| 351 |
+
"grad_norm": 1.046875,
|
| 352 |
+
"learning_rate": 1.999938092829122e-05,
|
| 353 |
+
"loss": 0.8757,
|
| 354 |
+
"step": 47
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.16161616161616163,
|
| 358 |
+
"grad_norm": 0.73046875,
|
| 359 |
+
"learning_rate": 1.999889943690545e-05,
|
| 360 |
+
"loss": 0.885,
|
| 361 |
+
"step": 48
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.16498316498316498,
|
| 365 |
+
"grad_norm": 0.86328125,
|
| 366 |
+
"learning_rate": 1.9998280387907845e-05,
|
| 367 |
+
"loss": 0.8664,
|
| 368 |
+
"step": 49
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.16835016835016836,
|
| 372 |
+
"grad_norm": 0.87109375,
|
| 373 |
+
"learning_rate": 1.9997523789814827e-05,
|
| 374 |
+
"loss": 0.8512,
|
| 375 |
+
"step": 50
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.1717171717171717,
|
| 379 |
+
"grad_norm": 0.80859375,
|
| 380 |
+
"learning_rate": 1.9996629653035128e-05,
|
| 381 |
+
"loss": 0.8428,
|
| 382 |
+
"step": 51
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.1750841750841751,
|
| 386 |
+
"grad_norm": 0.71875,
|
| 387 |
+
"learning_rate": 1.9995597989869625e-05,
|
| 388 |
+
"loss": 0.8536,
|
| 389 |
+
"step": 52
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.17845117845117844,
|
| 393 |
+
"grad_norm": 0.6171875,
|
| 394 |
+
"learning_rate": 1.9994428814511205e-05,
|
| 395 |
+
"loss": 0.82,
|
| 396 |
+
"step": 53
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.18181818181818182,
|
| 400 |
+
"grad_norm": 0.71875,
|
| 401 |
+
"learning_rate": 1.999312214304453e-05,
|
| 402 |
+
"loss": 0.8043,
|
| 403 |
+
"step": 54
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.18518518518518517,
|
| 407 |
+
"grad_norm": 0.8046875,
|
| 408 |
+
"learning_rate": 1.9991677993445832e-05,
|
| 409 |
+
"loss": 0.8525,
|
| 410 |
+
"step": 55
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.18855218855218855,
|
| 414 |
+
"grad_norm": 0.6015625,
|
| 415 |
+
"learning_rate": 1.9990096385582682e-05,
|
| 416 |
+
"loss": 0.7958,
|
| 417 |
+
"step": 56
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.1919191919191919,
|
| 421 |
+
"grad_norm": 0.79296875,
|
| 422 |
+
"learning_rate": 1.998837734121369e-05,
|
| 423 |
+
"loss": 0.8001,
|
| 424 |
+
"step": 57
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.19528619528619529,
|
| 428 |
+
"grad_norm": 0.7109375,
|
| 429 |
+
"learning_rate": 1.9986520883988233e-05,
|
| 430 |
+
"loss": 0.805,
|
| 431 |
+
"step": 58
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.19865319865319866,
|
| 435 |
+
"grad_norm": 0.84375,
|
| 436 |
+
"learning_rate": 1.9984527039446093e-05,
|
| 437 |
+
"loss": 0.7821,
|
| 438 |
+
"step": 59
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.20202020202020202,
|
| 442 |
+
"grad_norm": 1.1015625,
|
| 443 |
+
"learning_rate": 1.998239583501714e-05,
|
| 444 |
+
"loss": 0.7932,
|
| 445 |
+
"step": 60
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.2053872053872054,
|
| 449 |
+
"grad_norm": 0.6953125,
|
| 450 |
+
"learning_rate": 1.9980127300020946e-05,
|
| 451 |
+
"loss": 0.7925,
|
| 452 |
+
"step": 61
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.20875420875420875,
|
| 456 |
+
"grad_norm": 0.60546875,
|
| 457 |
+
"learning_rate": 1.9977721465666365e-05,
|
| 458 |
+
"loss": 0.7846,
|
| 459 |
+
"step": 62
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.21212121212121213,
|
| 463 |
+
"grad_norm": 0.7265625,
|
| 464 |
+
"learning_rate": 1.9975178365051123e-05,
|
| 465 |
+
"loss": 0.7671,
|
| 466 |
+
"step": 63
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.21548821548821548,
|
| 470 |
+
"grad_norm": 0.70703125,
|
| 471 |
+
"learning_rate": 1.997249803316136e-05,
|
| 472 |
+
"loss": 0.8194,
|
| 473 |
+
"step": 64
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.21885521885521886,
|
| 477 |
+
"grad_norm": 0.81640625,
|
| 478 |
+
"learning_rate": 1.9969680506871138e-05,
|
| 479 |
+
"loss": 0.7607,
|
| 480 |
+
"step": 65
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.2222222222222222,
|
| 484 |
+
"grad_norm": 0.61328125,
|
| 485 |
+
"learning_rate": 1.9966725824941933e-05,
|
| 486 |
+
"loss": 0.7858,
|
| 487 |
+
"step": 66
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.2255892255892256,
|
| 491 |
+
"grad_norm": 0.66015625,
|
| 492 |
+
"learning_rate": 1.9963634028022133e-05,
|
| 493 |
+
"loss": 0.7546,
|
| 494 |
+
"step": 67
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.22895622895622897,
|
| 498 |
+
"grad_norm": 0.6875,
|
| 499 |
+
"learning_rate": 1.9960405158646425e-05,
|
| 500 |
+
"loss": 0.7915,
|
| 501 |
+
"step": 68
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.23232323232323232,
|
| 505 |
+
"grad_norm": 0.59765625,
|
| 506 |
+
"learning_rate": 1.9957039261235263e-05,
|
| 507 |
+
"loss": 0.7324,
|
| 508 |
+
"step": 69
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.2356902356902357,
|
| 512 |
+
"grad_norm": 0.65625,
|
| 513 |
+
"learning_rate": 1.9953536382094218e-05,
|
| 514 |
+
"loss": 0.7735,
|
| 515 |
+
"step": 70
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.23905723905723905,
|
| 519 |
+
"grad_norm": 0.984375,
|
| 520 |
+
"learning_rate": 1.9949896569413368e-05,
|
| 521 |
+
"loss": 0.7356,
|
| 522 |
+
"step": 71
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.24242424242424243,
|
| 526 |
+
"grad_norm": 0.8125,
|
| 527 |
+
"learning_rate": 1.9946119873266615e-05,
|
| 528 |
+
"loss": 0.7407,
|
| 529 |
+
"step": 72
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.24579124579124578,
|
| 533 |
+
"grad_norm": 0.76953125,
|
| 534 |
+
"learning_rate": 1.9942206345611008e-05,
|
| 535 |
+
"loss": 0.7124,
|
| 536 |
+
"step": 73
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.24915824915824916,
|
| 540 |
+
"grad_norm": 0.6640625,
|
| 541 |
+
"learning_rate": 1.9938156040286027e-05,
|
| 542 |
+
"loss": 0.7295,
|
| 543 |
+
"step": 74
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.25252525252525254,
|
| 547 |
+
"grad_norm": 0.5546875,
|
| 548 |
+
"learning_rate": 1.993396901301283e-05,
|
| 549 |
+
"loss": 0.733,
|
| 550 |
+
"step": 75
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.2558922558922559,
|
| 554 |
+
"grad_norm": 0.5703125,
|
| 555 |
+
"learning_rate": 1.9929645321393505e-05,
|
| 556 |
+
"loss": 0.779,
|
| 557 |
+
"step": 76
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.2558922558922559,
|
| 561 |
+
"eval_loss": 0.7325140237808228,
|
| 562 |
+
"eval_runtime": 8.7635,
|
| 563 |
+
"eval_samples_per_second": 57.055,
|
| 564 |
+
"eval_steps_per_second": 3.652,
|
| 565 |
+
"step": 76
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.25925925925925924,
|
| 569 |
+
"grad_norm": 0.67578125,
|
| 570 |
+
"learning_rate": 1.992518502491028e-05,
|
| 571 |
+
"loss": 0.7345,
|
| 572 |
+
"step": 77
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.26262626262626265,
|
| 576 |
+
"grad_norm": 0.5,
|
| 577 |
+
"learning_rate": 1.9920588184924664e-05,
|
| 578 |
+
"loss": 0.7102,
|
| 579 |
+
"step": 78
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.265993265993266,
|
| 583 |
+
"grad_norm": 0.51953125,
|
| 584 |
+
"learning_rate": 1.9915854864676665e-05,
|
| 585 |
+
"loss": 0.7433,
|
| 586 |
+
"step": 79
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.26936026936026936,
|
| 590 |
+
"grad_norm": 0.48828125,
|
| 591 |
+
"learning_rate": 1.991098512928386e-05,
|
| 592 |
+
"loss": 0.7628,
|
| 593 |
+
"step": 80
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.2727272727272727,
|
| 597 |
+
"grad_norm": 0.55859375,
|
| 598 |
+
"learning_rate": 1.990597904574055e-05,
|
| 599 |
+
"loss": 0.6858,
|
| 600 |
+
"step": 81
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.2760942760942761,
|
| 604 |
+
"grad_norm": 0.5390625,
|
| 605 |
+
"learning_rate": 1.9900836682916796e-05,
|
| 606 |
+
"loss": 0.7169,
|
| 607 |
+
"step": 82
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.27946127946127947,
|
| 611 |
+
"grad_norm": 0.494140625,
|
| 612 |
+
"learning_rate": 1.9895558111557503e-05,
|
| 613 |
+
"loss": 0.6995,
|
| 614 |
+
"step": 83
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.2828282828282828,
|
| 618 |
+
"grad_norm": 0.4765625,
|
| 619 |
+
"learning_rate": 1.989014340428143e-05,
|
| 620 |
+
"loss": 0.7234,
|
| 621 |
+
"step": 84
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.28619528619528617,
|
| 625 |
+
"grad_norm": 0.5546875,
|
| 626 |
+
"learning_rate": 1.98845926355802e-05,
|
| 627 |
+
"loss": 0.7182,
|
| 628 |
+
"step": 85
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.2895622895622896,
|
| 632 |
+
"grad_norm": 0.51171875,
|
| 633 |
+
"learning_rate": 1.9878905881817254e-05,
|
| 634 |
+
"loss": 0.6837,
|
| 635 |
+
"step": 86
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.29292929292929293,
|
| 639 |
+
"grad_norm": 0.54296875,
|
| 640 |
+
"learning_rate": 1.9873083221226833e-05,
|
| 641 |
+
"loss": 0.6759,
|
| 642 |
+
"step": 87
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.2962962962962963,
|
| 646 |
+
"grad_norm": 0.52734375,
|
| 647 |
+
"learning_rate": 1.986712473391289e-05,
|
| 648 |
+
"loss": 0.7275,
|
| 649 |
+
"step": 88
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.2996632996632997,
|
| 653 |
+
"grad_norm": 0.51171875,
|
| 654 |
+
"learning_rate": 1.986103050184797e-05,
|
| 655 |
+
"loss": 0.728,
|
| 656 |
+
"step": 89
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.30303030303030304,
|
| 660 |
+
"grad_norm": 0.4609375,
|
| 661 |
+
"learning_rate": 1.9854800608872096e-05,
|
| 662 |
+
"loss": 0.6865,
|
| 663 |
+
"step": 90
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.3063973063973064,
|
| 667 |
+
"grad_norm": 0.44921875,
|
| 668 |
+
"learning_rate": 1.9848435140691627e-05,
|
| 669 |
+
"loss": 0.6816,
|
| 670 |
+
"step": 91
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.30976430976430974,
|
| 674 |
+
"grad_norm": 0.4453125,
|
| 675 |
+
"learning_rate": 1.9841934184878056e-05,
|
| 676 |
+
"loss": 0.6767,
|
| 677 |
+
"step": 92
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.31313131313131315,
|
| 681 |
+
"grad_norm": 0.546875,
|
| 682 |
+
"learning_rate": 1.9835297830866827e-05,
|
| 683 |
+
"loss": 0.7623,
|
| 684 |
+
"step": 93
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.3164983164983165,
|
| 688 |
+
"grad_norm": 0.45703125,
|
| 689 |
+
"learning_rate": 1.9828526169956083e-05,
|
| 690 |
+
"loss": 0.7193,
|
| 691 |
+
"step": 94
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.31986531986531985,
|
| 695 |
+
"grad_norm": 0.482421875,
|
| 696 |
+
"learning_rate": 1.9821619295305432e-05,
|
| 697 |
+
"loss": 0.6573,
|
| 698 |
+
"step": 95
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.32323232323232326,
|
| 702 |
+
"grad_norm": 0.43359375,
|
| 703 |
+
"learning_rate": 1.9814577301934647e-05,
|
| 704 |
+
"loss": 0.7,
|
| 705 |
+
"step": 96
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.3265993265993266,
|
| 709 |
+
"grad_norm": 0.56640625,
|
| 710 |
+
"learning_rate": 1.980740028672237e-05,
|
| 711 |
+
"loss": 0.7111,
|
| 712 |
+
"step": 97
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.32996632996632996,
|
| 716 |
+
"grad_norm": 0.47265625,
|
| 717 |
+
"learning_rate": 1.9800088348404778e-05,
|
| 718 |
+
"loss": 0.7025,
|
| 719 |
+
"step": 98
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.3333333333333333,
|
| 723 |
+
"grad_norm": 0.51953125,
|
| 724 |
+
"learning_rate": 1.9792641587574212e-05,
|
| 725 |
+
"loss": 0.6925,
|
| 726 |
+
"step": 99
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.3367003367003367,
|
| 730 |
+
"grad_norm": 0.484375,
|
| 731 |
+
"learning_rate": 1.9785060106677818e-05,
|
| 732 |
+
"loss": 0.6715,
|
| 733 |
+
"step": 100
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.3400673400673401,
|
| 737 |
+
"grad_norm": 0.443359375,
|
| 738 |
+
"learning_rate": 1.977734401001611e-05,
|
| 739 |
+
"loss": 0.6929,
|
| 740 |
+
"step": 101
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.3434343434343434,
|
| 744 |
+
"grad_norm": 0.5703125,
|
| 745 |
+
"learning_rate": 1.9769493403741556e-05,
|
| 746 |
+
"loss": 0.7126,
|
| 747 |
+
"step": 102
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.3468013468013468,
|
| 751 |
+
"grad_norm": 0.490234375,
|
| 752 |
+
"learning_rate": 1.9761508395857106e-05,
|
| 753 |
+
"loss": 0.7012,
|
| 754 |
+
"step": 103
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.3501683501683502,
|
| 758 |
+
"grad_norm": 0.54296875,
|
| 759 |
+
"learning_rate": 1.9753389096214716e-05,
|
| 760 |
+
"loss": 0.6819,
|
| 761 |
+
"step": 104
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.35353535353535354,
|
| 765 |
+
"grad_norm": 0.412109375,
|
| 766 |
+
"learning_rate": 1.974513561651382e-05,
|
| 767 |
+
"loss": 0.7073,
|
| 768 |
+
"step": 105
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.3569023569023569,
|
| 772 |
+
"grad_norm": 0.4296875,
|
| 773 |
+
"learning_rate": 1.9736748070299813e-05,
|
| 774 |
+
"loss": 0.6743,
|
| 775 |
+
"step": 106
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.3602693602693603,
|
| 779 |
+
"grad_norm": 0.4140625,
|
| 780 |
+
"learning_rate": 1.9728226572962474e-05,
|
| 781 |
+
"loss": 0.6746,
|
| 782 |
+
"step": 107
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.36363636363636365,
|
| 786 |
+
"grad_norm": 0.5078125,
|
| 787 |
+
"learning_rate": 1.9719571241734395e-05,
|
| 788 |
+
"loss": 0.6729,
|
| 789 |
+
"step": 108
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.367003367003367,
|
| 793 |
+
"grad_norm": 0.74609375,
|
| 794 |
+
"learning_rate": 1.9710782195689343e-05,
