Hugging Face's logo

Weni
/
ZeroShot-Multilanguage-3.0-noquantpeft-Llama2-13B

Text Generation
Transformers
PyTorch
TensorBoard
llama
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
Mel-Iza0 commited on
Commit
d21e52a
•
1 Parent(s): e1b5559

Training in progress, epoch 0

Browse files
Files changed (18) hide show
  1. adapter_config.json +21 -0
  2. adapter_model.bin +3 -0
  3. checkpoint-3500/README.md +9 -0
  4. checkpoint-3500/adapter_config.json +21 -0
  5. checkpoint-3500/adapter_model.bin +3 -0
  6. checkpoint-3500/optimizer.pt +3 -0
  7. checkpoint-3500/rng_state.pth +3 -0
  8. checkpoint-3500/scheduler.pt +3 -0
  9. checkpoint-3500/special_tokens_map.json +24 -0
  10. checkpoint-3500/tokenizer.json +0 -0
  11. checkpoint-3500/tokenizer_config.json +32 -0
  12. checkpoint-3500/trainer_state.json +2116 -0
  13. checkpoint-3500/training_args.bin +3 -0
  14. runs/Nov17_16-10-29_18e7f5055510/events.out.tfevents.1700237477.18e7f5055510.7008.0 +3 -0
  15. special_tokens_map.json +24 -0
  16. tokenizer.json +0 -0
  17. tokenizer_config.json +32 -0
  18. training_args.bin +3 -0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Nous-Hermes-Llama2-13b",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 16,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d241ac113bcbc627576a8bc414b60efc26f259dadd016438a5c54cd056036f5a
3
+ size 52486477
checkpoint-3500/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
checkpoint-3500/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Nous-Hermes-Llama2-13b",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 16,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-3500/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d241ac113bcbc627576a8bc414b60efc26f259dadd016438a5c54cd056036f5a
3
+ size 52486477
checkpoint-3500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62ef46aa39b9b52795dbe4eec0f2fea80c78b7f3e9cf3814edae8a0eb3c68211
3
+ size 104992261
checkpoint-3500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b223f3dac083f30dfc67ed1d98ba9461f111c1849743498918d61d0f65e447d
3
+ size 14511
checkpoint-3500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15d108474391e42bff96c66a3c002a5a5a8372e60c534e096e78d0f76a2274b4
3
+ size 627
checkpoint-3500/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-3500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3500/tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "__type": "AddedToken",
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "clean_up_tokenization_spaces": false,
11
+ "eos_token": {
12
+ "__type": "AddedToken",
13
+ "content": "</s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "legacy": false,
20
+ "model_max_length": 1000000000000000019884624838656,
21
+ "pad_token": null,
22
+ "sp_model_kwargs": {},
23
+ "tokenizer_class": "LlamaTokenizer",
24
+ "unk_token": {
25
+ "__type": "AddedToken",
26
+ "content": "<unk>",
27
+ "lstrip": false,
28
+ "normalized": true,
29
+ "rstrip": false,
30
+ "single_word": false
31
+ }
32
+ }
checkpoint-3500/trainer_state.json ADDED
@@ -0,0 +1,2116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.24560541735377706,
5
+ "global_step": 3500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 1.1428571428571429e-05,
13
+ "loss": 1.3156,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 2.2857142857142858e-05,
19
+ "loss": 1.2717,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.0,
24
+ "learning_rate": 3.428571428571429e-05,
25
+ "loss": 1.1819,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.0,
30
+ "learning_rate": 4.5714285714285716e-05,
31
+ "loss": 1.038,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.0,
36
+ "learning_rate": 5.714285714285714e-05,
37
+ "loss": 0.9192,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.0,
42
