Casper0508
commited on
Commit
•
7b9a13f
1
Parent(s):
336172d
End of training
Browse files
README.md
CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
-
- Loss: 0.
|
20 |
|
21 |
## Model description
|
22 |
|
@@ -32,20 +32,6 @@ More information needed
|
|
32 |
|
33 |
## Training procedure
|
34 |
|
35 |
-
|
36 |
-
The following `bitsandbytes` quantization config was used during training:
|
37 |
-
- quant_method: bitsandbytes
|
38 |
-
- _load_in_8bit: False
|
39 |
-
- _load_in_4bit: True
|
40 |
-
- llm_int8_threshold: 6.0
|
41 |
-
- llm_int8_skip_modules: None
|
42 |
-
- llm_int8_enable_fp32_cpu_offload: False
|
43 |
-
- llm_int8_has_fp16_weight: False
|
44 |
-
- bnb_4bit_quant_type: nf4
|
45 |
-
- bnb_4bit_use_double_quant: True
|
46 |
-
- bnb_4bit_compute_dtype: bfloat16
|
47 |
-
- load_in_4bit: True
|
48 |
-
- load_in_8bit: False
|
49 |
### Training hyperparameters
|
50 |
|
51 |
The following hyperparameters were used during training:
|
@@ -64,37 +50,37 @@ The following hyperparameters were used during training:
|
|
64 |
|
65 |
| Training Loss | Epoch | Step | Validation Loss |
|
66 |
|:-------------:|:-----:|:----:|:---------------:|
|
67 |
-
| 3.
|
68 |
-
|
|
69 |
-
|
|
70 |
-
| 1.
|
71 |
-
| 1.
|
72 |
-
|
|
73 |
-
| 0.
|
74 |
-
| 0.
|
75 |
-
| 0.
|
76 |
-
| 0.
|
77 |
-
| 0.
|
78 |
-
| 0.
|
79 |
-
| 0.
|
80 |
-
| 0.
|
81 |
-
| 0.
|
82 |
-
| 0.
|
83 |
-
| 0.
|
84 |
-
| 0.
|
85 |
-
| 0.
|
86 |
-
| 0.
|
87 |
-
| 0.
|
88 |
-
| 0.
|
89 |
-
| 0.
|
90 |
-
| 0.
|
91 |
-
| 0.
|
92 |
|
93 |
|
94 |
### Framework versions
|
95 |
|
96 |
- PEFT 0.4.0
|
97 |
- Transformers 4.38.2
|
98 |
-
- Pytorch 2.
|
99 |
- Datasets 2.13.1
|
100 |
- Tokenizers 0.15.2
|
|
|
16 |
|
17 |
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.7124
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
32 |
|
33 |
## Training procedure
|
34 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
### Training hyperparameters
|
36 |
|
37 |
The following hyperparameters were used during training:
|
|
|
50 |
|
51 |
| Training Loss | Epoch | Step | Validation Loss |
|
52 |
|:-------------:|:-----:|:----:|:---------------:|
|
53 |
+
| 3.8363 | 1.36 | 10 | 3.5698 |
|
54 |
+
| 3.2454 | 2.71 | 20 | 2.7356 |
|
55 |
+
| 2.2867 | 4.07 | 30 | 1.7205 |
|
56 |
+
| 1.4623 | 5.42 | 40 | 1.2840 |
|
57 |
+
| 1.1723 | 6.78 | 50 | 1.0982 |
|
58 |
+
| 1.0295 | 8.14 | 60 | 0.9766 |
|
59 |
+
| 0.9085 | 9.49 | 70 | 0.8723 |
|
60 |
+
| 0.784 | 10.85 | 80 | 0.7651 |
|
61 |
+
| 0.717 | 12.2 | 90 | 0.7394 |
|
62 |
+
| 0.6745 | 13.56 | 100 | 0.7235 |
|
63 |
+
| 0.6402 | 14.92 | 110 | 0.7157 |
|
64 |
+
| 0.6251 | 16.27 | 120 | 0.7089 |
|
65 |
+
| 0.5961 | 17.63 | 130 | 0.7100 |
|
66 |
+
| 0.5871 | 18.98 | 140 | 0.7042 |
|
67 |
+
| 0.5714 | 20.34 | 150 | 0.7070 |
|
68 |
+
| 0.5582 | 21.69 | 160 | 0.7062 |
|
69 |
+
| 0.5457 | 23.05 | 170 | 0.7076 |
|
70 |
+
| 0.5392 | 24.41 | 180 | 0.7094 |
|
71 |
+
| 0.5354 | 25.76 | 190 | 0.7100 |
|
72 |
+
| 0.5278 | 27.12 | 200 | 0.7105 |
|
73 |
+
| 0.5275 | 28.47 | 210 | 0.7110 |
|
74 |
+
| 0.5249 | 29.83 | 220 | 0.7123 |
|
75 |
+
| 0.5204 | 31.19 | 230 | 0.7123 |
|
76 |
+
| 0.5198 | 32.54 | 240 | 0.7123 |
|
77 |
+
| 0.5195 | 33.9 | 250 | 0.7124 |
|
78 |
|
79 |
|
80 |
### Framework versions
|
81 |
|
82 |
- PEFT 0.4.0
|
83 |
- Transformers 4.38.2
|
84 |
+
- Pytorch 2.4.0+cu121
|
85 |
- Datasets 2.13.1
|
86 |
- Tokenizers 0.15.2
|
adapter_config.json
CHANGED
@@ -7,11 +7,11 @@
|
|
7 |
"init_lora_weights": true,
|
8 |
"layers_pattern": null,
|
9 |
"layers_to_transform": null,
|
10 |
-
"lora_alpha":
|
11 |
"lora_dropout": 0.3,
|
12 |
"modules_to_save": null,
|
13 |
"peft_type": "LORA",
|
14 |
-
"r":
|
15 |
"revision": null,
|
16 |
"target_modules": [
|
17 |
"q_proj",
|
|
|
7 |
"init_lora_weights": true,
|
8 |
"layers_pattern": null,
|
9 |
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 32,
|
11 |
"lora_dropout": 0.3,
|
12 |
"modules_to_save": null,
|
13 |
"peft_type": "LORA",
|
14 |
+
"r": 16,
|
15 |
"revision": null,
|
16 |
"target_modules": [
|
17 |
"q_proj",
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:853047c7ee6f98c051a455fe53ed043a81d61b9f38d831e7f1d882e9b2d0c0a8
|
3 |
+
size 37774528
|
emissions.csv
CHANGED
@@ -1,2 +1,2 @@
|
|
1 |
timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
|
2 |
-
2024-07-
|
|
|
1 |
timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
|
2 |
+
2024-07-25T00:29:17,b81b783c-301a-439a-a0a5-4917c65bc6de,codecarbon,6290.977914571762,0.3557443410838277,0.5292955629092988,United Kingdom,GBR,scotland,N,,
|
runs/Jul24_22-44-22_msc-modeltrain-pod/events.out.tfevents.1721861066.msc-modeltrain-pod.678.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:64674eeea02a95ded93bfb6003762d75495f8ca12b599682b5fa2540e8d4fb08
|
3 |
+
size 17034
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4984
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:10b3b3a3d7323b4bda4c1a482867d25717c65236d1bd44bb96cd5c9ce33dd107
|
3 |
size 4984
|