|
| 795 |
+
"loss": 0.6781,
|
| 796 |
+
"step": 109
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.37037037037037035,
|
| 800 |
+
"grad_norm": 0.462890625,
|
| 801 |
+
"learning_rate": 1.9701859555740647e-05,
|
| 802 |
+
"loss": 0.6976,
|
| 803 |
+
"step": 110
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.37373737373737376,
|
| 807 |
+
"grad_norm": 0.396484375,
|
| 808 |
+
"learning_rate": 1.9692803444639517e-05,
|
| 809 |
+
"loss": 0.6464,
|
| 810 |
+
"step": 111
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.3771043771043771,
|
| 814 |
+
"grad_norm": 0.58203125,
|
| 815 |
+
"learning_rate": 1.9683613986973373e-05,
|
| 816 |
+
"loss": 0.6614,
|
| 817 |
+
"step": 112
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.38047138047138046,
|
| 821 |
+
"grad_norm": 0.416015625,
|
| 822 |
+
"learning_rate": 1.967429130916411e-05,
|
| 823 |
+
"loss": 0.6087,
|
| 824 |
+
"step": 113
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.3838383838383838,
|
| 828 |
+
"grad_norm": 0.431640625,
|
| 829 |
+
"learning_rate": 1.966483553946637e-05,
|
| 830 |
+
"loss": 0.6894,
|
| 831 |
+
"step": 114
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.3838383838383838,
|
| 835 |
+
"eval_loss": 0.6650684475898743,
|
| 836 |
+
"eval_runtime": 8.7352,
|
| 837 |
+
"eval_samples_per_second": 57.24,
|
| 838 |
+
"eval_steps_per_second": 3.663,
|
| 839 |
+
"step": 114
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 0.3872053872053872,
|
| 843 |
+
"grad_norm": 0.396484375,
|
| 844 |
+
"learning_rate": 1.9655246807965786e-05,
|
| 845 |
+
"loss": 0.6962,
|
| 846 |
+
"step": 115
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 0.39057239057239057,
|
| 850 |
+
"grad_norm": 0.39453125,
|
| 851 |
+
"learning_rate": 1.9645525246577168e-05,
|
| 852 |
+
"loss": 0.6614,
|
| 853 |
+
"step": 116
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.3939393939393939,
|
| 857 |
+
"grad_norm": 0.427734375,
|
| 858 |
+
"learning_rate": 1.963567098904272e-05,
|
| 859 |
+
"loss": 0.6889,
|
| 860 |
+
"step": 117
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.39730639730639733,
|
| 864 |
+
"grad_norm": 0.357421875,
|
| 865 |
+
"learning_rate": 1.9625684170930172e-05,
|
| 866 |
+
"loss": 0.642,
|
| 867 |
+
"step": 118
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.4006734006734007,
|
| 871 |
+
"grad_norm": 0.3828125,
|
| 872 |
+
"learning_rate": 1.9615564929630925e-05,
|
| 873 |
+
"loss": 0.6185,
|
| 874 |
+
"step": 119
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.40404040404040403,
|
| 878 |
+
"grad_norm": 0.396484375,
|
| 879 |
+
"learning_rate": 1.9605313404358176e-05,
|
| 880 |
+
"loss": 0.6624,
|
| 881 |
+
"step": 120
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.4074074074074074,
|
| 885 |
+
"grad_norm": 0.390625,
|
| 886 |
+
"learning_rate": 1.9594929736144978e-05,
|
| 887 |
+
"loss": 0.6632,
|
| 888 |
+
"step": 121
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.4107744107744108,
|
| 892 |
+
"grad_norm": 0.51953125,
|
| 893 |
+
"learning_rate": 1.9584414067842304e-05,
|
| 894 |
+
"loss": 0.6504,
|
| 895 |
+
"step": 122
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.41414141414141414,
|
| 899 |
+
"grad_norm": 0.44140625,
|
| 900 |
+
"learning_rate": 1.957376654411711e-05,
|
| 901 |
+
"loss": 0.6543,
|
| 902 |
+
"step": 123
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.4175084175084175,
|
| 906 |
+
"grad_norm": 0.48828125,
|
| 907 |
+
"learning_rate": 1.9562987311450298e-05,
|
| 908 |
+
"loss": 0.626,
|
| 909 |
+
"step": 124
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.4208754208754209,
|
| 913 |
+
"grad_norm": 0.38671875,
|
| 914 |
+
"learning_rate": 1.955207651813475e-05,
|
| 915 |
+
"loss": 0.7022,
|
| 916 |
+
"step": 125
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 0.42424242424242425,
|
| 920 |
+
"grad_norm": 0.34765625,
|
| 921 |
+
"learning_rate": 1.954103431427325e-05,
|
| 922 |
+
"loss": 0.6574,
|
| 923 |
+
"step": 126
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 0.4276094276094276,
|
| 927 |
+
"grad_norm": 0.3984375,
|
| 928 |
+
"learning_rate": 1.952986085177643e-05,
|
| 929 |
+
"loss": 0.6068,
|
| 930 |
+
"step": 127
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.43097643097643096,
|
| 934 |
+
"grad_norm": 0.380859375,
|
| 935 |
+
"learning_rate": 1.9518556284360696e-05,
|
| 936 |
+
"loss": 0.6538,
|
| 937 |
+
"step": 128
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.43434343434343436,
|
| 941 |
+
"grad_norm": 0.42578125,
|
| 942 |
+
"learning_rate": 1.950712076754609e-05,
|
| 943 |
+
"loss": 0.7139,
|
| 944 |
+
"step": 129
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.4377104377104377,
|
| 948 |
+
"grad_norm": 0.42578125,
|
| 949 |
+
"learning_rate": 1.949555445865417e-05,
|
| 950 |
+
"loss": 0.6362,
|
| 951 |
+
"step": 130
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 0.44107744107744107,
|
| 955 |
+
"grad_norm": 0.357421875,
|
| 956 |
+
"learning_rate": 1.9483857516805823e-05,
|
| 957 |
+
"loss": 0.6382,
|
| 958 |
+
"step": 131
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 0.4444444444444444,
|
| 962 |
+
"grad_norm": 0.337890625,
|
| 963 |
+
"learning_rate": 1.9472030102919102e-05,
|
| 964 |
+
"loss": 0.6522,
|
| 965 |
+
"step": 132
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.4478114478114478,
|
| 969 |
+
"grad_norm": 0.390625,
|
| 970 |
+
"learning_rate": 1.9460072379706995e-05,
|
| 971 |
+
"loss": 0.6583,
|
| 972 |
+
"step": 133
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.4511784511784512,
|
| 976 |
+
"grad_norm": 0.373046875,
|
| 977 |
+
"learning_rate": 1.9447984511675192e-05,
|
| 978 |
+
"loss": 0.6343,
|
| 979 |
+
"step": 134
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 0.45454545454545453,
|
| 983 |
+
"grad_norm": 0.392578125,
|
| 984 |
+
"learning_rate": 1.9435766665119823e-05,
|
| 985 |
+
"loss": 0.6442,
|
| 986 |
+
"step": 135
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.45791245791245794,
|
| 990 |
+
"grad_norm": 0.375,
|
| 991 |
+
"learning_rate": 1.942341900812516e-05,
|
| 992 |
+
"loss": 0.6286,
|
| 993 |
+
"step": 136
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.4612794612794613,
|
| 997 |
+
"grad_norm": 0.427734375,
|
| 998 |
+
"learning_rate": 1.9410941710561323e-05,
|
| 999 |
+
"loss": 0.6596,
|
| 1000 |
+
"step": 137
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.46464646464646464,
|
| 1004 |
+
"grad_norm": 0.423828125,
|
| 1005 |
+
"learning_rate": 1.939833494408193e-05,
|
| 1006 |
+
"loss": 0.6632,
|
| 1007 |
+
"step": 138
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.468013468013468,
|
| 1011 |
+
"grad_norm": 0.349609375,
|
| 1012 |
+
"learning_rate": 1.9385598882121735e-05,
|
| 1013 |
+
"loss": 0.6525,
|
| 1014 |
+
"step": 139
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.4713804713804714,
|
| 1018 |
+
"grad_norm": 0.431640625,
|
| 1019 |
+
"learning_rate": 1.9372733699894253e-05,
|
| 1020 |
+
"loss": 0.6136,
|
| 1021 |
+
"step": 140
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.47474747474747475,
|
| 1025 |
+
"grad_norm": 0.37109375,
|
| 1026 |
+
"learning_rate": 1.935973957438933e-05,
|
| 1027 |
+
"loss": 0.6908,
|
| 1028 |
+
"step": 141
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.4781144781144781,
|
| 1032 |
+
"grad_norm": 0.341796875,
|
| 1033 |
+
"learning_rate": 1.934661668437073e-05,
|
| 1034 |
+
"loss": 0.6374,
|
| 1035 |
+
"step": 142
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 0.48148148148148145,
|
| 1039 |
+
"grad_norm": 0.37890625,
|
| 1040 |
+
"learning_rate": 1.9333365210373668e-05,
|
| 1041 |
+
"loss": 0.6386,
|
| 1042 |
+
"step": 143
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.48484848484848486,
|
| 1046 |
+
"grad_norm": 0.3828125,
|
| 1047 |
+
"learning_rate": 1.9319985334702315e-05,
|
| 1048 |
+
"loss": 0.6775,
|
| 1049 |
+
"step": 144
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.4882154882154882,
|
| 1053 |
+
"grad_norm": 0.400390625,
|
| 1054 |
+
"learning_rate": 1.9306477241427303e-05,
|
| 1055 |
+
"loss": 0.6006,
|
| 1056 |
+
"step": 145
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.49158249158249157,
|
| 1060 |
+
"grad_norm": 0.37890625,
|
| 1061 |
+
"learning_rate": 1.9292841116383192e-05,
|
| 1062 |
+
"loss": 0.6123,
|
| 1063 |
+
"step": 146
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 0.494949494949495,
|
| 1067 |
+
"grad_norm": 0.416015625,
|
| 1068 |
+
"learning_rate": 1.9279077147165903e-05,
|
| 1069 |
+
"loss": 0.6198,
|
| 1070 |
+
"step": 147
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.4983164983164983,
|
| 1074 |
+
"grad_norm": 0.365234375,
|
| 1075 |
+
"learning_rate": 1.9265185523130156e-05,
|
| 1076 |
+
"loss": 0.624,
|
| 1077 |
+
"step": 148
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 0.5016835016835017,
|
| 1081 |
+
"grad_norm": 0.337890625,
|
| 1082 |
+
"learning_rate": 1.9251166435386837e-05,
|
| 1083 |
+
"loss": 0.6348,
|
| 1084 |
+
"step": 149
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 0.5050505050505051,
|
| 1088 |
+
"grad_norm": 0.361328125,
|
| 1089 |
+
"learning_rate": 1.9237020076800407e-05,
|
| 1090 |
+
"loss": 0.6379,
|
| 1091 |
+
"step": 150
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.5084175084175084,
|
| 1095 |
+
"grad_norm": 0.3828125,
|
| 1096 |
+
"learning_rate": 1.9222746641986207e-05,
|
| 1097 |
+
"loss": 0.6065,
|
| 1098 |
+
"step": 151
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.5117845117845118,
|
| 1102 |
+
"grad_norm": 0.419921875,
|
| 1103 |
+
"learning_rate": 1.920834632730781e-05,
|
| 1104 |
+
"loss": 0.6137,
|
| 1105 |
+
"step": 152
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.5117845117845118,
|
| 1109 |
+
"eval_loss": 0.6364830732345581,
|
| 1110 |
+
"eval_runtime": 8.689,
|
| 1111 |
+
"eval_samples_per_second": 57.544,
|
| 1112 |
+
"eval_steps_per_second": 3.683,
|
| 1113 |
+
"step": 152
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 0.5151515151515151,
|
| 1117 |
+
"grad_norm": 0.365234375,
|
| 1118 |
+
"learning_rate": 1.919381933087431e-05,
|
| 1119 |
+
"loss": 0.6596,
|
| 1120 |
+
"step": 153
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.5185185185185185,
|
| 1124 |
+
"grad_norm": 0.38671875,
|
| 1125 |
+
"learning_rate": 1.9179165852537596e-05,
|
| 1126 |
+
"loss": 0.6292,
|
| 1127 |
+
"step": 154
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 0.5218855218855218,
|
| 1131 |
+
"grad_norm": 0.345703125,
|
| 1132 |
+
"learning_rate": 1.9164386093889598e-05,
|
| 1133 |
+
"loss": 0.6555,
|
| 1134 |
+
"step": 155
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.5252525252525253,
|
| 1138 |
+
"grad_norm": 0.376953125,
|
| 1139 |
+
"learning_rate": 1.9149480258259535e-05,
|
| 1140 |
+
"loss": 0.632,
|
| 1141 |
+
"step": 156
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 0.5286195286195287,
|
| 1145 |
+
"grad_norm": 0.3515625,
|
| 1146 |
+
"learning_rate": 1.9134448550711077e-05,
|
| 1147 |
+
"loss": 0.644,
|
| 1148 |
+
"step": 157
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 0.531986531986532,
|
| 1152 |
+
"grad_norm": 0.353515625,
|
| 1153 |
+
"learning_rate": 1.9119291178039573e-05,
|
| 1154 |
+
"loss": 0.6549,
|
| 1155 |
+
"step": 158
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 0.5353535353535354,
|
| 1159 |
+
"grad_norm": 0.330078125,
|
| 1160 |
+
"learning_rate": 1.9104008348769164e-05,
|
| 1161 |
+
"loss": 0.6572,
|
| 1162 |
+
"step": 159
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 0.5387205387205387,
|
| 1166 |
+
"grad_norm": 0.3515625,
|
| 1167 |
+
"learning_rate": 1.9088600273149947e-05,
|
| 1168 |
+
"loss": 0.6284,
|
| 1169 |
+
"step": 160
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 0.5420875420875421,
|
| 1173 |
+
"grad_norm": 0.5,
|
| 1174 |
+
"learning_rate": 1.907306716315505e-05,
|
| 1175 |
+
"loss": 0.6495,
|
| 1176 |
+
"step": 161
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 0.5454545454545454,
|
| 1180 |
+
"grad_norm": 0.357421875,
|
| 1181 |
+
"learning_rate": 1.905740923247775e-05,
|
| 1182 |
+
"loss": 0.6357,
|
| 1183 |
+
"step": 162
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 0.5488215488215489,
|
| 1187 |
+
"grad_norm": 0.349609375,
|
| 1188 |
+
"learning_rate": 1.9041626696528503e-05,
|
| 1189 |
+
"loss": 0.6188,
|
| 1190 |
+
"step": 163
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.5521885521885522,
|
| 1194 |
+
"grad_norm": 0.34375,
|
| 1195 |
+
"learning_rate": 1.9025719772432006e-05,
|
| 1196 |
+
"loss": 0.6142,
|
| 1197 |
+
"step": 164
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 0.5555555555555556,
|
| 1201 |
+
"grad_norm": 0.337890625,
|
| 1202 |
+
"learning_rate": 1.900968867902419e-05,
|
| 1203 |
+
"loss": 0.6339,
|
| 1204 |
+
"step": 165
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 0.5589225589225589,
|
| 1208 |
+
"grad_norm": 0.412109375,
|
| 1209 |
+
"learning_rate": 1.8993533636849223e-05,
|
| 1210 |
+
"loss": 0.6067,
|
| 1211 |
+
"step": 166
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 0.5622895622895623,
|
| 1215 |
+
"grad_norm": 0.494140625,
|
| 1216 |
+
"learning_rate": 1.8977254868156465e-05,
|
| 1217 |
+
"loss": 0.5849,
|
| 1218 |
+
"step": 167
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 0.5656565656565656,
|
| 1222 |
+
"grad_norm": 0.31640625,
|
| 1223 |
+
"learning_rate": 1.896085259689741e-05,
|
| 1224 |
+
"loss": 0.6083,
|
| 1225 |
+
"step": 168
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.569023569023569,
|
| 1229 |
+
"grad_norm": 0.3984375,
|
| 1230 |