+ "learning_rate": 6.857142857142858e-05,
43
+ "loss": 0.8204,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.0,
48
+ "learning_rate": 8e-05,
49
+ "loss": 0.7139,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.01,
54
+ "learning_rate": 9.142857142857143e-05,
55
+ "loss": 0.6191,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.01,
60
+ "learning_rate": 0.00010285714285714286,
61
+ "loss": 0.559,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.01,
66
+ "learning_rate": 0.00011428571428571428,
67
+ "loss": 0.5244,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.01,
72
+ "learning_rate": 0.00012571428571428572,
73
+ "loss": 0.5149,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.01,
78
+ "learning_rate": 0.00013714285714285716,
79
+ "loss": 0.4988,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.01,
84
+ "learning_rate": 0.00014857142857142857,
85
+ "loss": 0.475,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.01,
90
+ "learning_rate": 0.00016,
91
+ "loss": 0.4527,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.01,
96
+ "learning_rate": 0.00017142857142857143,
97
+ "loss": 0.4563,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.01,
102
+ "learning_rate": 0.00018285714285714286,
103
+ "loss": 0.4443,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.01,
108
+ "learning_rate": 0.0001942857142857143,
109
+ "loss": 0.4247,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.01,
114
+ "learning_rate": 0.00020571428571428572,
115
+ "loss": 0.4369,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.01,
120
+ "learning_rate": 0.00021714285714285713,
121
+ "loss": 0.4371,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.01,
126
+ "learning_rate": 0.00022857142857142857,
127
+ "loss": 0.42,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.01,
132
+ "learning_rate": 0.00024,
133
+ "loss": 0.4221,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.02,
138
+ "learning_rate": 0.00025142857142857145,
139
+ "loss": 0.4162,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.02,
144
+ "learning_rate": 0.00026285714285714286,
145
+ "loss": 0.4065,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.02,
150
+ "learning_rate": 0.0002742857142857143,
151
+ "loss": 0.4094,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.02,
156
+ "learning_rate": 0.00028571428571428574,
157
+ "loss": 0.4017,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.02,
162
+ "learning_rate": 0.00029714285714285715,
163
+ "loss": 0.4291,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.02,
168
+ "learning_rate": 0.0003085714285714286,
169
+ "loss": 0.4198,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.02,
174
+ "learning_rate": 0.00032,
175
+ "loss": 0.4069,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.02,
180
+ "learning_rate": 0.0003314285714285715,
181
+ "loss": 0.4118,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.02,
186
+ "learning_rate": 0.00034285714285714285,
187
+ "loss": 0.3921,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.02,
192
+ "learning_rate": 0.00035428571428571426,
193
+ "loss": 0.3911,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.02,
198
+ "learning_rate": 0.00036571428571428573,
199
+ "loss": 0.4013,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.02,
204
+ "learning_rate": 0.00037714285714285714,
205
+ "loss": 0.4114,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.02,
210
+ "learning_rate": 0.0003885714285714286,
211
+ "loss": 0.4098,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.02,
216
+ "learning_rate": 0.0004,
217
+ "loss": 0.4114,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.03,
222
+ "learning_rate": 0.00039999005339118863,
223
+ "loss": 0.3901,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.03,
228
+ "learning_rate": 0.0003999602145541047,
229
+ "loss": 0.405,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.03,
234
+ "learning_rate": 0.00039991048645670067,
235
+ "loss": 0.3946,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.03,
240
+ "learning_rate": 0.00039984087404523587,
241
+ "loss": 0.398,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.03,
246
+ "learning_rate": 0.00039975138424378447,
247
+ "loss": 0.3979,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.03,
252
+ "learning_rate": 0.000399642025953547,
253
+ "loss": 0.3996,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.03,
258
+ "learning_rate": 0.0003995128100519649,
259
+ "loss": 0.3935,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.03,
264
+ "learning_rate": 0.0003993637493916382,
265
+ "loss": 0.4008,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.03,
270
+ "learning_rate": 0.00039919485879904784,
271
+ "loss": 0.3875,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.03,
276
+ "learning_rate": 0.00039900615507308033,
277
+ "loss": 0.3918,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.03,
282
+ "learning_rate": 0.00039879765698335704,
283
+ "loss": 0.3907,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.03,
288
+ "learning_rate": 0.0003985693852683675,
289
+ "loss": 0.3899,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.03,
294
+ "learning_rate": 0.00039832136263340603,
295
+ "loss": 0.3916,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.03,
300
+ "learning_rate": 0.0003980536137483141,
301
+ "loss": 0.3971,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.04,
306
+ "learning_rate": 0.0003977661652450257,
307
+ "loss": 0.3915,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.04,
312
+ "learning_rate": 0.00039745904571491917,
313
+ "loss": 0.3838,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.04,
318
+ "learning_rate": 0.0003971322857059726,
319
+ "loss": 0.378,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.04,
324
+ "learning_rate": 0.0003967859177197259,
325
+ "loss": 0.3795,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.04,
330
+ "learning_rate": 0.0003964199762080478,
331
+ "loss": 0.3697,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.04,
336
+ "learning_rate": 0.00039603449756970877,
337
+ "loss": 0.3785,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.04,
342
+ "learning_rate": 0.00039562952014676116,
343
+ "loss": 0.3868,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.04,
348
+ "learning_rate": 0.0003952050842207249,
349
+ "loss": 0.3938,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.04,
354
+ "learning_rate": 0.0003947612320085813,
355
+ "loss": 0.385,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 0.04,
360
+ "learning_rate": 0.0003942980076585735,
361
+ "loss": 0.3884,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 0.04,
366
+ "learning_rate": 0.0003938154572458156,
367
+ "loss": 0.3835,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 0.04,
372
+ "learning_rate": 0.00039331362876770945,
373
+ "loss": 0.3884,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 0.04,
378
+ "learning_rate": 0.00039279257213917066,
379
+ "loss": 0.3873,
380
+ "step": 620
381
+ },
382
+ {
383
+ "epoch": 0.04,
384
+ "learning_rate": 0.0003922523391876638,
385
+ "loss": 0.3824,
386
+ "step": 630
387
+ },
388
+ {
389
+ "epoch": 0.04,
390
+ "learning_rate": 0.00039169298364804715,
391
+ "loss": 0.3853,
392
+ "step": 640
393
+ },
394
+ {
395
+ "epoch": 0.05,
396
+ "learning_rate": 0.0003911145611572282,
397
+ "loss": 0.3783,
398
+ "step": 650
399
+ },
400
+ {
401
+ "epoch": 0.05,
402
+ "learning_rate": 0.0003905171292486292,
403
+ "loss": 0.3806,
404
+ "step": 660
405
+ },
406
+ {
407
+ "epoch": 0.05,
408
+ "learning_rate": 0.00038990074734646525,