+
"learning_rate": 1.8944327048722634e-05,
|
| 1231 |
+
"loss": 0.6319,
|
| 1232 |
+
"step": 169
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 0.5723905723905723,
|
| 1236 |
+
"grad_norm": 0.4375,
|
| 1237 |
+
"learning_rate": 1.892767845097864e-05,
|
| 1238 |
+
"loss": 0.6299,
|
| 1239 |
+
"step": 170
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.5757575757575758,
|
| 1243 |
+
"grad_norm": 0.302734375,
|
| 1244 |
+
"learning_rate": 1.891090703270477e-05,
|
| 1245 |
+
"loss": 0.5819,
|
| 1246 |
+
"step": 171
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 0.5791245791245792,
|
| 1250 |
+
"grad_norm": 0.34375,
|
| 1251 |
+
"learning_rate": 1.889401302463005e-05,
|
| 1252 |
+
"loss": 0.6166,
|
| 1253 |
+
"step": 172
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 0.5824915824915825,
|
| 1257 |
+
"grad_norm": 0.392578125,
|
| 1258 |
+
"learning_rate": 1.887699665916999e-05,
|
| 1259 |
+
"loss": 0.605,
|
| 1260 |
+
"step": 173
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 0.5858585858585859,
|
| 1264 |
+
"grad_norm": 0.33984375,
|
| 1265 |
+
"learning_rate": 1.8859858170423423e-05,
|
| 1266 |
+
"loss": 0.6573,
|
| 1267 |
+
"step": 174
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 0.5892255892255892,
|
| 1271 |
+
"grad_norm": 0.34375,
|
| 1272 |
+
"learning_rate": 1.8842597794169245e-05,
|
| 1273 |
+
"loss": 0.5891,
|
| 1274 |
+
"step": 175
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.5925925925925926,
|
| 1278 |
+
"grad_norm": 0.359375,
|
| 1279 |
+
"learning_rate": 1.8825215767863215e-05,
|
| 1280 |
+
"loss": 0.6383,
|
| 1281 |
+
"step": 176
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 0.5959595959595959,
|
| 1285 |
+
"grad_norm": 0.30859375,
|
| 1286 |
+
"learning_rate": 1.8807712330634645e-05,
|
| 1287 |
+
"loss": 0.5958,
|
| 1288 |
+
"step": 177
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 0.5993265993265994,
|
| 1292 |
+
"grad_norm": 0.3671875,
|
| 1293 |
+
"learning_rate": 1.879008772328314e-05,
|
| 1294 |
+
"loss": 0.5791,
|
| 1295 |
+
"step": 178
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 0.6026936026936027,
|
| 1299 |
+
"grad_norm": 0.37890625,
|
| 1300 |
+
"learning_rate": 1.877234218827528e-05,
|
| 1301 |
+
"loss": 0.6294,
|
| 1302 |
+
"step": 179
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 0.6060606060606061,
|
| 1306 |
+
"grad_norm": 0.40625,
|
| 1307 |
+
"learning_rate": 1.8754475969741272e-05,
|
| 1308 |
+
"loss": 0.5971,
|
| 1309 |
+
"step": 180
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 0.6094276094276094,
|
| 1313 |
+
"grad_norm": 0.34375,
|
| 1314 |
+
"learning_rate": 1.87364893134716e-05,
|
| 1315 |
+
"loss": 0.6078,
|
| 1316 |
+
"step": 181
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 0.6127946127946128,
|
| 1320 |
+
"grad_norm": 0.326171875,
|
| 1321 |
+
"learning_rate": 1.8718382466913642e-05,
|
| 1322 |
+
"loss": 0.6115,
|
| 1323 |
+
"step": 182
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 0.6161616161616161,
|
| 1327 |
+
"grad_norm": 0.3515625,
|
| 1328 |
+
"learning_rate": 1.8700155679168277e-05,
|
| 1329 |
+
"loss": 0.6263,
|
| 1330 |
+
"step": 183
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.6195286195286195,
|
| 1334 |
+
"grad_norm": 0.365234375,
|
| 1335 |
+
"learning_rate": 1.868180920098644e-05,
|
| 1336 |
+
"loss": 0.5933,
|
| 1337 |
+
"step": 184
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 0.622895622895623,
|
| 1341 |
+
"grad_norm": 0.419921875,
|
| 1342 |
+
"learning_rate": 1.866334328476568e-05,
|
| 1343 |
+
"loss": 0.6371,
|
| 1344 |
+
"step": 185
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.6262626262626263,
|
| 1348 |
+
"grad_norm": 0.333984375,
|
| 1349 |
+
"learning_rate": 1.864475818454669e-05,
|
| 1350 |
+
"loss": 0.6314,
|
| 1351 |
+
"step": 186
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 0.6296296296296297,
|
| 1355 |
+
"grad_norm": 0.302734375,
|
| 1356 |
+
"learning_rate": 1.8626054156009807e-05,
|
| 1357 |
+
"loss": 0.6802,
|
| 1358 |
+
"step": 187
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 0.632996632996633,
|
| 1362 |
+
"grad_norm": 0.359375,
|
| 1363 |
+
"learning_rate": 1.8607231456471505e-05,
|
| 1364 |
+
"loss": 0.6189,
|
| 1365 |
+
"step": 188
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 0.6363636363636364,
|
| 1369 |
+
"grad_norm": 0.337890625,
|
| 1370 |
+
"learning_rate": 1.858829034488084e-05,
|
| 1371 |
+
"loss": 0.5987,
|
| 1372 |
+
"step": 189
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 0.6397306397306397,
|
| 1376 |
+
"grad_norm": 0.328125,
|
| 1377 |
+
"learning_rate": 1.8569231081815895e-05,
|
| 1378 |
+
"loss": 0.6492,
|
| 1379 |
+
"step": 190
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 0.6397306397306397,
|
| 1383 |
+
"eval_loss": 0.6246641874313354,
|
| 1384 |
+
"eval_runtime": 8.7145,
|
| 1385 |
+
"eval_samples_per_second": 57.375,
|
| 1386 |
+
"eval_steps_per_second": 3.672,
|
| 1387 |
+
"step": 190
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.6430976430976431,
|
| 1391 |
+
"grad_norm": 0.326171875,
|
| 1392 |
+
"learning_rate": 1.8550053929480202e-05,
|
| 1393 |
+
"loss": 0.6198,
|
| 1394 |
+
"step": 191
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.6464646464646465,
|
| 1398 |
+
"grad_norm": 0.330078125,
|
| 1399 |
+
"learning_rate": 1.8530759151699122e-05,
|
| 1400 |
+
"loss": 0.6253,
|
| 1401 |
+
"step": 192
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.6498316498316499,
|
| 1405 |
+
"grad_norm": 0.330078125,
|
| 1406 |
+
"learning_rate": 1.8511347013916228e-05,
|
| 1407 |
+
"loss": 0.5804,
|
| 1408 |
+
"step": 193
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.6531986531986532,
|
| 1412 |
+
"grad_norm": 0.3671875,
|
| 1413 |
+
"learning_rate": 1.8491817783189636e-05,
|
| 1414 |
+
"loss": 0.5909,
|
| 1415 |
+
"step": 194
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 0.6565656565656566,
|
| 1419 |
+
"grad_norm": 0.3359375,
|
| 1420 |
+
"learning_rate": 1.8472171728188356e-05,
|
| 1421 |
+
"loss": 0.5996,
|
| 1422 |
+
"step": 195
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.6599326599326599,
|
| 1426 |
+
"grad_norm": 0.357421875,
|
| 1427 |
+
"learning_rate": 1.845240911918857e-05,
|
| 1428 |
+
"loss": 0.6401,
|
| 1429 |
+
"step": 196
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 0.6632996632996633,
|
| 1433 |
+
"grad_norm": 0.34375,
|
| 1434 |
+
"learning_rate": 1.843253022806993e-05,
|
| 1435 |
+
"loss": 0.6576,
|
| 1436 |
+
"step": 197
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.6666666666666666,
|
| 1440 |
+
"grad_norm": 0.30859375,
|
| 1441 |
+
"learning_rate": 1.8412535328311813e-05,
|
| 1442 |
+
"loss": 0.5715,
|
| 1443 |
+
"step": 198
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 0.67003367003367,
|
| 1447 |
+
"grad_norm": 0.3203125,
|
| 1448 |
+
"learning_rate": 1.839242469498956e-05,
|
| 1449 |
+
"loss": 0.5988,
|
| 1450 |
+
"step": 199
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.6734006734006734,
|
| 1454 |
+
"grad_norm": 0.31640625,
|
| 1455 |
+
"learning_rate": 1.837219860477069e-05,
|
| 1456 |
+
"loss": 0.5796,
|
| 1457 |
+
"step": 200
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.6767676767676768,
|
| 1461 |
+
"grad_norm": 0.33984375,
|
| 1462 |
+
"learning_rate": 1.8351857335911094e-05,
|
| 1463 |
+
"loss": 0.6329,
|
| 1464 |
+
"step": 201
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 0.6801346801346801,
|
| 1468 |
+
"grad_norm": 0.330078125,
|
| 1469 |
+
"learning_rate": 1.83314011682512e-05,
|
| 1470 |
+
"loss": 0.6374,
|
| 1471 |
+
"step": 202
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 0.6835016835016835,
|
| 1475 |
+
"grad_norm": 0.3125,
|
| 1476 |
+
"learning_rate": 1.831083038321215e-05,
|
| 1477 |
+
"loss": 0.6235,
|
| 1478 |
+
"step": 203
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.6868686868686869,
|
| 1482 |
+
"grad_norm": 0.349609375,
|
| 1483 |
+
"learning_rate": 1.8290145263791883e-05,
|
| 1484 |
+
"loss": 0.621,
|
| 1485 |
+
"step": 204
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 0.6902356902356902,
|
| 1489 |
+
"grad_norm": 0.359375,
|
| 1490 |
+
"learning_rate": 1.826934609456129e-05,
|
| 1491 |
+
"loss": 0.6396,
|
| 1492 |
+
"step": 205
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 0.6936026936026936,
|
| 1496 |
+
"grad_norm": 0.337890625,
|
| 1497 |
+
"learning_rate": 1.824843316166026e-05,
|
| 1498 |
+
"loss": 0.6197,
|
| 1499 |
+
"step": 206
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.696969696969697,
|
| 1503 |
+
"grad_norm": 0.361328125,
|
| 1504 |
+
"learning_rate": 1.822740675279377e-05,
|
| 1505 |
+
"loss": 0.6473,
|
| 1506 |
+
"step": 207
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.7003367003367004,
|
| 1510 |
+
"grad_norm": 0.34375,
|
| 1511 |
+
"learning_rate": 1.8206267157227918e-05,
|
| 1512 |
+
"loss": 0.6227,
|
| 1513 |
+
"step": 208
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.7037037037037037,
|
| 1517 |
+
"grad_norm": 0.330078125,
|
| 1518 |
+
"learning_rate": 1.8185014665785936e-05,
|
| 1519 |
+
"loss": 0.6009,
|
| 1520 |
+
"step": 209
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.7070707070707071,
|
| 1524 |
+
"grad_norm": 0.353515625,
|
| 1525 |
+
"learning_rate": 1.8163649570844198e-05,
|
| 1526 |
+
"loss": 0.592,
|
| 1527 |
+
"step": 210
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 0.7104377104377104,
|
| 1531 |
+
"grad_norm": 0.376953125,
|
| 1532 |
+
"learning_rate": 1.8142172166328198e-05,
|
| 1533 |
+
"loss": 0.6063,
|
| 1534 |
+
"step": 211
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 0.7138047138047138,
|
| 1538 |
+
"grad_norm": 0.34765625,
|
| 1539 |
+
"learning_rate": 1.8120582747708503e-05,
|
| 1540 |
+
"loss": 0.6315,
|
| 1541 |
+
"step": 212
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 0.7171717171717171,
|
| 1545 |
+
"grad_norm": 0.36328125,
|
| 1546 |
+
"learning_rate": 1.809888161199668e-05,
|
| 1547 |
+
"loss": 0.6297,
|
| 1548 |
+
"step": 213
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 0.7205387205387206,
|
| 1552 |
+
"grad_norm": 0.349609375,
|
| 1553 |
+
"learning_rate": 1.8077069057741235e-05,
|
| 1554 |
+
"loss": 0.6079,
|
| 1555 |
+
"step": 214
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 0.7239057239057239,
|
| 1559 |
+
"grad_norm": 0.318359375,
|
| 1560 |
+
"learning_rate": 1.8055145385023477e-05,
|
| 1561 |
+
"loss": 0.6192,
|
| 1562 |
+
"step": 215
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.7272727272727273,
|
| 1566 |
+
"grad_norm": 0.326171875,
|
| 1567 |
+
"learning_rate": 1.803311089545341e-05,
|
| 1568 |
+
"loss": 0.6104,
|
| 1569 |
+
"step": 216
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 0.7306397306397306,
|
| 1573 |
+
"grad_norm": 0.3359375,
|
| 1574 |
+
"learning_rate": 1.8010965892165568e-05,
|
| 1575 |
+
"loss": 0.6166,
|
| 1576 |
+
"step": 217
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 0.734006734006734,
|
| 1580 |
+
"grad_norm": 0.361328125,
|
| 1581 |
+
"learning_rate": 1.7988710679814857e-05,
|
| 1582 |
+
"loss": 0.6145,
|
| 1583 |
+
"step": 218
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 0.7373737373737373,
|
| 1587 |
+
"grad_norm": 0.32421875,
|
| 1588 |
+
"learning_rate": 1.796634556457236e-05,
|
| 1589 |
+
"loss": 0.5994,
|
| 1590 |
+
"step": 219
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 0.7407407407407407,
|
| 1594 |
+
"grad_norm": 0.34765625,
|
| 1595 |
+
"learning_rate": 1.7943870854121126e-05,
|
| 1596 |
+
"loss": 0.586,
|
| 1597 |
+
"step": 220
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 0.7441077441077442,
|
| 1601 |
+
"grad_norm": 0.32421875,
|
| 1602 |
+
"learning_rate": 1.792128685765193e-05,
|
| 1603 |
+
"loss": 0.582,
|
| 1604 |
+
"step": 221
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 0.7474747474747475,
|
| 1608 |
+
"grad_norm": 0.337890625,
|
| 1609 |
+
"learning_rate": 1.789859388585903e-05,
|
| 1610 |
+
"loss": 0.6017,
|
| 1611 |
+
"step": 222
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 0.7508417508417509,
|
| 1615 |
+
"grad_norm": 0.36328125,
|
| 1616 |
+
"learning_rate": 1.7875792250935883e-05,
|
| 1617 |
+
"loss": 0.5864,
|
| 1618 |
+
"step": 223
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 0.7542087542087542,
|
| 1622 |
+
"grad_norm": 0.373046875,
|
| 1623 |
+
"learning_rate": 1.785288226657086e-05,
|
| 1624 |
+
"loss": 0.5869,
|
| 1625 |
+
"step": 224
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 0.7575757575757576,
|
| 1629 |
+
"grad_norm": 0.3203125,
|
| 1630 |
+
"learning_rate": 1.782986424794292e-05,
|
| 1631 |
+
"loss": 0.6448,
|
| 1632 |
+
"step": 225
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 0.7609427609427609,
|
| 1636 |
+
"grad_norm": 0.3203125,
|
| 1637 |
+
"learning_rate": 1.780673851171728e-05,
|
| 1638 |
+
"loss": 0.5895,
|
| 1639 |
+
"step": 226
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 0.7643097643097643,
|
| 1643 |
+
"grad_norm": 0.326171875,
|
| 1644 |
+
"learning_rate": 1.7783505376041063e-05,
|
| 1645 |
+
"loss": 0.6289,
|
| 1646 |
+
"step": 227
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.7676767676767676,
|
| 1650 |
+
"grad_norm": 0.33984375,
|
| 1651 |
+
"learning_rate": 1.7760165160538907e-05,
|
| 1652 |
+
"loss": 0.6231,
|
| 1653 |
+
"step": 228
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 0.7676767676767676,
|
| 1657 |
+
"eval_loss": 0.6190042495727539,
|
| 1658 |
+
"eval_runtime": 8.7179,
|
| 1659 |
+
"eval_samples_per_second": 57.353,
|
| 1660 |
+
"eval_steps_per_second": 3.671,
|