409
+ "loss": 0.3794,
410
+ "step": 670
411
+ },
412
+ {
413
+ "epoch": 0.05,
414
+ "learning_rate": 0.00038926547675983286,
415
+ "loss": 0.3883,
416
+ "step": 680
417
+ },
418
+ {
419
+ "epoch": 0.05,
420
+ "learning_rate": 0.0003886113806766121,
421
+ "loss": 0.3764,
422
+ "step": 690
423
+ },
424
+ {
425
+ "epoch": 0.05,
426
+ "learning_rate": 0.0003879385241571817,
427
+ "loss": 0.3872,
428
+ "step": 700
429
+ },
430
+ {
431
+ "epoch": 0.05,
432
+ "learning_rate": 0.00038724697412794747,
433
+ "loss": 0.381,
434
+ "step": 710
435
+ },
436
+ {
437
+ "epoch": 0.05,
438
+ "learning_rate": 0.00038653679937468557,
439
+ "loss": 0.3777,
440
+ "step": 720
441
+ },
442
+ {
443
+ "epoch": 0.05,
444
+ "learning_rate": 0.0003858080705357005,
445
+ "loss": 0.378,
446
+ "step": 730
447
+ },
448
+ {
449
+ "epoch": 0.05,
450
+ "learning_rate": 0.00038506086009479937,
451
+ "loss": 0.3838,
452
+ "step": 740
453
+ },
454
+ {
455
+ "epoch": 0.05,
456
+ "learning_rate": 0.0003842952423740815,
457
+ "loss": 0.3748,
458
+ "step": 750
459
+ },
460
+ {
461
+ "epoch": 0.05,
462
+ "learning_rate": 0.00038351129352654677,
463
+ "loss": 0.3732,
464
+ "step": 760
465
+ },
466
+ {
467
+ "epoch": 0.05,
468
+ "learning_rate": 0.0003827090915285202,
469
+ "loss": 0.3677,
470
+ "step": 770
471
+ },
472
+ {
473
+ "epoch": 0.05,
474
+ "learning_rate": 0.00038188871617189646,
475
+ "loss": 0.3635,
476
+ "step": 780
477
+ },
478
+ {
479
+ "epoch": 0.06,
480
+ "learning_rate": 0.000381050249056203,
481
+ "loss": 0.369,
482
+ "step": 790
483
+ },
484
+ {
485
+ "epoch": 0.06,
486
+ "learning_rate": 0.0003801937735804838,
487
+ "loss": 0.3823,
488
+ "step": 800
489
+ },
490
+ {
491
+ "epoch": 0.06,
492
+ "learning_rate": 0.0003793193749350042,
493
+ "loss": 0.385,
494
+ "step": 810
495
+ },
496
+ {
497
+ "epoch": 0.06,
498
+ "learning_rate": 0.0003784271400927767,
499
+ "loss": 0.3785,
500
+ "step": 820
501
+ },
502
+ {
503
+ "epoch": 0.06,
504
+ "learning_rate": 0.00037751715780091086,
505
+ "loss": 0.3779,
506
+ "step": 830
507
+ },
508
+ {
509
+ "epoch": 0.06,
510
+ "learning_rate": 0.00037658951857178543,
511
+ "loss": 0.3604,
512
+ "step": 840
513
+ },
514
+ {
515
+ "epoch": 0.06,
516
+ "learning_rate": 0.00037564431467404573,
517
+ "loss": 0.3742,
518
+ "step": 850
519
+ },
520
+ {
521
+ "epoch": 0.06,
522
+ "learning_rate": 0.00037468164012342597,
523
+ "loss": 0.365,
524
+ "step": 860
525
+ },
526
+ {
527
+ "epoch": 0.06,
528
+ "learning_rate": 0.0003737015906733978,
529
+ "loss": 0.382,
530
+ "step": 870
531
+ },
532
+ {
533
+ "epoch": 0.06,
534
+ "learning_rate": 0.0003727042638056462,
535
+ "loss": 0.3804,
536
+ "step": 880
537
+ },
538
+ {
539
+ "epoch": 0.06,
540
+ "learning_rate": 0.00037168975872037323,
541
+ "loss": 0.3794,
542
+ "step": 890
543
+ },
544
+ {
545
+ "epoch": 0.06,
546
+ "learning_rate": 0.00037065817632643116,
547
+ "loss": 0.3794,
548
+ "step": 900
549
+ },
550
+ {
551
+ "epoch": 0.06,
552
+ "learning_rate": 0.0003696096192312852,
553
+ "loss": 0.3725,
554
+ "step": 910
555
+ },
556
+ {
557
+ "epoch": 0.06,
558
+ "learning_rate": 0.0003685441917308078,
559
+ "loss": 0.3698,
560
+ "step": 920
561
+ },
562
+ {
563
+ "epoch": 0.07,
564
+ "learning_rate": 0.0003674619997989047,
565
+ "loss": 0.3679,
566
+ "step": 930
567
+ },
568
+ {
569
+ "epoch": 0.07,
570
+ "learning_rate": 0.0003663631510769738,
571
+ "loss": 0.3671,
572
+ "step": 940
573
+ },
574
+ {
575
+ "epoch": 0.07,
576
+ "learning_rate": 0.000365247754863199,
577
+ "loss": 0.3719,
578
+ "step": 950
579
+ },
580
+ {
581
+ "epoch": 0.07,
582
+ "learning_rate": 0.0003641159221016783,
583
+ "loss": 0.3685,
584
+ "step": 960
585
+ },
586
+ {
587
+ "epoch": 0.07,
588
+ "learning_rate": 0.00036296776537138905,
589
+ "loss": 0.3703,
590
+ "step": 970
591
+ },
592
+ {
593
+ "epoch": 0.07,
594
+ "learning_rate": 0.0003618033988749895,
595
+ "loss": 0.3708,
596
+ "step": 980
597
+ },
598
+ {
599
+ "epoch": 0.07,
600
+ "learning_rate": 0.00036062293842746036,
601
+ "loss": 0.3642,
602
+ "step": 990
603
+ },
604
+ {
605
+ "epoch": 0.07,
606
+ "learning_rate": 0.0003594265014445845,
607
+ "loss": 0.3692,
608
+ "step": 1000
609
+ },
610
+ {
611
+ "epoch": 0.07,
612
+ "learning_rate": 0.0003582142069312683,
613
+ "loss": 0.3753,
614
+ "step": 1010
615
+ },
616
+ {
617
+ "epoch": 0.07,
618
+ "learning_rate": 0.0003569861754697045,
619
+ "loss": 0.3673,
620
+ "step": 1020
621
+ },
622
+ {
623
+ "epoch": 0.07,
624
+ "learning_rate": 0.0003557425292073788,
625
+ "loss": 0.3728,
626
+ "step": 1030
627
+ },
628
+ {
629
+ "epoch": 0.07,
630
+ "learning_rate": 0.0003544833918449199,
631
+ "loss": 0.3711,
632
+ "step": 1040
633
+ },
634
+ {
635
+ "epoch": 0.07,
636
+ "learning_rate": 0.0003532088886237956,
637
+ "loss": 0.3657,
638
+ "step": 1050
639
+ },
640
+ {
641
+ "epoch": 0.07,
642
+ "learning_rate": 0.00035191914631385563,
643
+ "loss": 0.3714,
644
+ "step": 1060
645
+ },
646
+ {
647
+ "epoch": 0.08,
648
+ "learning_rate": 0.00035061429320072223,
649
+ "loss": 0.3705,
650
+ "step": 1070
651
+ },
652
+ {
653
+ "epoch": 0.08,
654
+ "learning_rate": 0.00034929445907303007,
655
+ "loss": 0.3672,
656
+ "step": 1080
657
+ },
658
+ {
659
+ "epoch": 0.08,
660
+ "learning_rate": 0.0003479597752095168,
661
+ "loss": 0.3628,
662
+ "step": 1090
663
+ },
664
+ {
665
+ "epoch": 0.08,
666
+ "learning_rate": 0.0003466103743659653,
667
+ "loss": 0.3647,
668
+ "step": 1100
669
+ },
670
+ {
671
+ "epoch": 0.08,
672
+ "learning_rate": 0.0003452463907619986,
673
+ "loss": 0.3699,
674
+ "step": 1110
675
+ },
676
+ {
677
+ "epoch": 0.08,
678
+ "learning_rate": 0.0003438679600677302,
679
+ "loss": 0.3699,
680
+ "step": 1120
681
+ },
682
+ {
683
+ "epoch": 0.08,
684
+ "learning_rate": 0.000342475219390269,
685
+ "loss": 0.3659,
686
+ "step": 1130
687
+ },
688
+ {
689
+ "epoch": 0.08,
690
+ "learning_rate": 0.0003410683072600818,
691
+ "loss": 0.3708,
692
+ "step": 1140
693
+ },
694
+ {
695
+ "epoch": 0.08,
696
+ "learning_rate": 0.00033964736361721457,
697
+ "loss": 0.3667,
698
+ "step": 1150
699
+ },
700
+ {
701
+ "epoch": 0.08,
702
+ "learning_rate": 0.00033821252979737297,
703
+ "loss": 0.3628,
704
+ "step": 1160
705
+ },
706
+ {
707
+ "epoch": 0.08,
708
+ "learning_rate": 0.00033676394851786404,
709
+ "loss": 0.3618,
710
+ "step": 1170
711
+ },
712
+ {
713
+ "epoch": 0.08,
714
+ "learning_rate": 0.00033530176386340106,
715
+ "loss": 0.372,
716
+ "step": 1180
717
+ },
718
+ {
719
+ "epoch": 0.08,
720
+ "learning_rate": 0.00033382612127177166,
721
+ "loss": 0.3637,
722
+ "step": 1190
723
+ },
724
+ {
725
+ "epoch": 0.08,
726
+ "learning_rate": 0.0003323371675193719,
727
+ "loss": 0.3623,
728
+ "step": 1200
729
+ },
730
+ {
731
+ "epoch": 0.08,
732
+ "learning_rate": 0.0003308350507066069,
733
+ "loss": 0.3684,
734
+ "step": 1210
735
+ },
736
+ {
737
+ "epoch": 0.09,
738
+ "learning_rate": 0.0003293199202431599,
739
+ "loss": 0.3648,
740
+ "step": 1220
741
+ },
742
+ {
743
+ "epoch": 0.09,
744
+ "learning_rate": 0.0003277919268331312,
745
+ "loss": 0.3555,
746
+ "step": 1230
747
+ },
748
+ {
749
+ "epoch": 0.09,
750
+ "learning_rate": 0.0003262512224600478,
751
+ "loss": 0.3592,
752
+ "step": 1240
753
+ },
754
+ {
755
+ "epoch": 0.09,
756
+ "learning_rate": 0.00032469796037174674,
757
+ "loss": 0.3587,
758
+ "step": 1250
759
+ },
760
+ {
761
+ "epoch": 0.09,
762
+ "learning_rate": 0.0003231322950651316,
763
+ "loss": 0.3645,
764
+ "step": 1260
765
+ },
766
+ {
767
+ "epoch": 0.09,
768
+ "learning_rate": 0.00032155438227080604,
769
+ "loss": 0.3647,
770
+ "step": 1270
771
+ },
772
+ {
773
+ "epoch": 0.09,
774