| 1661 |
+
"step": 228
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7710437710437711,
|
| 1665 |
+
"grad_norm": 0.384765625,
|
| 1666 |
+
"learning_rate": 1.7736718186308585e-05,
|
| 1667 |
+
"loss": 0.6327,
|
| 1668 |
+
"step": 229
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7744107744107744,
|
| 1672 |
+
"grad_norm": 0.330078125,
|
| 1673 |
+
"learning_rate": 1.7713164775916573e-05,
|
| 1674 |
+
"loss": 0.5811,
|
| 1675 |
+
"step": 230
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7777777777777778,
|
| 1679 |
+
"grad_norm": 0.314453125,
|
| 1680 |
+
"learning_rate": 1.768950525339362e-05,
|
| 1681 |
+
"loss": 0.5816,
|
| 1682 |
+
"step": 231
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7811447811447811,
|
| 1686 |
+
"grad_norm": 0.345703125,
|
| 1687 |
+
"learning_rate": 1.7665739944230296e-05,
|
| 1688 |
+
"loss": 0.6046,
|
| 1689 |
+
"step": 232
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7845117845117845,
|
| 1693 |
+
"grad_norm": 0.361328125,
|
| 1694 |
+
"learning_rate": 1.7641869175372493e-05,
|
| 1695 |
+
"loss": 0.5985,
|
| 1696 |
+
"step": 233
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7878787878787878,
|
| 1700 |
+
"grad_norm": 0.36328125,
|
| 1701 |
+
"learning_rate": 1.7617893275216953e-05,
|
| 1702 |
+
"loss": 0.6226,
|
| 1703 |
+
"step": 234
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7912457912457912,
|
| 1707 |
+
"grad_norm": 0.353515625,
|
| 1708 |
+
"learning_rate": 1.759381257360673e-05,
|
| 1709 |
+
"loss": 0.6361,
|
| 1710 |
+
"step": 235
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7946127946127947,
|
| 1714 |
+
"grad_norm": 0.328125,
|
| 1715 |
+
"learning_rate": 1.7569627401826673e-05,
|
| 1716 |
+
"loss": 0.665,
|
| 1717 |
+
"step": 236
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.797979797979798,
|
| 1721 |
+
"grad_norm": 0.3671875,
|
| 1722 |
+
"learning_rate": 1.754533809259884e-05,
|
| 1723 |
+
"loss": 0.6078,
|
| 1724 |
+
"step": 237
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8013468013468014,
|
| 1728 |
+
"grad_norm": 0.302734375,
|
| 1729 |
+
"learning_rate": 1.7520944980077948e-05,
|
| 1730 |
+
"loss": 0.6043,
|
| 1731 |
+
"step": 238
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8047138047138047,
|
| 1735 |
+
"grad_norm": 0.3125,
|
| 1736 |
+
"learning_rate": 1.7496448399846757e-05,
|
| 1737 |
+
"loss": 0.614,
|
| 1738 |
+
"step": 239
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8080808080808081,
|
| 1742 |
+
"grad_norm": 0.330078125,
|
| 1743 |
+
"learning_rate": 1.7471848688911465e-05,
|
| 1744 |
+
"loss": 0.5756,
|
| 1745 |
+
"step": 240
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8114478114478114,
|
| 1749 |
+
"grad_norm": 0.40234375,
|
| 1750 |
+
"learning_rate": 1.744714618569707e-05,
|
| 1751 |
+
"loss": 0.5982,
|
| 1752 |
+
"step": 241
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8148148148148148,
|
| 1756 |
+
"grad_norm": 0.359375,
|
| 1757 |
+
"learning_rate": 1.74223412300427e-05,
|
| 1758 |
+
"loss": 0.6126,
|
| 1759 |
+
"step": 242
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8181818181818182,
|
| 1763 |
+
"grad_norm": 0.296875,
|
| 1764 |
+
"learning_rate": 1.7397434163196967e-05,
|
| 1765 |
+
"loss": 0.6281,
|
| 1766 |
+
"step": 243
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8215488215488216,
|
| 1770 |
+
"grad_norm": 0.294921875,
|
| 1771 |
+
"learning_rate": 1.737242532781324e-05,
|
| 1772 |
+
"loss": 0.5821,
|
| 1773 |
+
"step": 244
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8249158249158249,
|
| 1777 |
+
"grad_norm": 0.31640625,
|
| 1778 |
+
"learning_rate": 1.7347315067944955e-05,
|
| 1779 |
+
"loss": 0.6546,
|
| 1780 |
+
"step": 245
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8282828282828283,
|
| 1784 |
+
"grad_norm": 0.34375,
|
| 1785 |
+
"learning_rate": 1.7322103729040868e-05,
|
| 1786 |
+
"loss": 0.6301,
|
| 1787 |
+
"step": 246
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8316498316498316,
|
| 1791 |
+
"grad_norm": 0.396484375,
|
| 1792 |
+
"learning_rate": 1.72967916579403e-05,
|
| 1793 |
+
"loss": 0.585,
|
| 1794 |
+
"step": 247
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.835016835016835,
|
| 1798 |
+
"grad_norm": 0.33203125,
|
| 1799 |
+
"learning_rate": 1.7271379202868394e-05,
|
| 1800 |
+
"loss": 0.6277,
|
| 1801 |
+
"step": 248
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8383838383838383,
|
| 1805 |
+
"grad_norm": 0.37109375,
|
| 1806 |
+
"learning_rate": 1.7245866713431278e-05,
|
| 1807 |
+
"loss": 0.6536,
|
| 1808 |
+
"step": 249
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8417508417508418,
|
| 1812 |
+
"grad_norm": 0.33984375,
|
| 1813 |
+
"learning_rate": 1.72202545406113e-05,
|
| 1814 |
+
"loss": 0.6231,
|
| 1815 |
+
"step": 250
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8451178451178452,
|
| 1819 |
+
"grad_norm": 0.30078125,
|
| 1820 |
+
"learning_rate": 1.7194543036762165e-05,
|
| 1821 |
+
"loss": 0.6102,
|
| 1822 |
+
"step": 251
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8484848484848485,
|
| 1826 |
+
"grad_norm": 0.33203125,
|
| 1827 |
+
"learning_rate": 1.7168732555604114e-05,
|
| 1828 |
+
"loss": 0.6118,
|
| 1829 |
+
"step": 252
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8518518518518519,
|
| 1833 |
+
"grad_norm": 0.328125,
|
| 1834 |
+
"learning_rate": 1.7142823452219036e-05,
|
| 1835 |
+
"loss": 0.6323,
|
| 1836 |
+
"step": 253
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8552188552188552,
|
| 1840 |
+
"grad_norm": 0.32421875,
|
| 1841 |
+
"learning_rate": 1.7116816083045603e-05,
|
| 1842 |
+
"loss": 0.6393,
|
| 1843 |
+
"step": 254
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8585858585858586,
|
| 1847 |
+
"grad_norm": 0.34375,
|
| 1848 |
+
"learning_rate": 1.709071080587435e-05,
|
| 1849 |
+
"loss": 0.5845,
|
| 1850 |
+
"step": 255
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8619528619528619,
|
| 1854 |
+
"grad_norm": 0.33203125,
|
| 1855 |
+
"learning_rate": 1.7064507979842755e-05,
|
| 1856 |
+
"loss": 0.6026,
|
| 1857 |
+
"step": 256
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8653198653198653,
|
| 1861 |
+
"grad_norm": 0.302734375,
|
| 1862 |
+
"learning_rate": 1.7038207965430316e-05,
|
| 1863 |
+
"loss": 0.5665,
|
| 1864 |
+
"step": 257
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8686868686868687,
|
| 1868 |
+
"grad_norm": 0.3203125,
|
| 1869 |
+
"learning_rate": 1.7011811124453567e-05,
|
| 1870 |
+
"loss": 0.5787,
|
| 1871 |
+
"step": 258
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8720538720538721,
|
| 1875 |
+
"grad_norm": 0.359375,
|
| 1876 |
+
"learning_rate": 1.6985317820061116e-05,
|
| 1877 |
+
"loss": 0.5506,
|
| 1878 |
+
"step": 259
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8754208754208754,
|
| 1882 |
+
"grad_norm": 0.35546875,
|
| 1883 |
+
"learning_rate": 1.6958728416728644e-05,
|
| 1884 |
+
"loss": 0.5902,
|
| 1885 |
+
"step": 260
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8787878787878788,
|
| 1889 |
+
"grad_norm": 0.404296875,
|
| 1890 |
+
"learning_rate": 1.6932043280253892e-05,
|
| 1891 |
+
"loss": 0.6037,
|
| 1892 |
+
"step": 261
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8821548821548821,
|
| 1896 |
+
"grad_norm": 0.314453125,
|
| 1897 |
+
"learning_rate": 1.6905262777751627e-05,
|
| 1898 |
+
"loss": 0.5533,
|
| 1899 |
+
"step": 262
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8855218855218855,
|
| 1903 |
+
"grad_norm": 0.296875,
|
| 1904 |
+
"learning_rate": 1.6878387277648593e-05,
|
| 1905 |
+
"loss": 0.5924,
|
| 1906 |
+
"step": 263
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8888888888888888,
|
| 1910 |
+
"grad_norm": 0.310546875,
|
| 1911 |
+
"learning_rate": 1.6851417149678442e-05,
|
| 1912 |
+
"loss": 0.5834,
|
| 1913 |
+
"step": 264
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8922558922558923,
|
| 1917 |
+
"grad_norm": 0.345703125,
|
| 1918 |
+
"learning_rate": 1.6824352764876653e-05,
|
| 1919 |
+
"loss": 0.6527,
|
| 1920 |
+
"step": 265
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8956228956228957,
|
| 1924 |
+
"grad_norm": 0.3359375,
|
| 1925 |
+
"learning_rate": 1.6797194495575412e-05,
|
| 1926 |
+
"loss": 0.6428,
|
| 1927 |
+
"step": 266
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8956228956228957,
|
| 1931 |
+
"eval_loss": 0.6151465773582458,
|
| 1932 |
+
"eval_runtime": 8.6578,
|
| 1933 |
+
"eval_samples_per_second": 57.751,
|
| 1934 |
+
"eval_steps_per_second": 3.696,
|
| 1935 |
+
"step": 266
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.898989898989899,
|
| 1939 |
+
"grad_norm": 0.298828125,
|
| 1940 |
+
"learning_rate": 1.676994271539851e-05,
|
| 1941 |
+
"loss": 0.5817,
|
| 1942 |
+
"step": 267
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.9023569023569024,
|
| 1946 |
+
"grad_norm": 0.33203125,
|
| 1947 |
+
"learning_rate": 1.6742597799256182e-05,
|
| 1948 |
+
"loss": 0.6313,
|
| 1949 |
+
"step": 268
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.9057239057239057,
|
| 1953 |
+
"grad_norm": 0.353515625,
|
| 1954 |
+
"learning_rate": 1.671516012333997e-05,
|
| 1955 |
+
"loss": 0.5895,
|
| 1956 |
+
"step": 269
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.9090909090909091,
|
| 1960 |
+
"grad_norm": 0.36328125,
|
| 1961 |
+
"learning_rate": 1.6687630065117536e-05,
|
| 1962 |
+
"loss": 0.6106,
|
| 1963 |
+
"step": 270
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.9124579124579124,
|
| 1967 |
+
"grad_norm": 0.302734375,
|
| 1968 |
+
"learning_rate": 1.666000800332747e-05,
|
| 1969 |
+
"loss": 0.567,
|
| 1970 |
+
"step": 271
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.9158249158249159,
|
| 1974 |
+
"grad_norm": 0.302734375,
|
| 1975 |
+
"learning_rate": 1.663229431797407e-05,
|
| 1976 |
+
"loss": 0.6024,
|
| 1977 |
+
"step": 272
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.9191919191919192,
|
| 1981 |
+
"grad_norm": 0.359375,
|
| 1982 |
+
"learning_rate": 1.660448939032214e-05,
|
| 1983 |
+
"loss": 0.6348,
|
| 1984 |
+
"step": 273
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.9225589225589226,
|
| 1988 |
+
"grad_norm": 0.3125,
|
| 1989 |
+
"learning_rate": 1.6576593602891727e-05,
|
| 1990 |
+
"loss": 0.6119,
|
| 1991 |
+
"step": 274
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.9259259259259259,
|
| 1995 |
+
"grad_norm": 0.314453125,
|
| 1996 |
+
"learning_rate": 1.6548607339452853e-05,
|
| 1997 |
+
"loss": 0.599,
|
| 1998 |
+
"step": 275
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.9292929292929293,
|
| 2002 |
+
"grad_norm": 0.3203125,
|
| 2003 |
+
"learning_rate": 1.6520530985020243e-05,
|
| 2004 |
+
"loss": 0.587,
|
| 2005 |
+
"step": 276
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.9326599326599326,
|
| 2009 |
+
"grad_norm": 0.318359375,
|
| 2010 |
+
"learning_rate": 1.6492364925848045e-05,
|
| 2011 |
+
"loss": 0.6113,
|
| 2012 |
+
"step": 277
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.936026936026936,
|
| 2016 |
+
"grad_norm": 0.330078125,
|
| 2017 |
+
"learning_rate": 1.6464109549424477e-05,
|
| 2018 |
+
"loss": 0.594,
|
| 2019 |
+
"step": 278
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.9393939393939394,
|
| 2023 |
+
"grad_norm": 0.283203125,
|
| 2024 |
+
"learning_rate": 1.6435765244466536e-05,
|
| 2025 |
+
"loss": 0.5823,
|
| 2026 |
+
"step": 279
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.9427609427609428,
|
| 2030 |
+
"grad_norm": 0.3203125,
|
| 2031 |
+
"learning_rate": 1.6407332400914625e-05,
|
| 2032 |
+
"loss": 0.5871,
|
| 2033 |
+
"step": 280
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.9461279461279462,
|
| 2037 |
+
"grad_norm": 0.32421875,
|
| 2038 |
+
"learning_rate": 1.63788114099272e-05,
|
| 2039 |
+
"loss": 0.6126,
|
| 2040 |
+
"step": 281
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.9494949494949495,
|
| 2044 |
+
"grad_norm": 0.31640625,
|
| 2045 |
+
"learning_rate": 1.6350202663875385e-05,
|
| 2046 |
+
"loss": 0.6343,
|
| 2047 |
+
"step": 282
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.9528619528619529,
|
| 2051 |
+
"grad_norm": 0.30078125,
|
| 2052 |
+
"learning_rate": 1.6321506556337575e-05,
|
| 2053 |
+
"loss": 0.6033,
|
| 2054 |
+
"step": 283
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.9562289562289562,
|
| 2058 |
+
"grad_norm": 0.400390625,
|
| 2059 |
+
"learning_rate": 1.6292723482094013e-05,
|
| 2060 |
+
"loss": 0.6216,
|
| 2061 |
+
"step": 284
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.9595959595959596,
|
| 2065 |
+
"grad_norm": 0.296875,
|
| 2066 |
+
"learning_rate": 1.6263853837121384e-05,
|
| 2067 |
+
"loss": 0.5599,
|
| 2068 |
+
"step": 285
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.9629629629629629,
|
| 2072 |
+
"grad_norm": 0.349609375,
|
| 2073 |
+
"learning_rate": 1.6234898018587336e-05,
|
| 2074 |
+
"loss": 0.632,
|
| 2075 |
+
"step": 286
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.9663299663299664,
|
| 2079 |
+
"grad_norm": 0.328125,
|
| 2080 |
+
"learning_rate": 1.6205856424845038e-05,
|
| 2081 |
+
"loss": 0.5824,
|
| 2082 |
+
"step": 287
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.9696969696969697,
|
| 2086 |
+
"grad_norm": 0.3359375,
|
| 2087 |
+
"learning_rate": 1.617672945542768e-05,
|
| 2088 |
+
"loss": 0.628,
|
| 2089 |
+
"step": 288
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.9730639730639731,
|
| 2093 |
+
"grad_norm": 0.275390625,
|
| 2094 |
+
"learning_rate": 1.614751751104301e-05,
|
| 2095 |
+
"loss": 0.5691,
|
| 2096 |
+
"step": 289