+ "learning_rate": 0.00031996437893758276,
775
+ "loss": 0.3563,
776
+ "step": 1280
777
+ },
778
+ {
779
+ "epoch": 0.09,
780
+ "learning_rate": 0.00031836244321687356,
781
+ "loss": 0.3613,
782
+ "step": 1290
783
+ },
784
+ {
785
+ "epoch": 0.09,
786
+ "learning_rate": 0.000316748734446958,
787
+ "loss": 0.3726,
788
+ "step": 1300
789
+ },
790
+ {
791
+ "epoch": 0.09,
792
+ "learning_rate": 0.00031512341313713474,
793
+ "loss": 0.377,
794
+ "step": 1310
795
+ },
796
+ {
797
+ "epoch": 0.09,
798
+ "learning_rate": 0.0003134866409517564,
799
+ "loss": 0.3629,
800
+ "step": 1320
801
+ },
802
+ {
803
+ "epoch": 0.09,
804
+ "learning_rate": 0.0003118385806941494,
805
+ "loss": 0.363,
806
+ "step": 1330
807
+ },
808
+ {
809
+ "epoch": 0.09,
810
+ "learning_rate": 0.0003101793962904205,
811
+ "loss": 0.353,
812
+ "step": 1340
813
+ },
814
+ {
815
+ "epoch": 0.09,
816
+ "learning_rate": 0.0003085092527731519,
817
+ "loss": 0.3537,
818
+ "step": 1350
819
+ },
820
+ {
821
+ "epoch": 0.1,
822
+ "learning_rate": 0.00030682831626498575,
823
+ "loss": 0.3606,
824
+ "step": 1360
825
+ },
826
+ {
827
+ "epoch": 0.1,
828
+ "learning_rate": 0.00030513675396210094,
829
+ "loss": 0.357,
830
+ "step": 1370
831
+ },
832
+ {
833
+ "epoch": 0.1,
834
+ "learning_rate": 0.00030343473411758253,
835
+ "loss": 0.3603,
836
+ "step": 1380
837
+ },
838
+ {
839
+ "epoch": 0.1,
840
+ "learning_rate": 0.0003017224260246863,
841
+ "loss": 0.3618,
842
+ "step": 1390
843
+ },
844
+ {
845
+ "epoch": 0.1,
846
+ "learning_rate": 0.00030000000000000003,
847
+ "loss": 0.3674,
848
+ "step": 1400
849
+ },
850
+ {
851
+ "epoch": 0.1,
852
+ "learning_rate": 0.0002982676273665023,
853
+ "loss": 0.3711,
854
+ "step": 1410
855
+ },
856
+ {
857
+ "epoch": 0.1,
858
+ "learning_rate": 0.0002965254804365222,
859
+ "loss": 0.366,
860
+ "step": 1420
861
+ },
862
+ {
863
+ "epoch": 0.1,
864
+ "learning_rate": 0.0002947737324945997,
865
+ "loss": 0.3727,
866
+ "step": 1430
867
+ },
868
+ {
869
+ "epoch": 0.1,
870
+ "learning_rate": 0.00029318909447116496,
871
+ "loss": 0.372,
872
+ "step": 1440
873
+ },
874
+ {
875
+ "epoch": 0.1,
876
+ "learning_rate": 0.0002914195854117389,
877
+ "loss": 0.3683,
878
+ "step": 1450
879
+ },
880
+ {
881
+ "epoch": 0.1,
882
+ "learning_rate": 0.0002896409832037748,
883
+ "loss": 0.356,
884
+ "step": 1460
885
+ },
886
+ {
887
+ "epoch": 0.1,
888
+ "learning_rate": 0.0002878534647578768,
889
+ "loss": 0.3599,
890
+ "step": 1470
891
+ },
892
+ {
893
+ "epoch": 0.1,
894
+ "learning_rate": 0.0002860572078715121,
895
+ "loss": 0.3605,
896
+ "step": 1480
897
+ },
898
+ {
899
+ "epoch": 0.1,
900
+ "learning_rate": 0.0002842523912113264,
901
+ "loss": 0.3595,
902
+ "step": 1490
903
+ },
904
+ {
905
+ "epoch": 0.11,
906
+ "learning_rate": 0.00028243919429537265,
907
+ "loss": 0.3592,
908
+ "step": 1500
909
+ },
910
+ {
911
+ "epoch": 0.11,
912
+ "learning_rate": 0.00028061779747525504,
913
+ "loss": 0.3582,
914
+ "step": 1510
915
+ },
916
+ {
917
+ "epoch": 0.11,
918
+ "learning_rate": 0.00027878838191819024,
919
+ "loss": 0.3473,
920
+ "step": 1520
921
+ },
922
+ {
923
+ "epoch": 0.11,
924
+ "learning_rate": 0.0002769511295889872,
925
+ "loss": 0.3577,
926
+ "step": 1530
927
+ },
928
+ {
929
+ "epoch": 0.11,
930
+ "learning_rate": 0.000275106223231948,
931
+ "loss": 0.3465,
932
+ "step": 1540
933
+ },
934
+ {
935
+ "epoch": 0.11,
936
+ "learning_rate": 0.00027325384635269094,
937
+ "loss": 0.3547,
938
+ "step": 1550
939
+ },
940
+ {
941
+ "epoch": 0.11,
942
+ "learning_rate": 0.00027139418319989785,
943
+ "loss": 0.3583,
944
+ "step": 1560
945
+ },
946
+ {
947
+ "epoch": 0.11,
948
+ "learning_rate": 0.0002695274187469878,
949
+ "loss": 0.3558,
950
+ "step": 1570
951
+ },
952
+ {
953
+ "epoch": 0.11,
954
+ "learning_rate": 0.0002676537386737183,
955
+ "loss": 0.3547,
956
+ "step": 1580
957
+ },
958
+ {
959
+ "epoch": 0.11,
960
+ "learning_rate": 0.00026577332934771665,
961
+ "loss": 0.3591,
962
+ "step": 1590
963
+ },
964
+ {
965
+ "epoch": 0.11,
966
+ "learning_rate": 0.0002638863778059426,
967
+ "loss": 0.3443,
968
+ "step": 1600
969
+ },
970
+ {
971
+ "epoch": 0.11,
972
+ "learning_rate": 0.00026199307173608433,
973
+ "loss": 0.3481,
974
+ "step": 1610
975
+ },
976
+ {
977
+ "epoch": 0.11,
978
+ "learning_rate": 0.0002600935994578904,
979
+ "loss": 0.349,
980
+ "step": 1620
981
+ },
982
+ {
983
+ "epoch": 0.11,
984
+ "learning_rate": 0.0002581881499044377,
985
+ "loss": 0.3565,
986
+ "step": 1630
987
+ },
988
+ {
989
+ "epoch": 0.12,
990
+ "learning_rate": 0.0002562769126033394,
991
+ "loss": 0.3539,
992
+ "step": 1640
993
+ },
994
+ {
995
+ "epoch": 0.12,
996
+ "learning_rate": 0.00025436007765789327,
997
+ "loss": 0.353,
998
+ "step": 1650
999
+ },
1000
+ {
1001
+ "epoch": 0.12,
1002
+ "learning_rate": 0.00025243783572817297,
1003
+ "loss": 0.3612,
1004
+ "step": 1660
1005
+ },
1006
+ {
1007
+ "epoch": 0.12,
1008
+ "learning_rate": 0.0002505103780120636,
1009
+ "loss": 0.3573,
1010
+ "step": 1670
1011
+ },
1012
+ {
1013
+ "epoch": 0.12,
1014
+ "learning_rate": 0.0002485778962262443,
1015
+ "loss": 0.3536,
1016
+ "step": 1680
1017
+ },
1018
+ {
1019
+ "epoch": 0.12,
1020
+ "learning_rate": 0.00024664058258711853,
1021
+ "loss": 0.3632,
1022
+ "step": 1690
1023
+ },
1024
+ {
1025
+ "epoch": 0.12,
1026
+ "learning_rate": 0.0002446986297916954,
1027
+ "loss": 0.3522,
1028
+ "step": 1700
1029
+ },
1030
+ {
1031
+ "epoch": 0.12,
1032
+ "learning_rate": 0.00024275223099842291,
1033
+ "loss": 0.3509,
1034
+ "step": 1710
1035
+ },
1036
+ {
1037
+ "epoch": 0.12,
1038
+ "learning_rate": 0.00024080157980797484,
1039
+ "loss": 0.348,
1040
+ "step": 1720
1041
+ },
1042
+ {
1043
+ "epoch": 0.12,
1044
+ "learning_rate": 0.0002388468702439944,
1045
+ "loss": 0.3556,
1046
+ "step": 1730
1047
+ },
1048
+ {
1049
+ "epoch": 0.12,
1050
+ "learning_rate": 0.00023688829673379534,
1051
+ "loss": 0.3561,
1052
+ "step": 1740
1053
+ },
1054
+ {
1055
+ "epoch": 0.12,
1056
+ "learning_rate": 0.00023492605408902297,
1057
+ "loss": 0.3532,
1058
+ "step": 1750
1059
+ },
1060
+ {
1061
+ "epoch": 0.12,
1062
+ "learning_rate": 0.00023296033748627712,
1063
+ "loss": 0.346,
1064
+ "step": 1760
1065
+ },
1066
+ {
1067
+ "epoch": 0.12,
1068
+ "learning_rate": 0.0002309913424476986,
1069
+ "loss": 0.3521,
1070
+ "step": 1770
1071
+ },
1072
+ {
1073
+ "epoch": 0.12,
1074
+ "learning_rate": 0.00022901926482152138,
1075
+ "loss": 0.3505,
1076
+ "step": 1780
1077
+ },
1078
+ {
1079
+ "epoch": 0.13,
1080
+ "learning_rate": 0.00022704430076259246,
1081
+ "loss": 0.3444,
1082
+ "step": 1790
1083
+ },
1084
+ {
1085
+ "epoch": 0.13,
1086
+ "learning_rate": 0.00022506664671286087,
1087
+ "loss": 0.3467,
1088
+ "step": 1800
1089
+ },
1090
+ {
1091
+ "epoch": 0.13,
1092
+ "learning_rate": 0.00022308649938183864,
1093
+ "loss": 0.3561,
1094
+ "step": 1810
1095
+ },
1096
+ {
1097
+ "epoch": 0.13,
1098
+ "learning_rate": 0.00022110405572703466,
1099
+ "loss": 0.3529,
1100
+ "step": 1820
1101
+ },
1102
+ {
1103
+ "epoch": 0.13,
1104
+ "learning_rate": 0.00021911951293436416,
1105
+ "loss": 0.348,
1106
+ "step": 1830
1107
+ },
1108
+ {
1109
+ "epoch": 0.13,
1110
+ "learning_rate": 0.00021713306839853545,
1111
+ "loss": 0.3449,
1112
+ "step": 1840
1113
+ },
1114
+ {
1115
+ "epoch": 0.13,
1116
+ "learning_rate": 0.0002151449197034157,
1117
+ "loss": 0.3525,
1118
+ "step": 1850
1119
+ },
1120
+ {
1121
+ "epoch": 0.13,
1122
+ "learning_rate": 0.0002131552646023783,
1123
+ "loss": 0.3431,
1124
+ "step": 1860
1125
+ },
1126
+ {
1127
+ "epoch": 0.13,
1128
+ "learning_rate": 0.00021116430099863277,
1129
+ "loss": 0.35,
1130