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.9764309764309764,
|
| 2100 |
+
"grad_norm": 0.310546875,
|
| 2101 |
+
"learning_rate": 1.6118220993567783e-05,
|
| 2102 |
+
"loss": 0.5816,
|
| 2103 |
+
"step": 290
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.9797979797979798,
|
| 2107 |
+
"grad_norm": 0.302734375,
|
| 2108 |
+
"learning_rate": 1.6088840306042247e-05,
|
| 2109 |
+
"loss": 0.5907,
|
| 2110 |
+
"step": 291
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.9831649831649831,
|
| 2114 |
+
"grad_norm": 0.341796875,
|
| 2115 |
+
"learning_rate": 1.605937585266462e-05,
|
| 2116 |
+
"loss": 0.6129,
|
| 2117 |
+
"step": 292
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.9865319865319865,
|
| 2121 |
+
"grad_norm": 0.333984375,
|
| 2122 |
+
"learning_rate": 1.6029828038785486e-05,
|
| 2123 |
+
"loss": 0.5916,
|
| 2124 |
+
"step": 293
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.98989898989899,
|
| 2128 |
+
"grad_norm": 0.38671875,
|
| 2129 |
+
"learning_rate": 1.600019727090226e-05,
|
| 2130 |
+
"loss": 0.5984,
|
| 2131 |
+
"step": 294
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.9932659932659933,
|
| 2135 |
+
"grad_norm": 0.345703125,
|
| 2136 |
+
"learning_rate": 1.5970483956653572e-05,
|
| 2137 |
+
"loss": 0.5549,
|
| 2138 |
+
"step": 295
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.9966329966329966,
|
| 2142 |
+
"grad_norm": 0.3671875,
|
| 2143 |
+
"learning_rate": 1.5940688504813664e-05,
|
| 2144 |
+
"loss": 0.6102,
|
| 2145 |
+
"step": 296
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.0,
|
| 2149 |
+
"grad_norm": 0.4140625,
|
| 2150 |
+
"learning_rate": 1.5910811325286768e-05,
|
| 2151 |
+
"loss": 0.6109,
|
| 2152 |
+
"step": 297
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.0033670033670035,
|
| 2156 |
+
"grad_norm": 0.3671875,
|
| 2157 |
+
"learning_rate": 1.5880852829101464e-05,
|
| 2158 |
+
"loss": 0.6281,
|
| 2159 |
+
"step": 298
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.0067340067340067,
|
| 2163 |
+
"grad_norm": 0.32421875,
|
| 2164 |
+
"learning_rate": 1.5850813428405036e-05,
|
| 2165 |
+
"loss": 0.6149,
|
| 2166 |
+
"step": 299
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.0101010101010102,
|
| 2170 |
+
"grad_norm": 0.3359375,
|
| 2171 |
+
"learning_rate": 1.5820693536457787e-05,
|
| 2172 |
+
"loss": 0.6244,
|
| 2173 |
+
"step": 300
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.0134680134680134,
|
| 2177 |
+
"grad_norm": 0.318359375,
|
| 2178 |
+
"learning_rate": 1.5790493567627357e-05,
|
| 2179 |
+
"loss": 0.6052,
|
| 2180 |
+
"step": 301
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.0168350168350169,
|
| 2184 |
+
"grad_norm": 0.341796875,
|
| 2185 |
+
"learning_rate": 1.5760213937383032e-05,
|
| 2186 |
+
"loss": 0.5706,
|
| 2187 |
+
"step": 302
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.02020202020202,
|
| 2191 |
+
"grad_norm": 0.333984375,
|
| 2192 |
+
"learning_rate": 1.5729855062290024e-05,
|
| 2193 |
+
"loss": 0.5531,
|
| 2194 |
+
"step": 303
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.0235690235690236,
|
| 2198 |
+
"grad_norm": 0.30859375,
|
| 2199 |
+
"learning_rate": 1.5699417360003725e-05,
|
| 2200 |
+
"loss": 0.6111,
|
| 2201 |
+
"step": 304
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.0235690235690236,
|
| 2205 |
+
"eval_loss": 0.6120156049728394,
|
| 2206 |
+
"eval_runtime": 8.7268,
|
| 2207 |
+
"eval_samples_per_second": 57.295,
|
| 2208 |
+
"eval_steps_per_second": 3.667,
|
| 2209 |
+
"step": 304
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 1.026936026936027,
|
| 2213 |
+
"grad_norm": 0.314453125,
|
| 2214 |
+
"learning_rate": 1.5668901249263996e-05,
|
| 2215 |
+
"loss": 0.5895,
|
| 2216 |
+
"step": 305
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 1.0303030303030303,
|
| 2220 |
+
"grad_norm": 0.310546875,
|
| 2221 |
+
"learning_rate": 1.563830714988936e-05,
|
| 2222 |
+
"loss": 0.6005,
|
| 2223 |
+
"step": 306
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 1.0336700336700337,
|
| 2227 |
+
"grad_norm": 0.337890625,
|
| 2228 |
+
"learning_rate": 1.5607635482771272e-05,
|
| 2229 |
+
"loss": 0.6132,
|
| 2230 |
+
"step": 307
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 1.037037037037037,
|
| 2234 |
+
"grad_norm": 0.287109375,
|
| 2235 |
+
"learning_rate": 1.5576886669868297e-05,
|
| 2236 |
+
"loss": 0.5946,
|
| 2237 |
+
"step": 308
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 1.0404040404040404,
|
| 2241 |
+
"grad_norm": 0.30078125,
|
| 2242 |
+
"learning_rate": 1.5546061134200316e-05,
|
| 2243 |
+
"loss": 0.5838,
|
| 2244 |
+
"step": 309
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 1.0437710437710437,
|
| 2248 |
+
"grad_norm": 0.306640625,
|
| 2249 |
+
"learning_rate": 1.551515929984271e-05,
|
| 2250 |
+
"loss": 0.5992,
|
| 2251 |
+
"step": 310
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 1.0471380471380471,
|
| 2255 |
+
"grad_norm": 0.298828125,
|
| 2256 |
+
"learning_rate": 1.5484181591920516e-05,
|
| 2257 |
+
"loss": 0.5773,
|
| 2258 |
+
"step": 311
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 1.0505050505050506,
|
| 2262 |
+
"grad_norm": 0.318359375,
|
| 2263 |
+
"learning_rate": 1.5453128436602597e-05,
|
| 2264 |
+
"loss": 0.5757,
|
| 2265 |
+
"step": 312
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 1.0538720538720538,
|
| 2269 |
+
"grad_norm": 0.3515625,
|
| 2270 |
+
"learning_rate": 1.542200026109575e-05,
|
| 2271 |
+
"loss": 0.6248,
|
| 2272 |
+
"step": 313
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 1.0572390572390573,
|
| 2276 |
+
"grad_norm": 0.326171875,
|
| 2277 |
+
"learning_rate": 1.5390797493638862e-05,
|
| 2278 |
+
"loss": 0.6123,
|
| 2279 |
+
"step": 314
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 1.0606060606060606,
|
| 2283 |
+
"grad_norm": 0.322265625,
|
| 2284 |
+
"learning_rate": 1.5359520563496985e-05,
|
| 2285 |
+
"loss": 0.5815,
|
| 2286 |
+
"step": 315
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 1.063973063973064,
|
| 2290 |
+
"grad_norm": 0.287109375,
|
| 2291 |
+
"learning_rate": 1.5328169900955463e-05,
|
| 2292 |
+
"loss": 0.589,
|
| 2293 |
+
"step": 316
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 1.0673400673400673,
|
| 2297 |
+
"grad_norm": 0.328125,
|
| 2298 |
+
"learning_rate": 1.529674593731399e-05,
|
| 2299 |
+
"loss": 0.6343,
|
| 2300 |
+
"step": 317
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 1.0707070707070707,
|
| 2304 |
+
"grad_norm": 0.36328125,
|
| 2305 |
+
"learning_rate": 1.5265249104880675e-05,
|
| 2306 |
+
"loss": 0.6045,
|
| 2307 |
+
"step": 318
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 1.074074074074074,
|
| 2311 |
+
"grad_norm": 0.37890625,
|
| 2312 |
+
"learning_rate": 1.5233679836966122e-05,
|
| 2313 |
+
"loss": 0.5939,
|
| 2314 |
+
"step": 319
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 1.0774410774410774,
|
| 2318 |
+
"grad_norm": 0.328125,
|
| 2319 |
+
"learning_rate": 1.5202038567877436e-05,
|
| 2320 |
+
"loss": 0.6627,
|
| 2321 |
+
"step": 320
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 1.0808080808080809,
|
| 2325 |
+
"grad_norm": 0.326171875,
|
| 2326 |
+
"learning_rate": 1.517032573291226e-05,
|
| 2327 |
+
"loss": 0.616,
|
| 2328 |
+
"step": 321
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 1.0841750841750841,
|
| 2332 |
+
"grad_norm": 0.369140625,
|
| 2333 |
+
"learning_rate": 1.5138541768352802e-05,
|
| 2334 |
+
"loss": 0.587,
|
| 2335 |
+
"step": 322
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 1.0875420875420876,
|
| 2339 |
+
"grad_norm": 0.3359375,
|
| 2340 |
+
"learning_rate": 1.5106687111459809e-05,
|
| 2341 |
+
"loss": 0.586,
|
| 2342 |
+
"step": 323
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 1.0909090909090908,
|
| 2346 |
+
"grad_norm": 0.322265625,
|
| 2347 |
+
"learning_rate": 1.5074762200466557e-05,
|
| 2348 |
+
"loss": 0.6117,
|
| 2349 |
+
"step": 324
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 1.0942760942760943,
|
| 2353 |
+
"grad_norm": 0.34375,
|
| 2354 |
+
"learning_rate": 1.5042767474572846e-05,
|
| 2355 |
+
"loss": 0.6171,
|
| 2356 |
+
"step": 325
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 1.0976430976430978,
|
| 2360 |
+
"grad_norm": 0.349609375,
|
| 2361 |
+
"learning_rate": 1.5010703373938915e-05,
|
| 2362 |
+
"loss": 0.617,
|
| 2363 |
+
"step": 326
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 1.101010101010101,
|
| 2367 |
+
"grad_norm": 0.298828125,
|
| 2368 |
+
"learning_rate": 1.4978570339679435e-05,
|
| 2369 |
+
"loss": 0.5944,
|
| 2370 |
+
"step": 327
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 1.1043771043771045,
|
| 2374 |
+
"grad_norm": 0.30859375,
|
| 2375 |
+
"learning_rate": 1.4946368813857393e-05,
|
| 2376 |
+
"loss": 0.5867,
|
| 2377 |
+
"step": 328
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 1.1077441077441077,
|
| 2381 |
+
"grad_norm": 0.337890625,
|
| 2382 |
+
"learning_rate": 1.4914099239478046e-05,
|
| 2383 |
+
"loss": 0.5639,
|
| 2384 |
+
"step": 329
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 1.1111111111111112,
|
| 2388 |
+
"grad_norm": 0.34375,
|
| 2389 |
+
"learning_rate": 1.4881762060482814e-05,
|
| 2390 |
+
"loss": 0.6106,
|
| 2391 |
+
"step": 330
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 1.1144781144781144,
|
| 2395 |
+
"grad_norm": 0.328125,
|
| 2396 |
+
"learning_rate": 1.4849357721743169e-05,
|
| 2397 |
+
"loss": 0.5997,
|
| 2398 |
+
"step": 331
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 1.1178451178451179,
|
| 2402 |
+
"grad_norm": 0.33984375,
|
| 2403 |
+
"learning_rate": 1.4816886669054514e-05,
|
| 2404 |
+
"loss": 0.6194,
|
| 2405 |
+
"step": 332
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 1.121212121212121,
|
| 2409 |
+
"grad_norm": 0.310546875,
|
| 2410 |
+
"learning_rate": 1.4784349349130063e-05,
|
| 2411 |
+
"loss": 0.5837,
|
| 2412 |
+
"step": 333
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 1.1245791245791246,
|
| 2416 |
+
"grad_norm": 0.322265625,
|
| 2417 |
+
"learning_rate": 1.4751746209594683e-05,
|
| 2418 |
+
"loss": 0.5881,
|
| 2419 |
+
"step": 334
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 1.127946127946128,
|
| 2423 |
+
"grad_norm": 0.3046875,
|
| 2424 |
+
"learning_rate": 1.4719077698978737e-05,
|
| 2425 |
+
"loss": 0.5973,
|
| 2426 |
+
"step": 335
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 1.1313131313131313,
|
| 2430 |
+
"grad_norm": 0.314453125,
|
| 2431 |
+
"learning_rate": 1.4686344266711916e-05,
|
| 2432 |
+
"loss": 0.5861,
|
| 2433 |
+
"step": 336
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 1.1346801346801347,
|
| 2437 |
+
"grad_norm": 0.310546875,
|
| 2438 |
+
"learning_rate": 1.4653546363117063e-05,
|
| 2439 |
+
"loss": 0.5911,
|
| 2440 |
+
"step": 337
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 1.138047138047138,
|
| 2444 |
+
"grad_norm": 0.30078125,
|
| 2445 |
+
"learning_rate": 1.4620684439403962e-05,
|
| 2446 |
+
"loss": 0.5749,
|
| 2447 |
+
"step": 338
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 1.1414141414141414,
|
| 2451 |
+
"grad_norm": 0.30078125,
|
| 2452 |
+
"learning_rate": 1.4587758947663146e-05,
|
| 2453 |
+
"loss": 0.6134,
|
| 2454 |
+
"step": 339
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 1.144781144781145,
|
| 2458 |
+
"grad_norm": 0.318359375,
|
| 2459 |
+
"learning_rate": 1.4554770340859661e-05,
|
| 2460 |
+
"loss": 0.5948,
|
| 2461 |
+
"step": 340
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 1.1481481481481481,
|
| 2465 |
+
"grad_norm": 0.296875,
|
| 2466 |
+
"learning_rate": 1.4521719072826858e-05,
|
| 2467 |
+
"loss": 0.5687,
|
| 2468 |
+
"step": 341
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 1.1515151515151516,
|
| 2472 |
+
"grad_norm": 0.31640625,
|
| 2473 |
+
"learning_rate": 1.4488605598260129e-05,
|
| 2474 |
+
"loss": 0.6138,
|
| 2475 |
+
"step": 342
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 1.1515151515151516,
|
| 2479 |
+
"eval_loss": 0.6101293563842773,
|
| 2480 |
+
"eval_runtime": 8.7555,
|
| 2481 |
+
"eval_samples_per_second": 57.107,
|
| 2482 |
+
"eval_steps_per_second": 3.655,
|
| 2483 |
+
"step": 342
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 1.1548821548821548,
|
| 2487 |
+
"grad_norm": 0.30078125,
|
| 2488 |
+
"learning_rate": 1.4455430372710652e-05,
|
| 2489 |
+
"loss": 0.5619,
|
| 2490 |
+
"step": 343
|
| 2491 |
+
},
|
| 2492 |
+
{
|
| 2493 |
+
"epoch": 1.1582491582491583,
|
| 2494 |
+
"grad_norm": 0.34765625,
|
| 2495 |
+
"learning_rate": 1.4422193852579144e-05,
|
| 2496 |
+
"loss": 0.6349,
|
| 2497 |
+
"step": 344
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"epoch": 1.1616161616161615,
|
| 2501 |
+
"grad_norm": 0.32421875,
|
| 2502 |
+
"learning_rate": 1.438889649510956e-05,
|
| 2503 |
+
"loss": 0.5807,
|
| 2504 |
+
"step": 345
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 1.164983164983165,
|
| 2508 |
+
"grad_norm": 0.328125,
|
| 2509 |
+
"learning_rate": 1.4355538758382805e-05,
|
| 2510 |
+
"loss": 0.5874,
|
| 2511 |
+
"step": 346
|
| 2512 |
+
},
|
| 2513 |
+
{
|
| 2514 |
+
"epoch": 1.1683501683501682,
|
| 2515 |
+
"grad_norm": 0.357421875,
|
| 2516 |
+
"learning_rate": 1.4322121101310454e-05,
|
| 2517 |
+
"loss": 0.6472,
|
| 2518 |
+
"step": 347
|
| 2519 |
+
},
|
| 2520 |
+
{
|
| 2521 |
+
"epoch": 1.1717171717171717,
|
| 2522 |
+
"grad_norm": 0.369140625,
|
| 2523 |
+
"learning_rate": 1.428864398362841e-05,
|
| 2524 |
+
"loss": 0.638,
|
| 2525 |
+
"step": 348
|
| 2526 |
+
},
|
| 2527 |
+
{
|
| 2528 |
+
"epoch": 1.1750841750841752,
|
| 2529 |
+
"grad_norm": 0.318359375,
|
| 2530 |
+
"learning_rate": 1.4255107865890597e-05,
|
| 2531 |
+
"loss": 0.5989,
|
| 2532 |
+
"step": 349
|
| 2533 |
+
},
|
| 2534 |
+
{
|
| 2535 |
+
"epoch": 1.1784511784511784,
|
| 2536 |
+
"grad_norm": 0.302734375,
|
| 2537 |
+
"learning_rate": 1.4221513209462615e-05,
|
| 2538 |
+
"loss": 0.6196,
|
| 2539 |
+
"step": 350
|
| 2540 |
+
},
|
| 2541 |
+
{
|
| 2542 |
+
"epoch": 1.1818181818181819,
|
| 2543 |
+
"grad_norm": 0.37109375,
|
| 2544 |
+
"learning_rate": 1.41878604765154e-05,
|
| 2545 |
+
"loss": 0.6152,
|
| 2546 |
+
"step": 351
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 1.1851851851851851,
|
| 2550 |
+
"grad_norm": 0.2890625,
|
| 2551 |
+
"learning_rate": 1.4154150130018867e-05,
|
| 2552 |
+
"loss": 0.6104,
|
| 2553 |
+
"step": 352
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 1.1885521885521886,
|
| 2557 |
+
"grad_norm": 0.326171875,
|
| 2558 |
+
"learning_rate": 1.4120382633735528e-05,
|
| 2559 |
+
"loss": 0.6012,
|
| 2560 |
+
"step": 353
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 1.1919191919191918,
|
| 2564 |
+
"grad_norm": 0.318359375,
|
| 2565 |
+
"learning_rate": 1.4086558452214121e-05,
|
| 2566 |
+
"loss": 0.5788,
|
| 2567 |
+
"step": 354
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 1.1952861952861953,
|
| 2571 |
+
"grad_norm": 0.369140625,
|
| 2572 |
+
"learning_rate": 1.405267805078323e-05,
|
| 2573 |
+
"loss": 0.6059,
|
| 2574 |
+
"step": 355
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 1.1986531986531987,
|
| 2578 |
+
"grad_norm": 0.3359375,
|
| 2579 |
+
"learning_rate": 1.401874189554486e-05,
|
| 2580 |
+
"loss": 0.6137,
|
| 2581 |
+
"step": 356
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 1.202020202020202,
|
| 2585 |
+
"grad_norm": 0.322265625,
|
| 2586 |
+
"learning_rate": 1.3984750453368033e-05,
|
| 2587 |
+
"loss": 0.5951,
|
| 2588 |
+
"step": 357
|
| 2589 |
+
},
|
| 2590 |
+
{
|
| 2591 |
+
"epoch": 1.2053872053872055,
|
| 2592 |
+
"grad_norm": 0.34765625,
|
| 2593 |
+
"learning_rate": 1.3950704191882388e-05,
|
| 2594 |
+
"loss": 0.6272,
|
| 2595 |
+
"step": 358
|
| 2596 |
+
},
|
| 2597 |
+
{
|
| 2598 |
+
"epoch": 1.2087542087542087,
|
| 2599 |
+
"grad_norm": 0.3125,
|
| 2600 |
+
"learning_rate": 1.3916603579471705e-05,
|
| 2601 |
+
"loss": 0.5916,
|
| 2602 |
+
"step": 359
|
| 2603 |
+
},
|
| 2604 |
+
{
|
| 2605 |
+
"epoch": 1.2121212121212122,
|
| 2606 |
+
"grad_norm": 0.3203125,
|
| 2607 |
+
"learning_rate": 1.3882449085267497e-05,
|
| 2608 |
+
"loss": 0.5745,
|
| 2609 |
+
"step": 360
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 1.2154882154882154,
|
| 2613 |
+
"grad_norm": 0.34375,
|
| 2614 |
+
"learning_rate": 1.384824117914255e-05,
|
| 2615 |
+
"loss": 0.6189,
|
| 2616 |
+
"step": 361
|
| 2617 |
+
},
|
| 2618 |
+
{
|
| 2619 |
+
"epoch": 1.2188552188552189,
|
| 2620 |
+
"grad_norm": 0.302734375,
|
| 2621 |
+
"learning_rate": 1.3813980331704437e-05,
|
| 2622 |
+
"loss": 0.6032,
|
| 2623 |
+
"step": 362
|
| 2624 |
+
},
|
| 2625 |
+
{
|
| 2626 |
+
"epoch": 1.2222222222222223,
|
| 2627 |
+
"grad_norm": 0.314453125,
|
| 2628 |
+
"learning_rate": 1.3779667014289067e-05,
|
| 2629 |
+
"loss": 0.6202,
|
| 2630 |
+
"step": 363
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 1.2255892255892256,
|
| 2634 |
+
"grad_norm": 0.328125,
|
| 2635 |
+
"learning_rate": 1.3745301698954197e-05,
|
| 2636 |
+
"loss": 0.6272,
|
| 2637 |
+
"step": 364
|
| 2638 |
+
},
|
| 2639 |
+
{
|
| 2640 |
+
"epoch": 1.228956228956229,
|
| 2641 |
+
"grad_norm": 0.33984375,
|
| 2642 |
+
"learning_rate": 1.3710884858472926e-05,
|
| 2643 |
+
"loss": 0.6054,
|
| 2644 |
+
"step": 365
|
| 2645 |
+
},
|
| 2646 |
+
{
|
| 2647 |
+
"epoch": 1.2323232323232323,
|
| 2648 |
+
"grad_norm": 0.322265625,
|
| 2649 |
+
"learning_rate": 1.3676416966327201e-05,
|
| 2650 |
+
"loss": 0.5688,
|
| 2651 |
+
"step": 366
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"epoch": 1.2356902356902357,
|
| 2655 |
+
"grad_norm": 0.375,
|
| 2656 |
+
"learning_rate": 1.3641898496701306e-05,
|
| 2657 |
+
"loss": 0.6228,
|
| 2658 |
+
"step": 367
|
| 2659 |
+
},
|
| 2660 |
+
{
|
| 2661 |
+
"epoch": 1.239057239057239,
|
| 2662 |
+
"grad_norm": 0.302734375,
|
| 2663 |
+
"learning_rate": 1.360732992447533e-05,
|
| 2664 |
+
"loss": 0.5925,
|
| 2665 |
+
"step": 368
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 1.2424242424242424,
|
| 2669 |
+
"grad_norm": 0.326171875,
|
| 2670 |
+
"learning_rate": 1.357271172521863e-05,
|
| 2671 |
+
"loss": 0.5687,
|
| 2672 |
+
"step": 369
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 1.2457912457912457,
|
| 2676 |
+
"grad_norm": 0.283203125,
|
| 2677 |
+
"learning_rate": 1.3538044375183308e-05,
|
| 2678 |
+
"loss": 0.5483,
|
| 2679 |
+
"step": 370
|
| 2680 |
+
},
|
| 2681 |
+
{
|
| 2682 |
+
"epoch": 1.2491582491582491,
|
| 2683 |
+
"grad_norm": 0.32421875,
|
| 2684 |
+
"learning_rate": 1.3503328351297635e-05,
|
| 2685 |
+
"loss": 0.5925,
|
| 2686 |
+
"step": 371
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 1.2525252525252526,
|
| 2690 |
+
"grad_norm": 0.34375,
|
| 2691 |
+
"learning_rate": 1.3468564131159515e-05,
|
| 2692 |
+
"loss": 0.5782,
|
| 2693 |
+
"step": 372
|
| 2694 |
+
},
|
| 2695 |
+
{
|
| 2696 |
+
"epoch": 1.2558922558922558,
|
| 2697 |
+
"grad_norm": 0.318359375,
|
| 2698 |
+
"learning_rate": 1.3433752193029888e-05,
|
| 2699 |
+
"loss": 0.595,
|
| 2700 |
+
"step": 373
|
| 2701 |
+
},
|
| 2702 |
+
{
|
| 2703 |
+
"epoch": 1.2592592592592593,
|
| 2704 |
+
"grad_norm": 0.341796875,
|
| 2705 |
+
"learning_rate": 1.3398893015826166e-05,
|
| 2706 |
+
"loss": 0.6001,
|
| 2707 |
+
"step": 374
|
| 2708 |
+
},
|
| 2709 |
+
{
|
| 2710 |
+
"epoch": 1.2626262626262625,
|
| 2711 |
+
"grad_norm": 0.33984375,
|
| 2712 |
+
"learning_rate": 1.3363987079115656e-05,
|
| 2713 |
+
"loss": 0.6185,
|
| 2714 |
+
"step": 375
|
| 2715 |
+
},
|
| 2716 |
+
{
|
| 2717 |
+
"epoch": 1.265993265993266,
|
| 2718 |
+
"grad_norm": 0.287109375,
|
| 2719 |
+
"learning_rate": 1.3329034863108932e-05,
|
| 2720 |
+
"loss": 0.5738,
|
| 2721 |
+
"step": 376
|
| 2722 |
+
},
|
| 2723 |
+
{
|
| 2724 |
+
"epoch": 1.2693602693602695,
|
| 2725 |
+
"grad_norm": 0.302734375,
|
| 2726 |
+
"learning_rate": 1.3294036848653247e-05,
|
| 2727 |
+
"loss": 0.6121,
|
| 2728 |
+
"step": 377
|
| 2729 |
+
},
|
| 2730 |
+
{
|
| 2731 |
+
"epoch": 1.2727272727272727,
|
| 2732 |
+
"grad_norm": 0.31640625,
|
| 2733 |
+
"learning_rate": 1.3258993517225923e-05,
|
| 2734 |
+
"loss": 0.5751,
|
| 2735 |
+
"step": 378
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 1.2760942760942762,
|
| 2739 |
+
"grad_norm": 0.33203125,
|
| 2740 |
+
"learning_rate": 1.3223905350927718e-05,
|
| 2741 |
+
"loss": 0.6198,
|
| 2742 |
+
"step": 379
|
| 2743 |
+
},
|
| 2744 |
+
{
|
| 2745 |
+
"epoch": 1.2794612794612794,
|
| 2746 |
+
"grad_norm": 0.3359375,
|
| 2747 |
+
"learning_rate": 1.318877283247619e-05,
|
| 2748 |
+
"loss": 0.5938,
|
| 2749 |
+
"step": 380
|
| 2750 |
+
},
|
| 2751 |
+
{
|
| 2752 |
+
"epoch": 1.2794612794612794,
|
| 2753 |
+
"eval_loss": 0.6084044575691223,
|
| 2754 |
+
"eval_runtime": 8.685,
|
| 2755 |
+
"eval_samples_per_second": 57.57,
|
| 2756 |
+
"eval_steps_per_second": 3.685,
|
| 2757 |
+
"step": 380
|
| 2758 |
+
},
|
| 2759 |
+
{
|
| 2760 |
+
"epoch": 1.2828282828282829,
|
| 2761 |
+
"grad_norm": 0.375,
|
| 2762 |
+
"learning_rate": 1.3153596445199063e-05,
|
| 2763 |
+
"loss": 0.6104,
|
| 2764 |
+
"step": 381
|
| 2765 |
+
},
|
| 2766 |
+
{
|
| 2767 |
+
"epoch": 1.2861952861952861,
|
| 2768 |
+
"grad_norm": 0.31640625,
|
| 2769 |
+
"learning_rate": 1.3118376673027588e-05,
|
| 2770 |
+
"loss": 0.5953,
|
| 2771 |
+
"step": 382
|
| 2772 |
+
},
|
| 2773 |
+
{
|
| 2774 |
+
"epoch": 1.2895622895622896,
|
| 2775 |
+
"grad_norm": 0.326171875,
|
| 2776 |
+
"learning_rate": 1.3083114000489863e-05,
|
| 2777 |
+
"loss": 0.5892,
|
| 2778 |
+
"step": 383
|
| 2779 |
+
},
|
| 2780 |
+
{
|
| 2781 |
+
"epoch": 1.2929292929292928,
|
| 2782 |
+
"grad_norm": 0.36328125,
|
| 2783 |
+
"learning_rate": 1.3047808912704178e-05,
|
| 2784 |
+
"loss": 0.6184,
|
| 2785 |
+
"step": 384
|
| 2786 |
+
},
|
| 2787 |
+
{
|
| 2788 |
+
"epoch": 1.2962962962962963,
|
| 2789 |
+
"grad_norm": 0.30078125,
|
| 2790 |
+
"learning_rate": 1.3012461895372343e-05,
|
| 2791 |
+
"loss": 0.5926,
|
| 2792 |
+
"step": 385
|
| 2793 |
+
},
|
| 2794 |
+
{
|
| 2795 |
+
"epoch": 1.2996632996632997,
|
| 2796 |
+
"grad_norm": 0.296875,
|
| 2797 |
+
"learning_rate": 1.2977073434773009e-05,
|
| 2798 |
+
"loss": 0.571,
|
| 2799 |
+
"step": 386
|
| 2800 |
+
},
|
| 2801 |
+
{
|
| 2802 |
+
"epoch": 1.303030303030303,
|
| 2803 |
+
"grad_norm": 0.3203125,
|
| 2804 |
+
"learning_rate": 1.2941644017754964e-05,
|
| 2805 |
+
"loss": 0.6072,
|
| 2806 |
+
"step": 387
|
| 2807 |
+
},
|
| 2808 |
+
{
|
| 2809 |
+
"epoch": 1.3063973063973064,
|
| 2810 |
+
"grad_norm": 0.287109375,
|
| 2811 |
+
"learning_rate": 1.2906174131730458e-05,
|
| 2812 |
+
"loss": 0.5968,
|
| 2813 |
+
"step": 388
|
| 2814 |
+
},
|
| 2815 |
+
{
|
| 2816 |
+
"epoch": 1.3097643097643097,
|
| 2817 |
+
"grad_norm": 0.3125,
|
| 2818 |
+
"learning_rate": 1.2870664264668467e-05,
|
| 2819 |
+
"loss": 0.5982,
|
| 2820 |
+
"step": 389
|
| 2821 |
+
},
|
| 2822 |
+
{
|
| 2823 |
+
"epoch": 1.3131313131313131,
|
| 2824 |
+
"grad_norm": 0.326171875,
|
| 2825 |
+
"learning_rate": 1.2835114905088013e-05,
|
| 2826 |
+
"loss": 0.5947,
|
| 2827 |
+
"step": 390
|
| 2828 |
+
},
|
| 2829 |
+
{
|
| 2830 |
+
"epoch": 1.3164983164983166,
|
| 2831 |
+
"grad_norm": 0.33203125,
|
| 2832 |
+
"learning_rate": 1.2799526542051419e-05,
|
| 2833 |
+
"loss": 0.5942,
|
| 2834 |
+
"step": 391
|
| 2835 |
+
},
|
| 2836 |
+
{
|
| 2837 |
+
"epoch": 1.3198653198653199,
|
| 2838 |
+
"grad_norm": 0.380859375,
|
| 2839 |
+
"learning_rate": 1.2763899665157591e-05,
|
| 2840 |
+
"loss": 0.6106,
|
| 2841 |
+
"step": 392
|
| 2842 |
+
},
|
| 2843 |
+
{
|
| 2844 |
+
"epoch": 1.3232323232323233,
|
| 2845 |
+
"grad_norm": 0.31640625,
|
| 2846 |
+
"learning_rate": 1.2728234764535283e-05,
|
| 2847 |
+
"loss": 0.5929,
|
| 2848 |
+
"step": 393
|
| 2849 |
+
},
|
| 2850 |
+
{
|
| 2851 |
+
"epoch": 1.3265993265993266,
|
| 2852 |
+
"grad_norm": 0.29296875,
|
| 2853 |
+
"learning_rate": 1.2692532330836346e-05,
|
| 2854 |
+
"loss": 0.5774,
|
| 2855 |
+
"step": 394
|
| 2856 |
+
},
|
| 2857 |
+
{
|
| 2858 |
+
"epoch": 1.32996632996633,
|
| 2859 |
+
"grad_norm": 0.32421875,
|
| 2860 |
+
"learning_rate": 1.2656792855228993e-05,
|
| 2861 |
+
"loss": 0.6024,
|
| 2862 |
+
"step": 395
|
| 2863 |
+
},
|
| 2864 |
+
{
|
| 2865 |
+
"epoch": 1.3333333333333333,
|
| 2866 |
+
"grad_norm": 0.33203125,
|
| 2867 |
+
"learning_rate": 1.2621016829391022e-05,
|
| 2868 |
+
"loss": 0.6026,
|
| 2869 |
+
"step": 396
|
| 2870 |
+
},
|
| 2871 |
+
{
|
| 2872 |
+
"epoch": 1.3367003367003367,
|
| 2873 |
+
"grad_norm": 0.326171875,
|
| 2874 |
+
"learning_rate": 1.2585204745503072e-05,
|
| 2875 |
+
"loss": 0.5987,
|
| 2876 |
+
"step": 397
|
| 2877 |
+
},
|
| 2878 |
+
{
|
| 2879 |
+
"epoch": 1.34006734006734,
|
| 2880 |
+
"grad_norm": 0.30859375,
|
| 2881 |
+
"learning_rate": 1.2549357096241841e-05,
|
| 2882 |
+
"loss": 0.6175,
|
| 2883 |
+
"step": 398
|
| 2884 |
+
},
|
| 2885 |
+
{
|
| 2886 |
+
"epoch": 1.3434343434343434,
|
| 2887 |
+
"grad_norm": 0.33203125,
|
| 2888 |
+
"learning_rate": 1.25134743747733e-05,
|
| 2889 |
+
"loss": 0.606,
|
| 2890 |
+
"step": 399
|
| 2891 |
+
},
|
| 2892 |
+
{
|
| 2893 |
+
"epoch": 1.3468013468013469,
|
| 2894 |
+
"grad_norm": 0.306640625,
|
| 2895 |
+
"learning_rate": 1.2477557074745932e-05,
|
| 2896 |
+
"loss": 0.5935,
|
| 2897 |
+
"step": 400
|
| 2898 |
+
},
|
| 2899 |
+
{
|
| 2900 |
+
"epoch": 1.3501683501683501,
|
| 2901 |
+
"grad_norm": 0.294921875,
|
| 2902 |
+
"learning_rate": 1.2441605690283915e-05,
|
| 2903 |
+
"loss": 0.6087,
|
| 2904 |
+
"step": 401
|
| 2905 |
+
},
|
| 2906 |
+
{
|
| 2907 |
+
"epoch": 1.3535353535353536,
|
| 2908 |
+
"grad_norm": 0.326171875,
|
| 2909 |
+
"learning_rate": 1.2405620715980345e-05,
|
| 2910 |
+
"loss": 0.6107,
|
| 2911 |
+
"step": 402
|
| 2912 |
+
},
|
| 2913 |
+
{
|
| 2914 |
+
"epoch": 1.3569023569023568,
|
| 2915 |
+
"grad_norm": 0.294921875,
|
| 2916 |
+
"learning_rate": 1.2369602646890415e-05,
|
| 2917 |
+
"loss": 0.5924,
|
| 2918 |
+
"step": 403
|
| 2919 |
+
},
|
| 2920 |
+
{
|
| 2921 |
+
"epoch": 1.3602693602693603,
|
| 2922 |
+
"grad_norm": 0.314453125,
|
| 2923 |
+
"learning_rate": 1.2333551978524618e-05,
|
| 2924 |
+
"loss": 0.6308,
|
| 2925 |
+
"step": 404
|
| 2926 |
+
},
|
| 2927 |
+
{
|
| 2928 |
+
"epoch": 1.3636363636363638,
|
| 2929 |
+
"grad_norm": 0.328125,
|
| 2930 |
+
"learning_rate": 1.2297469206841921e-05,
|
| 2931 |
+
"loss": 0.6212,
|
| 2932 |
+
"step": 405
|
| 2933 |
+
},
|
| 2934 |
+
{
|
| 2935 |
+
"epoch": 1.367003367003367,
|
| 2936 |
+
"grad_norm": 0.30078125,
|
| 2937 |
+
"learning_rate": 1.2261354828242948e-05,
|
| 2938 |
+
"loss": 0.5882,
|
| 2939 |
+
"step": 406
|
| 2940 |
+
},
|
| 2941 |
+
{
|
| 2942 |
+
"epoch": 1.3703703703703702,
|
| 2943 |
+
"grad_norm": 0.296875,
|
| 2944 |
+
"learning_rate": 1.2225209339563144e-05,
|
| 2945 |
+
"loss": 0.6231,
|
| 2946 |
+
"step": 407
|
| 2947 |
+
},
|
| 2948 |
+
{
|
| 2949 |
+
"epoch": 1.3737373737373737,
|
| 2950 |
+
"grad_norm": 0.353515625,
|
| 2951 |
+
"learning_rate": 1.218903323806595e-05,
|
| 2952 |
+
"loss": 0.6054,
|
| 2953 |
+
"step": 408
|
| 2954 |
+
},
|
| 2955 |
+
{
|
| 2956 |
+
"epoch": 1.3771043771043772,
|
| 2957 |
+
"grad_norm": 0.38671875,
|
| 2958 |
+
"learning_rate": 1.2152827021435946e-05,
|
| 2959 |
+
"loss": 0.6463,
|
| 2960 |
+
"step": 409
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 1.3804713804713804,
|
| 2964 |
+
"grad_norm": 0.30078125,
|
| 2965 |
+
"learning_rate": 1.2116591187772026e-05,
|
| 2966 |
+
"loss": 0.6015,
|
| 2967 |
+
"step": 410
|
| 2968 |
+
},
|
| 2969 |
+
{
|
| 2970 |
+
"epoch": 1.3838383838383839,
|
| 2971 |
+
"grad_norm": 0.298828125,
|
| 2972 |
+
"learning_rate": 1.2080326235580521e-05,
|
| 2973 |
+
"loss": 0.5653,
|
| 2974 |
+
"step": 411
|
| 2975 |
+
},
|
| 2976 |
+
{
|
| 2977 |
+
"epoch": 1.387205387205387,
|
| 2978 |
+
"grad_norm": 0.349609375,
|
| 2979 |
+
"learning_rate": 1.204403266376837e-05,
|
| 2980 |
+
"loss": 0.5852,
|
| 2981 |
+
"step": 412
|
| 2982 |
+
},
|
| 2983 |
+
{
|
| 2984 |
+
"epoch": 1.3905723905723906,
|
| 2985 |
+
"grad_norm": 0.37890625,
|
| 2986 |
+
"learning_rate": 1.2007710971636221e-05,
|
| 2987 |
+
"loss": 0.5886,
|
| 2988 |
+
"step": 413
|
| 2989 |
+
},
|
| 2990 |
+
{
|
| 2991 |
+
"epoch": 1.393939393939394,
|
| 2992 |
+
"grad_norm": 0.349609375,
|
| 2993 |
+
"learning_rate": 1.1971361658871597e-05,
|
| 2994 |
+
"loss": 0.5933,
|
| 2995 |
+
"step": 414
|
| 2996 |
+
},
|
| 2997 |
+
{
|
| 2998 |
+
"epoch": 1.3973063973063973,
|
| 2999 |
+
"grad_norm": 0.333984375,
|
| 3000 |
+
"learning_rate": 1.1934985225541998e-05,
|
| 3001 |
+
"loss": 0.6065,
|
| 3002 |
+
"step": 415
|
| 3003 |
+
},
|
| 3004 |
+
{
|
| 3005 |
+
"epoch": 1.4006734006734007,
|
| 3006 |
+
"grad_norm": 0.33984375,
|
| 3007 |
+
"learning_rate": 1.1898582172088027e-05,
|
| 3008 |
+
"loss": 0.595,
|
| 3009 |
+
"step": 416
|
| 3010 |
+
},
|
| 3011 |
+
{
|
| 3012 |
+
"epoch": 1.404040404040404,
|
| 3013 |
+
"grad_norm": 0.30859375,
|
| 3014 |
+
"learning_rate": 1.1862152999316517e-05,
|
| 3015 |
+
"loss": 0.6061,
|
| 3016 |
+
"step": 417
|
| 3017 |
+
},
|
| 3018 |
+
{
|
| 3019 |
+
"epoch": 1.4074074074074074,
|
| 3020 |
+
"grad_norm": 0.31640625,
|
| 3021 |
+
"learning_rate": 1.182569820839362e-05,
|
| 3022 |
+
"loss": 0.5592,
|
| 3023 |
+
"step": 418
|
| 3024 |
+
},
|
| 3025 |
+
{
|
| 3026 |
+
"epoch": 1.4074074074074074,
|
| 3027 |
+
"eval_loss": 0.6071019768714905,
|
| 3028 |
+
"eval_runtime": 8.651,
|
| 3029 |
+
"eval_samples_per_second": 57.797,
|
| 3030 |
+
"eval_steps_per_second": 3.699,
|
| 3031 |
+
"step": 418
|
| 3032 |
+
},
|
| 3033 |
+
{
|
| 3034 |
+
"epoch": 1.410774410774411,
|
| 3035 |
+
"grad_norm": 0.333984375,
|
| 3036 |
+
"learning_rate": 1.1789218300837929e-05,
|
| 3037 |
+
"loss": 0.5964,
|
| 3038 |
+
"step": 419
|
| 3039 |
+
},
|
| 3040 |
+
{
|
| 3041 |
+
"epoch": 1.4141414141414141,
|
| 3042 |
+
"grad_norm": 0.296875,
|
| 3043 |
+
"learning_rate": 1.1752713778513576e-05,
|
| 3044 |
+
"loss": 0.588,
|
| 3045 |
+
"step": 420
|
| 3046 |
+
},
|
| 3047 |
+
{
|
| 3048 |
+
"epoch": 1.4175084175084174,
|
| 3049 |
+
"grad_norm": 0.28515625,
|
| 3050 |
+
"learning_rate": 1.1716185143623322e-05,
|
| 3051 |
+
"loss": 0.5954,
|
| 3052 |
+
"step": 421
|
| 3053 |
+
},
|
| 3054 |
+
{
|
| 3055 |
+
"epoch": 1.4208754208754208,
|
| 3056 |
+
"grad_norm": 0.296875,
|
| 3057 |
+
"learning_rate": 1.1679632898701649e-05,
|
| 3058 |
+
"loss": 0.5887,
|
| 3059 |
+
"step": 422
|
| 3060 |
+
},
|
| 3061 |
+
{
|
| 3062 |
+
"epoch": 1.4242424242424243,
|
| 3063 |
+
"grad_norm": 0.302734375,
|
| 3064 |
+
"learning_rate": 1.1643057546607858e-05,
|
| 3065 |
+
"loss": 0.6194,
|
| 3066 |
+
"step": 423
|
| 3067 |
+
},
|
| 3068 |
+
{
|
| 3069 |
+
"epoch": 1.4276094276094276,
|
| 3070 |
+
"grad_norm": 0.29296875,
|
| 3071 |
+
"learning_rate": 1.1606459590519132e-05,
|
| 3072 |
+
"loss": 0.5773,
|
| 3073 |
+
"step": 424
|
| 3074 |
+
},
|
| 3075 |
+
{
|
| 3076 |
+
"epoch": 1.430976430976431,
|
| 3077 |
+
"grad_norm": 0.328125,
|
| 3078 |
+
"learning_rate": 1.1569839533923626e-05,
|
| 3079 |
+
"loss": 0.6031,
|
| 3080 |
+
"step": 425
|
| 3081 |
+
},
|
| 3082 |
+
{
|
| 3083 |
+
"epoch": 1.4343434343434343,
|
| 3084 |
+
"grad_norm": 0.318359375,
|
| 3085 |
+
"learning_rate": 1.153319788061354e-05,
|
| 3086 |
+
"loss": 0.6102,
|
| 3087 |
+
"step": 426
|
| 3088 |
+
},
|
| 3089 |
+
{
|
| 3090 |
+
"epoch": 1.4377104377104377,
|
| 3091 |
+
"grad_norm": 0.31640625,
|
| 3092 |
+
"learning_rate": 1.1496535134678186e-05,
|
| 3093 |
+
"loss": 0.6135,
|
| 3094 |
+
"step": 427
|
| 3095 |
+
},
|
| 3096 |
+
{
|
| 3097 |
+
"epoch": 1.4410774410774412,
|
| 3098 |
+
"grad_norm": 0.279296875,
|
| 3099 |
+
"learning_rate": 1.1459851800497048e-05,
|
| 3100 |
+
"loss": 0.5686,
|
| 3101 |
+
"step": 428
|
| 3102 |
+
},
|
| 3103 |
+
{
|
| 3104 |
+
"epoch": 1.4444444444444444,
|
| 3105 |
+
"grad_norm": 0.3125,
|
| 3106 |
+
"learning_rate": 1.1423148382732854e-05,
|
| 3107 |
+
"loss": 0.5578,
|
| 3108 |
+
"step": 429
|
| 3109 |
+
},
|
| 3110 |
+
{
|
| 3111 |
+
"epoch": 1.4478114478114479,
|
| 3112 |
+
"grad_norm": 0.421875,
|
| 3113 |
+
"learning_rate": 1.1386425386324622e-05,
|
| 3114 |
+
"loss": 0.6305,
|
| 3115 |
+
"step": 430
|
| 3116 |
+
},
|
| 3117 |
+
{
|
| 3118 |
+
"epoch": 1.4511784511784511,
|
| 3119 |
+
"grad_norm": 0.30859375,
|
| 3120 |
+
"learning_rate": 1.134968331648073e-05,
|
| 3121 |
+
"loss": 0.6194,
|
| 3122 |
+
"step": 431
|
| 3123 |
+
},
|
| 3124 |
+
{
|
| 3125 |
+
"epoch": 1.4545454545454546,
|
| 3126 |
+
"grad_norm": 0.318359375,
|
| 3127 |
+
"learning_rate": 1.1312922678671935e-05,
|
| 3128 |
+
"loss": 0.5963,
|
| 3129 |
+
"step": 432
|
| 3130 |
+
},
|
| 3131 |
+
{
|
| 3132 |
+
"epoch": 1.457912457912458,
|
| 3133 |
+
"grad_norm": 0.3203125,
|
| 3134 |
+
"learning_rate": 1.1276143978624457e-05,
|
| 3135 |
+
"loss": 0.627,
|
| 3136 |
+
"step": 433
|
| 3137 |
+
},
|
| 3138 |
+
{
|
| 3139 |
+
"epoch": 1.4612794612794613,
|
| 3140 |
+
"grad_norm": 0.302734375,
|
| 3141 |
+
"learning_rate": 1.1239347722312997e-05,
|
| 3142 |
+
"loss": 0.587,
|
| 3143 |
+
"step": 434
|
| 3144 |
+
},
|
| 3145 |
+
{
|
| 3146 |
+
"epoch": 1.4646464646464645,
|
| 3147 |
+
"grad_norm": 0.283203125,
|
| 3148 |
+
"learning_rate": 1.120253441595378e-05,
|
| 3149 |
+
"loss": 0.5845,
|
| 3150 |
+
"step": 435
|
| 3151 |
+
},
|
| 3152 |
+
{
|
| 3153 |
+
"epoch": 1.468013468013468,
|
| 3154 |
+
"grad_norm": 0.3359375,
|
| 3155 |
+
"learning_rate": 1.1165704565997593e-05,
|
| 3156 |
+
"loss": 0.5788,
|
| 3157 |
+
"step": 436
|
| 3158 |
+
},
|
| 3159 |
+
{
|
| 3160 |
+
"epoch": 1.4713804713804715,
|
| 3161 |
+
"grad_norm": 0.328125,
|
| 3162 |
+
"learning_rate": 1.1128858679122822e-05,
|
| 3163 |
+
"loss": 0.571,
|
| 3164 |
+
"step": 437
|
| 3165 |
+
},
|
| 3166 |
+
{
|
| 3167 |
+
"epoch": 1.4747474747474747,
|
| 3168 |
+
"grad_norm": 0.4140625,
|
| 3169 |
+
"learning_rate": 1.1091997262228473e-05,
|
| 3170 |
+
"loss": 0.6235,
|
| 3171 |
+
"step": 438
|
| 3172 |
+
},
|
| 3173 |
+
{
|
| 3174 |
+
"epoch": 1.4781144781144782,
|
| 3175 |
+
"grad_norm": 0.296875,
|
| 3176 |
+
"learning_rate": 1.1055120822427208e-05,
|
| 3177 |
+
"loss": 0.5438,
|
| 3178 |
+
"step": 439
|
| 3179 |
+
},
|
| 3180 |
+
{
|
| 3181 |
+
"epoch": 1.4814814814814814,
|
| 3182 |
+
"grad_norm": 0.32421875,
|
| 3183 |
+
"learning_rate": 1.1018229867038358e-05,
|
| 3184 |
+
"loss": 0.5704,
|
| 3185 |
+
"step": 440
|
| 3186 |
+
},
|
| 3187 |
+
{
|
| 3188 |
+
"epoch": 1.4848484848484849,
|
| 3189 |
+
"grad_norm": 0.3046875,
|
| 3190 |
+
"learning_rate": 1.0981324903580945e-05,
|
| 3191 |
+
"loss": 0.5693,
|
| 3192 |
+
"step": 441
|
| 3193 |
+
},
|
| 3194 |
+
{
|
| 3195 |
+
"epoch": 1.4882154882154883,
|
| 3196 |
+
"grad_norm": 0.318359375,
|
| 3197 |
+
"learning_rate": 1.0944406439766719e-05,
|
| 3198 |
+
"loss": 0.6106,
|
| 3199 |
+
"step": 442
|
| 3200 |
+
},
|
| 3201 |
+
{
|
| 3202 |
+
"epoch": 1.4915824915824916,
|
| 3203 |
+
"grad_norm": 0.333984375,
|
| 3204 |
+
"learning_rate": 1.0907474983493144e-05,
|
| 3205 |
+
"loss": 0.6313,
|
| 3206 |
+
"step": 443
|
| 3207 |
+
},
|
| 3208 |
+
{
|
| 3209 |
+
"epoch": 1.494949494949495,
|
| 3210 |
+
"grad_norm": 0.287109375,
|
| 3211 |
+
"learning_rate": 1.087053104283643e-05,
|
| 3212 |
+
"loss": 0.5699,
|
| 3213 |
+
"step": 444
|
| 3214 |
+
},
|
| 3215 |
+
{
|
| 3216 |
+
"epoch": 1.4983164983164983,
|
| 3217 |
+
"grad_norm": 0.29296875,
|
| 3218 |
+
"learning_rate": 1.083357512604454e-05,
|
| 3219 |
+
"loss": 0.5806,
|
| 3220 |
+
"step": 445
|
| 3221 |
+
},
|
| 3222 |
+
{
|
| 3223 |
+
"epoch": 1.5016835016835017,
|
| 3224 |
+
"grad_norm": 0.298828125,
|
| 3225 |
+
"learning_rate": 1.0796607741530191e-05,
|
| 3226 |
+
"loss": 0.5922,
|
| 3227 |
+
"step": 446
|
| 3228 |
+
},
|
| 3229 |
+
{
|
| 3230 |
+
"epoch": 1.5050505050505052,
|
| 3231 |
+
"grad_norm": 0.314453125,
|
| 3232 |
+
"learning_rate": 1.075962939786387e-05,
|
| 3233 |
+
"loss": 0.6123,
|
| 3234 |
+
"step": 447
|
| 3235 |
+
},
|
| 3236 |
+
{
|
| 3237 |
+
"epoch": 1.5084175084175084,
|
| 3238 |
+
"grad_norm": 0.2890625,
|
| 3239 |
+
"learning_rate": 1.0722640603766825e-05,
|
| 3240 |
+
"loss": 0.5926,
|
| 3241 |
+
"step": 448
|
| 3242 |
+
},
|
| 3243 |
+
{
|
| 3244 |
+
"epoch": 1.5117845117845117,
|
| 3245 |
+
"grad_norm": 0.412109375,
|
| 3246 |
+
"learning_rate": 1.0685641868104085e-05,
|
| 3247 |
+
"loss": 0.6335,
|
| 3248 |
+
"step": 449
|
| 3249 |
+
},
|
| 3250 |
+
{
|
| 3251 |
+
"epoch": 1.5151515151515151,
|
| 3252 |
+
"grad_norm": 0.330078125,
|
| 3253 |
+
"learning_rate": 1.064863369987743e-05,
|
| 3254 |
+
"loss": 0.5997,
|
| 3255 |
+
"step": 450
|
| 3256 |
+
},
|
| 3257 |
+
{
|
| 3258 |
+
"epoch": 1.5185185185185186,
|
| 3259 |
+
"grad_norm": 0.330078125,
|
| 3260 |
+
"learning_rate": 1.0611616608218429e-05,
|
| 3261 |
+
"loss": 0.6001,
|
| 3262 |
+
"step": 451
|
| 3263 |
+
},
|
| 3264 |
+
{
|
| 3265 |
+
"epoch": 1.5218855218855218,
|
| 3266 |
+
"grad_norm": 0.298828125,
|
| 3267 |
+
"learning_rate": 1.0574591102381395e-05,
|
| 3268 |
+
"loss": 0.6193,
|
| 3269 |
+
"step": 452
|
| 3270 |
+
},
|
| 3271 |
+
{
|
| 3272 |
+
"epoch": 1.5252525252525253,
|
| 3273 |
+
"grad_norm": 0.365234375,
|
| 3274 |
+
"learning_rate": 1.0537557691736402e-05,
|
| 3275 |
+
"loss": 0.5833,
|
| 3276 |
+
"step": 453
|
| 3277 |
+
},
|
| 3278 |
+
{
|
| 3279 |
+
"epoch": 1.5286195286195285,
|
| 3280 |
+
"grad_norm": 0.359375,
|
| 3281 |
+
"learning_rate": 1.0500516885762278e-05,
|
| 3282 |
+
"loss": 0.6065,
|
| 3283 |
+
"step": 454
|
| 3284 |
+
},
|
| 3285 |
+
{
|
| 3286 |
+
"epoch": 1.531986531986532,
|
| 3287 |
+
"grad_norm": 0.349609375,
|
| 3288 |
+
"learning_rate": 1.0463469194039584e-05,
|
| 3289 |
+
"loss": 0.6208,
|
| 3290 |
+
"step": 455
|
| 3291 |
+
},
|
| 3292 |
+
{
|
| 3293 |
+
"epoch": 1.5353535353535355,
|
| 3294 |
+
"grad_norm": 0.291015625,
|
| 3295 |
+
"learning_rate": 1.0426415126243615e-05,
|
| 3296 |
+
"loss": 0.6041,
|
| 3297 |
+
"step": 456
|
| 3298 |
+
},
|
| 3299 |
+
{
|
| 3300 |
+
"epoch": 1.5353535353535355,
|
| 3301 |
+
"eval_loss": 0.6061655282974243,
|
| 3302 |
+
"eval_runtime": 8.7167,
|
| 3303 |
+
"eval_samples_per_second": 57.361,
|
| 3304 |
+
"eval_steps_per_second": 3.671,
|
| 3305 |
+
"step": 456
|
| 3306 |
+
},
|
| 3307 |
+
{
|
| 3308 |
+
"epoch": 1.5387205387205387,
|
| 3309 |
+
"grad_norm": 0.330078125,
|
| 3310 |
+
"learning_rate": 1.0389355192137379e-05,
|
| 3311 |
+
"loss": 0.5762,
|
| 3312 |
+
"step": 457
|
| 3313 |
+
},
|
| 3314 |
+
{
|
| 3315 |
+
"epoch": 1.542087542087542,
|
| 3316 |
+
"grad_norm": 0.306640625,
|
| 3317 |
+
"learning_rate": 1.0352289901564592e-05,
|
| 3318 |
+
"loss": 0.6152,
|
| 3319 |
+
"step": 458
|
| 3320 |
+
},
|
| 3321 |
+
{
|
| 3322 |
+
"epoch": 1.5454545454545454,
|
| 3323 |
+
"grad_norm": 0.3203125,
|
| 3324 |
+
"learning_rate": 1.0315219764442657e-05,
|
| 3325 |
+
"loss": 0.617,
|
| 3326 |
+
"step": 459
|
| 3327 |
+
},
|
| 3328 |
+
{
|
| 3329 |
+
"epoch": 1.5488215488215489,
|
| 3330 |
+
"grad_norm": 0.291015625,
|
| 3331 |
+
"learning_rate": 1.0278145290755657e-05,
|
| 3332 |
+
"loss": 0.5459,
|
| 3333 |
+
"step": 460
|
| 3334 |
+
},
|
| 3335 |
+
{
|
| 3336 |
+
"epoch": 1.5521885521885523,
|
| 3337 |
+
"grad_norm": 0.330078125,
|
| 3338 |
+
"learning_rate": 1.0241066990547328e-05,
|
| 3339 |
+
"loss": 0.5702,
|
| 3340 |
+
"step": 461
|
| 3341 |
+
},
|
| 3342 |
+
{
|
| 3343 |
+
"epoch": 1.5555555555555556,
|
| 3344 |
+
"grad_norm": 0.30859375,
|
| 3345 |
+
"learning_rate": 1.0203985373914056e-05,
|
| 3346 |
+
"loss": 0.591,
|
| 3347 |
+
"step": 462
|
| 3348 |
+
},
|
| 3349 |
+
{
|
| 3350 |
+
"epoch": 1.5589225589225588,
|
| 3351 |
+
"grad_norm": 0.33984375,
|
| 3352 |
+
"learning_rate": 1.0166900950997845e-05,
|
| 3353 |
+
"loss": 0.622,
|
| 3354 |
+
"step": 463
|
| 3355 |
+
},
|
| 3356 |
+
{
|
| 3357 |
+
"epoch": 1.5622895622895623,
|
| 3358 |
+
"grad_norm": 0.318359375,
|
| 3359 |
+
"learning_rate": 1.012981423197931e-05,
|
| 3360 |
+
"loss": 0.6427,
|
| 3361 |
+
"step": 464
|
| 3362 |
+
},
|
| 3363 |
+
{
|
| 3364 |
+
"epoch": 1.5656565656565657,
|
| 3365 |
+
"grad_norm": 0.310546875,
|
| 3366 |
+
"learning_rate": 1.0092725727070653e-05,
|
| 3367 |
+
"loss": 0.5941,
|
| 3368 |
+
"step": 465
|
| 3369 |
+
},
|
| 3370 |
+
{
|
| 3371 |
+
"epoch": 1.569023569023569,
|
| 3372 |
+
"grad_norm": 0.30859375,
|
| 3373 |
+
"learning_rate": 1.0055635946508649e-05,
|
| 3374 |
+
"loss": 0.622,
|
| 3375 |
+
"step": 466
|
| 3376 |
+
},
|
| 3377 |
+
{
|
| 3378 |
+
"epoch": 1.5723905723905722,
|
| 3379 |
+
"grad_norm": 0.484375,
|
| 3380 |
+
"learning_rate": 1.0018545400547609e-05,
|
| 3381 |
+
"loss": 0.591,
|
| 3382 |
+
"step": 467
|
| 3383 |
+
},
|
| 3384 |
+
{
|
| 3385 |
+
"epoch": 1.5757575757575757,
|
| 3386 |
+
"grad_norm": 0.291015625,
|
| 3387 |
+
"learning_rate": 9.981454599452391e-06,
|
| 3388 |
+
"loss": 0.5814,
|
| 3389 |
+
"step": 468
|
| 3390 |
+
},
|
| 3391 |
+
{
|
| 3392 |
+
"epoch": 1.5791245791245792,
|
| 3393 |
+
"grad_norm": 0.330078125,
|
| 3394 |
+
"learning_rate": 9.944364053491356e-06,
|
| 3395 |
+
"loss": 0.6045,
|
| 3396 |
+
"step": 469
|
| 3397 |
+
},
|
| 3398 |
+
{
|
| 3399 |
+
"epoch": 1.5824915824915826,
|
| 3400 |
+
"grad_norm": 0.330078125,
|
| 3401 |
+
"learning_rate": 9.90727427292935e-06,
|
| 3402 |
+
"loss": 0.562,
|
| 3403 |
+
"step": 470
|
| 3404 |
+
},
|
| 3405 |
+
{
|
| 3406 |
+
"epoch": 1.5858585858585859,
|
| 3407 |
+
"grad_norm": 0.3125,
|
| 3408 |
+
"learning_rate": 9.870185768020694e-06,
|
| 3409 |
+
"loss": 0.5714,
|
| 3410 |
+
"step": 471
|
| 3411 |
+
},
|
| 3412 |
+
{
|
| 3413 |
+
"epoch": 1.589225589225589,
|
| 3414 |
+
"grad_norm": 0.31640625,
|
| 3415 |
+
"learning_rate": 9.83309904900216e-06,
|
| 3416 |
+
"loss": 0.5765,
|
| 3417 |
+
"step": 472
|
| 3418 |
+
},
|
| 3419 |
+
{
|
| 3420 |
+
"epoch": 1.5925925925925926,
|
| 3421 |
+
"grad_norm": 0.287109375,
|
| 3422 |
+
"learning_rate": 9.79601462608595e-06,
|
| 3423 |
+
"loss": 0.5943,
|
| 3424 |
+
"step": 473
|
| 3425 |
+
},
|
| 3426 |
+
{
|
| 3427 |
+
"epoch": 1.595959595959596,
|
| 3428 |
+
"grad_norm": 0.30078125,
|
| 3429 |
+
"learning_rate": 9.758933009452674e-06,
|
| 3430 |
+
"loss": 0.6049,
|
| 3431 |
+
"step": 474
|
| 3432 |
+
},
|
| 3433 |
+
{
|
| 3434 |
+
"epoch": 1.5993265993265995,
|
| 3435 |
+
"grad_norm": 0.32421875,
|
| 3436 |
+
"learning_rate": 9.721854709244346e-06,
|
| 3437 |
+
"loss": 0.6075,
|
| 3438 |
+
"step": 475
|
| 3439 |
+
},
|
| 3440 |
+
{
|
| 3441 |
+
"epoch": 1.6026936026936027,
|
| 3442 |
+
"grad_norm": 0.37109375,
|
| 3443 |
+
"learning_rate": 9.684780235557346e-06,
|
| 3444 |
+
"loss": 0.6078,
|
| 3445 |
+
"step": 476
|
| 3446 |
+
},
|
| 3447 |
+
{
|
| 3448 |
+
"epoch": 1.606060606060606,
|
| 3449 |
+
"grad_norm": 0.328125,
|
| 3450 |
+
"learning_rate": 9.647710098435413e-06,
|
| 3451 |
+
"loss": 0.6221,
|
| 3452 |
+
"step": 477
|
| 3453 |
+
},
|
| 3454 |
+
{
|
| 3455 |
+
"epoch": 1.6094276094276094,
|
| 3456 |
+
"grad_norm": 0.328125,
|
| 3457 |
+
"learning_rate": 9.610644807862625e-06,
|
| 3458 |
+
"loss": 0.6269,
|
| 3459 |
+
"step": 478
|
| 3460 |
+
},
|
| 3461 |
+
{
|
| 3462 |
+
"epoch": 1.612794612794613,
|
| 3463 |
+
"grad_norm": 0.32421875,
|
| 3464 |
+
"learning_rate": 9.573584873756387e-06,
|
| 3465 |
+
"loss": 0.6095,
|
| 3466 |
+
"step": 479
|
| 3467 |
+
},
|
| 3468 |
+
{
|
| 3469 |
+
"epoch": 1.6161616161616161,
|
| 3470 |
+
"grad_norm": 0.310546875,
|
| 3471 |
+
"learning_rate": 9.536530805960418e-06,
|
| 3472 |
+
"loss": 0.5573,
|
| 3473 |
+
"step": 480
|
| 3474 |
+
},
|
| 3475 |
+
{
|
| 3476 |
+
"epoch": 1.6195286195286194,
|
| 3477 |
+
"grad_norm": 0.30859375,
|
| 3478 |
+
"learning_rate": 9.499483114237726e-06,
|
| 3479 |
+
"loss": 0.6072,
|
| 3480 |
+
"step": 481
|
| 3481 |
+
},
|
| 3482 |
+
{
|
| 3483 |
+
"epoch": 1.6228956228956228,
|
| 3484 |
+
"grad_norm": 0.3125,
|
| 3485 |
+
"learning_rate": 9.4624423082636e-06,
|
| 3486 |
+
"loss": 0.5876,
|
| 3487 |
+
"step": 482
|
| 3488 |
+
},
|
| 3489 |
+
{
|
| 3490 |
+
"epoch": 1.6262626262626263,
|
| 3491 |
+
"grad_norm": 0.287109375,
|
| 3492 |
+
"learning_rate": 9.42540889761861e-06,
|
| 3493 |
+
"loss": 0.605,
|
| 3494 |
+
"step": 483
|
| 3495 |
+
},
|
| 3496 |
+
{
|
| 3497 |
+
"epoch": 1.6296296296296298,
|
| 3498 |
+
"grad_norm": 0.32421875,
|
| 3499 |
+
"learning_rate": 9.388383391781576e-06,
|
| 3500 |
+
"loss": 0.5952,
|
| 3501 |
+
"step": 484
|
| 3502 |
+
},
|
| 3503 |
+
{
|
| 3504 |
+
"epoch": 1.632996632996633,
|
| 3505 |
+
"grad_norm": 0.2890625,
|
| 3506 |
+
"learning_rate": 9.351366300122569e-06,
|
| 3507 |
+
"loss": 0.6113,
|
| 3508 |
+
"step": 485
|
| 3509 |
+
},
|
| 3510 |
+
{
|
| 3511 |
+
"epoch": 1.6363636363636362,
|
| 3512 |
+
"grad_norm": 0.330078125,
|
| 3513 |
+
"learning_rate": 9.314358131895919e-06,
|
| 3514 |
+
"loss": 0.6207,
|
| 3515 |
+
"step": 486
|
| 3516 |
+
},
|
| 3517 |
+
{
|
| 3518 |
+
"epoch": 1.6397306397306397,
|
| 3519 |
+
"grad_norm": 0.35546875,
|
| 3520 |
+
"learning_rate": 9.277359396233176e-06,
|
| 3521 |
+
"loss": 0.6038,
|
| 3522 |
+
"step": 487
|
| 3523 |
+
},
|
| 3524 |
+
{
|
| 3525 |
+
"epoch": 1.6430976430976432,
|
| 3526 |
+
"grad_norm": 0.330078125,
|
| 3527 |
+
"learning_rate": 9.240370602136134e-06,
|
| 3528 |
+
"loss": 0.5764,
|
| 3529 |
+
"step": 488
|
| 3530 |
+
},
|
| 3531 |
+
{
|
| 3532 |
+
"epoch": 1.6464646464646466,
|
| 3533 |
+
"grad_norm": 0.3125,
|
| 3534 |
+
"learning_rate": 9.203392258469814e-06,
|
| 3535 |
+
"loss": 0.5874,
|
| 3536 |
+
"step": 489
|
| 3537 |
+
},
|
| 3538 |
+
{
|
| 3539 |
+
"epoch": 1.6498316498316499,
|
| 3540 |
+
"grad_norm": 0.3125,
|
| 3541 |
+
"learning_rate": 9.166424873955462e-06,
|
| 3542 |
+
"loss": 0.6086,
|
| 3543 |
+
"step": 490
|
| 3544 |
+
},
|
| 3545 |
+
{
|
| 3546 |
+
"epoch": 1.6531986531986531,
|
| 3547 |
+
"grad_norm": 0.326171875,
|
| 3548 |
+
"learning_rate": 9.12946895716357e-06,
|
| 3549 |
+
"loss": 0.6301,
|
| 3550 |
+
"step": 491
|
| 3551 |
+
},
|
| 3552 |
+
{
|
| 3553 |
+
"epoch": 1.6565656565656566,
|
| 3554 |
+
"grad_norm": 0.330078125,
|
| 3555 |
+
"learning_rate": 9.092525016506858e-06,
|
| 3556 |
+
"loss": 0.631,
|
| 3557 |
+
"step": 492
|
| 3558 |
+
},
|
| 3559 |
+
{
|
| 3560 |
+
"epoch": 1.65993265993266,
|
| 3561 |
+
"grad_norm": 0.353515625,
|
| 3562 |
+
"learning_rate": 9.055593560233284e-06,
|
| 3563 |
+
"loss": 0.6312,
|
| 3564 |
+
"step": 493
|
| 3565 |
+
},
|
| 3566 |
+
{
|
| 3567 |
+
"epoch": 1.6632996632996633,
|
| 3568 |
+
"grad_norm": 0.310546875,
|
| 3569 |
+
"learning_rate": 9.018675096419059e-06,
|
| 3570 |
+
"loss": 0.5768,
|
| 3571 |
+
"step": 494
|
| 3572 |
+
},
|
| 3573 |
+
{
|
| 3574 |
+
"epoch": 1.6632996632996633,
|
| 3575 |
+
"eval_loss": 0.6054856777191162,
|
| 3576 |
+
"eval_runtime": 8.7005,
|
| 3577 |
+
"eval_samples_per_second": 57.468,
|
| 3578 |
+
"eval_steps_per_second": 3.678,
|
| 3579 |
+
"step": 494
|
| 3580 |
+
},
|
| 3581 |
+
{
|
| 3582 |
+
"epoch": 1.6666666666666665,
|
| 3583 |
+
"grad_norm": 0.30078125,
|
| 3584 |
+
"learning_rate": 8.981770132961649e-06,
|
| 3585 |
+
"loss": 0.5698,
|
| 3586 |
+
"step": 495
|
| 3587 |
+
},
|
| 3588 |
+
{
|
| 3589 |
+
"epoch": 1.67003367003367,
|
| 3590 |
+
"grad_norm": 0.3359375,
|
| 3591 |
+
"learning_rate": 8.944879177572796e-06,
|
| 3592 |
+
"loss": 0.6081,
|
| 3593 |
+
"step": 496
|
| 3594 |
+
},
|
| 3595 |
+
{
|
| 3596 |
+
"epoch": 1.6734006734006734,
|
| 3597 |
+
"grad_norm": 0.318359375,
|
| 3598 |
+
"learning_rate": 8.908002737771529e-06,
|
| 3599 |
+
"loss": 0.5791,
|
| 3600 |
+
"step": 497
|
| 3601 |
+
},
|
| 3602 |
+
{
|
| 3603 |
+
"epoch": 1.676767676767677,
|
| 3604 |
+
"grad_norm": 0.30078125,
|
| 3605 |
+
"learning_rate": 8.871141320877181e-06,
|
| 3606 |
+
"loss": 0.5912,
|
| 3607 |
+
"step": 498
|
| 3608 |
+
},
|
| 3609 |
+
{
|
| 3610 |
+
"epoch": 1.6801346801346801,
|
| 3611 |
+
"grad_norm": 0.35546875,
|
| 3612 |
+
"learning_rate": 8.83429543400241e-06,
|
| 3613 |
+
"loss": 0.606,
|
| 3614 |
+
"step": 499
|
| 3615 |
+
},
|
| 3616 |
+
{
|
| 3617 |
+
"epoch": 1.6835016835016834,
|
| 3618 |
+
"grad_norm": 0.3046875,
|
| 3619 |
+
"learning_rate": 8.797465584046225e-06,
|
| 3620 |
+
"loss": 0.5852,
|
| 3621 |
+
"step": 500
|
| 3622 |
+
},
|
| 3623 |
+
{
|
| 3624 |
+
"epoch": 1.6868686868686869,
|
| 3625 |
+
"grad_norm": 0.380859375,
|
| 3626 |
+
"learning_rate": 8.760652277687007e-06,
|
| 3627 |
+
"loss": 0.6325,
|
| 3628 |
+
"step": 501
|
| 3629 |
+
},
|
| 3630 |
+
{
|
| 3631 |
+
"epoch": 1.6902356902356903,
|
| 3632 |
+
"grad_norm": 0.337890625,
|
| 3633 |
+
"learning_rate": 8.723856021375543e-06,
|
| 3634 |
+
"loss": 0.6013,
|
| 3635 |
+
"step": 502
|
| 3636 |
+
},
|
| 3637 |
+
{
|
| 3638 |
+
"epoch": 1.6936026936026936,
|
| 3639 |
+
"grad_norm": 0.31640625,
|
| 3640 |
+
"learning_rate": 8.687077321328066e-06,
|
| 3641 |
+
"loss": 0.591,
|
| 3642 |
+
"step": 503
|
| 3643 |
+
},
|
| 3644 |
+
{
|
| 3645 |
+
"epoch": 1.696969696969697,
|
| 3646 |
+
"grad_norm": 0.3359375,
|
| 3647 |
+
"learning_rate": 8.650316683519276e-06,
|
| 3648 |
+
"loss": 0.6132,
|
| 3649 |
+
"step": 504
|
| 3650 |
+
},
|
| 3651 |
+
{
|
| 3652 |
+
"epoch": 1.7003367003367003,
|
| 3653 |
+
"grad_norm": 0.294921875,
|
| 3654 |
+
"learning_rate": 8.61357461367538e-06,
|
| 3655 |
+
"loss": 0.5836,
|
| 3656 |
+
"step": 505
|
| 3657 |
+
},
|
| 3658 |
+
{
|
| 3659 |
+
"epoch": 1.7037037037037037,
|
| 3660 |
+
"grad_norm": 0.302734375,
|
| 3661 |
+
"learning_rate": 8.576851617267151e-06,
|
| 3662 |
+
"loss": 0.6312,
|
| 3663 |
+
"step": 506
|
| 3664 |
+
},
|
| 3665 |
+
{
|
| 3666 |
+
"epoch": 1.7070707070707072,
|
| 3667 |
+
"grad_norm": 0.41796875,
|
| 3668 |
+
"learning_rate": 8.540148199502955e-06,
|
| 3669 |
+
"loss": 0.5807,
|
| 3670 |
+
"step": 507
|
| 3671 |
+
},
|
| 3672 |
+
{
|
| 3673 |
+
"epoch": 1.7104377104377104,
|
| 3674 |
+
"grad_norm": 0.3046875,
|
| 3675 |
+
"learning_rate": 8.503464865321817e-06,
|
| 3676 |
+
"loss": 0.6041,
|
| 3677 |
+
"step": 508
|
| 3678 |
+
},
|
| 3679 |
+
{
|
| 3680 |
+
"epoch": 1.7138047138047137,
|
| 3681 |
+
"grad_norm": 0.287109375,
|
| 3682 |
+
"learning_rate": 8.466802119386462e-06,
|
| 3683 |
+
"loss": 0.5789,
|
| 3684 |
+
"step": 509
|
| 3685 |
+
},
|
| 3686 |
+
{
|
| 3687 |
+
"epoch": 1.7171717171717171,
|
| 3688 |
+
"grad_norm": 0.287109375,
|
| 3689 |
+
"learning_rate": 8.430160466076378e-06,
|
| 3690 |
+
"loss": 0.5908,
|
| 3691 |
+
"step": 510
|
| 3692 |
+
},
|
| 3693 |
+
{
|
| 3694 |
+
"epoch": 1.7205387205387206,
|
| 3695 |
+
"grad_norm": 0.30859375,
|
| 3696 |
+
"learning_rate": 8.393540409480873e-06,
|
| 3697 |
+
"loss": 0.6105,
|
| 3698 |
+
"step": 511
|
| 3699 |
+
},
|
| 3700 |
+
{
|
| 3701 |
+
"epoch": 1.723905723905724,
|
| 3702 |
+
"grad_norm": 0.310546875,
|
| 3703 |
+
"learning_rate": 8.356942453392147e-06,
|
| 3704 |
+
"loss": 0.5964,
|
| 3705 |
+
"step": 512
|
| 3706 |
+
},
|
| 3707 |
+
{
|
| 3708 |
+
"epoch": 1.7272727272727273,
|
| 3709 |
+
"grad_norm": 0.337890625,
|
| 3710 |
+
"learning_rate": 8.320367101298351e-06,
|
| 3711 |
+
"loss": 0.636,
|
| 3712 |
+
"step": 513
|
| 3713 |
+
},
|
| 3714 |
+
{
|
| 3715 |
+
"epoch": 1.7306397306397305,
|
| 3716 |
+
"grad_norm": 0.31640625,
|
| 3717 |
+
"learning_rate": 8.283814856376681e-06,
|
| 3718 |
+
"loss": 0.6001,
|
| 3719 |
+
"step": 514
|
| 3720 |
+
},
|
| 3721 |
+
{
|
| 3722 |
+
"epoch": 1.734006734006734,
|
| 3723 |
+
"grad_norm": 0.2890625,
|
| 3724 |
+
"learning_rate": 8.247286221486429e-06,
|
| 3725 |
+
"loss": 0.5764,
|
| 3726 |
+
"step": 515
|
| 3727 |
+
},
|
| 3728 |
+
{
|
| 3729 |
+
"epoch": 1.7373737373737375,
|
| 3730 |
+
"grad_norm": 0.31640625,
|
| 3731 |
+
"learning_rate": 8.210781699162075e-06,
|
| 3732 |
+
"loss": 0.5564,
|
| 3733 |
+
"step": 516
|
| 3734 |
+
},
|
| 3735 |
+
{
|
| 3736 |
+
"epoch": 1.7407407407407407,
|
| 3737 |
+
"grad_norm": 0.3046875,
|
| 3738 |
+
"learning_rate": 8.174301791606384e-06,
|
| 3739 |
+
"loss": 0.5797,
|
| 3740 |
+
"step": 517
|
| 3741 |
+
},
|
| 3742 |
+
{
|
| 3743 |
+
"epoch": 1.7441077441077442,
|
| 3744 |
+
"grad_norm": 0.318359375,
|
| 3745 |
+
"learning_rate": 8.137847000683485e-06,
|
| 3746 |
+
"loss": 0.5987,
|
| 3747 |
+
"step": 518
|
| 3748 |
+
},
|
| 3749 |
+
{
|
| 3750 |
+
"epoch": 1.7474747474747474,
|
| 3751 |
+
"grad_norm": 0.326171875,
|
| 3752 |
+
"learning_rate": 8.101417827911975e-06,
|
| 3753 |
+
"loss": 0.6197,
|
| 3754 |
+
"step": 519
|
| 3755 |
+
},
|
| 3756 |
+
{
|
| 3757 |
+
"epoch": 1.7508417508417509,
|
| 3758 |
+
"grad_norm": 0.375,
|
| 3759 |
+
"learning_rate": 8.065014774458004e-06,
|
| 3760 |
+
"loss": 0.6284,
|
| 3761 |
+
"step": 520
|
| 3762 |
+
},
|
| 3763 |
+
{
|
| 3764 |
+
"epoch": 1.7542087542087543,
|
| 3765 |
+
"grad_norm": 0.3125,
|
| 3766 |
+
"learning_rate": 8.028638341128405e-06,
|
| 3767 |
+
"loss": 0.5972,
|
| 3768 |
+
"step": 521
|
| 3769 |
+
},
|
| 3770 |
+
{
|
| 3771 |
+
"epoch": 1.7575757575757576,
|
| 3772 |
+
"grad_norm": 0.2890625,
|
| 3773 |
+
"learning_rate": 7.992289028363782e-06,
|
| 3774 |
+
"loss": 0.5666,
|
| 3775 |
+
"step": 522
|
| 3776 |
+
},
|
| 3777 |
+
{
|
| 3778 |
+
"epoch": 1.7609427609427608,
|
| 3779 |
+
"grad_norm": 0.296875,
|
| 3780 |
+
"learning_rate": 7.955967336231635e-06,
|
| 3781 |
+
"loss": 0.5555,
|
| 3782 |
+
"step": 523
|
| 3783 |
+
},
|
| 3784 |
+
{
|
| 3785 |
+
"epoch": 1.7643097643097643,
|
| 3786 |
+
"grad_norm": 0.357421875,
|
| 3787 |
+
"learning_rate": 7.919673764419479e-06,
|
| 3788 |
+
"loss": 0.5559,
|
| 3789 |
+
"step": 524
|
| 3790 |
+
},
|
| 3791 |
+
{
|
| 3792 |
+
"epoch": 1.7676767676767677,
|
| 3793 |
+
"grad_norm": 0.302734375,
|
| 3794 |
+
"learning_rate": 7.883408812227977e-06,
|
| 3795 |
+
"loss": 0.6168,
|
| 3796 |
+
"step": 525
|
| 3797 |
+
}
|
| 3798 |
+
],
|
| 3799 |
+
"logging_steps": 1,
|
| 3800 |
+
"max_steps": 891,
|
| 3801 |
+
"num_input_tokens_seen": 0,
|
| 3802 |
+
"num_train_epochs": 3,
|
| 3803 |
+
"save_steps": 75,
|
| 3804 |
+
"stateful_callbacks": {
|
| 3805 |
+
"TrainerControl": {
|
| 3806 |
+
"args": {
|
| 3807 |
+
"should_epoch_stop": false,
|
| 3808 |
+
"should_evaluate": false,
|
| 3809 |
+
"should_log": false,
|
| 3810 |
+
"should_save": true,
|
| 3811 |
+
"should_training_stop": false
|
| 3812 |
+
},
|
| 3813 |
+
"attributes": {}
|
| 3814 |
+
}
|
| 3815 |
+
},
|
| 3816 |
+
"total_flos": 9.55063707107328e+17,
|
| 3817 |
+
"train_batch_size": 2,
|
| 3818 |
+
"trial_name": null,
|
| 3819 |
+
"trial_params": null
|
| 3820 |
+
}
|
checkpoint-525/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3fe3399e1833e4baeabfb150d83c5b0e5f796c95cfa06424bda5f2c881c0a1e7
|
| 3 |
+
size 6392
|