+ "step": 1870
1131
+ },
1132
+ {
1133
+ "epoch": 0.13,
1134
+ "learning_rate": 0.0002091722269255404,
1135
+ "loss": 0.3397,
1136
+ "step": 1880
1137
+ },
1138
+ {
1139
+ "epoch": 0.13,
1140
+ "learning_rate": 0.0002071792405269165,
1141
+ "loss": 0.3532,
1142
+ "step": 1890
1143
+ },
1144
+ {
1145
+ "epoch": 0.13,
1146
+ "learning_rate": 0.00020518554003732167,
1147
+ "loss": 0.3487,
1148
+ "step": 1900
1149
+ },
1150
+ {
1151
+ "epoch": 0.13,
1152
+ "learning_rate": 0.00020319132376234462,
1153
+ "loss": 0.3508,
1154
+ "step": 1910
1155
+ },
1156
+ {
1157
+ "epoch": 0.13,
1158
+ "learning_rate": 0.00020119679005887702,
1159
+ "loss": 0.35,
1160
+ "step": 1920
1161
+ },
1162
+ {
1163
+ "epoch": 0.14,
1164
+ "learning_rate": 0.00019920213731538394,
1165
+ "loss": 0.3479,
1166
+ "step": 1930
1167
+ },
1168
+ {
1169
+ "epoch": 0.14,
1170
+ "learning_rate": 0.00019720756393217098,
1171
+ "loss": 0.3439,
1172
+ "step": 1940
1173
+ },
1174
+ {
1175
+ "epoch": 0.14,
1176
+ "learning_rate": 0.00019521326830164998,
1177
+ "loss": 0.337,
1178
+ "step": 1950
1179
+ },
1180
+ {
1181
+ "epoch": 0.14,
1182
+ "learning_rate": 0.00019321944878860587,
1183
+ "loss": 0.3419,
1184
+ "step": 1960
1185
+ },
1186
+ {
1187
+ "epoch": 0.14,
1188
+ "learning_rate": 0.000191226303710466,
1189
+ "loss": 0.3508,
1190
+ "step": 1970
1191
+ },
1192
+ {
1193
+ "epoch": 0.14,
1194
+ "learning_rate": 0.00018923403131757439,
1195
+ "loss": 0.3474,
1196
+ "step": 1980
1197
+ },
1198
+ {
1199
+ "epoch": 0.14,
1200
+ "learning_rate": 0.00018724282977347235,
1201
+ "loss": 0.3451,
1202
+ "step": 1990
1203
+ },
1204
+ {
1205
+ "epoch": 0.14,
1206
+ "learning_rate": 0.00018525289713518817,
1207
+ "loss": 0.3449,
1208
+ "step": 2000
1209
+ },
1210
+ {
1211
+ "epoch": 0.14,
1212
+ "learning_rate": 0.00018326443133353693,
1213
+ "loss": 0.3437,
1214
+ "step": 2010
1215
+ },
1216
+ {
1217
+ "epoch": 0.14,
1218
+ "learning_rate": 0.00018127763015343332,
1219
+ "loss": 0.3425,
1220
+ "step": 2020
1221
+ },
1222
+ {
1223
+ "epoch": 0.14,
1224
+ "learning_rate": 0.00017929269121421857,
1225
+ "loss": 0.3478,
1226
+ "step": 2030
1227
+ },
1228
+ {
1229
+ "epoch": 0.14,
1230
+ "learning_rate": 0.00017730981195000406,
1231
+ "loss": 0.3408,
1232
+ "step": 2040
1233
+ },
1234
+ {
1235
+ "epoch": 0.14,
1236
+ "learning_rate": 0.00017532918959003353,
1237
+ "loss": 0.3452,
1238
+ "step": 2050
1239
+ },
1240
+ {
1241
+ "epoch": 0.14,
1242
+ "learning_rate": 0.00017335102113906505,
1243
+ "loss": 0.3395,
1244
+ "step": 2060
1245
+ },
1246
+ {
1247
+ "epoch": 0.15,
1248
+ "learning_rate": 0.00017137550335777612,
1249
+ "loss": 0.3462,
1250
+ "step": 2070
1251
+ },
1252
+ {
1253
+ "epoch": 0.15,
1254
+ "learning_rate": 0.0001694028327431924,
1255
+ "loss": 0.3397,
1256
+ "step": 2080
1257
+ },
1258
+ {
1259
+ "epoch": 0.15,
1260
+ "learning_rate": 0.0001674332055091431,
1261
+ "loss": 0.3434,
1262
+ "step": 2090
1263
+ },
1264
+ {
1265
+ "epoch": 0.15,
1266
+ "learning_rate": 0.0001654668175667442,
1267
+ "loss": 0.3431,
1268
+ "step": 2100
1269
+ },
1270
+ {
1271
+ "epoch": 0.15,
1272
+ "learning_rate": 0.00016350386450491208,
1273
+ "loss": 0.3326,
1274
+ "step": 2110
1275
+ },
1276
+ {
1277
+ "epoch": 0.15,
1278
+ "learning_rate": 0.00016154454157090884,
1279
+ "loss": 0.3392,
1280
+ "step": 2120
1281
+ },
1282
+ {
1283
+ "epoch": 0.15,
1284
+ "learning_rate": 0.00015958904365092225,
1285
+ "loss": 0.339,
1286
+ "step": 2130
1287
+ },
1288
+ {
1289
+ "epoch": 0.15,
1290
+ "learning_rate": 0.00015763756525068065,
1291
+ "loss": 0.336,
1292
+ "step": 2140
1293
+ },
1294
+ {
1295
+ "epoch": 0.15,
1296
+ "learning_rate": 0.00015569030047610656,
1297
+ "loss": 0.3363,
1298
+ "step": 2150
1299
+ },
1300
+ {
1301
+ "epoch": 0.15,
1302
+ "learning_rate": 0.0001537474430140096,
1303
+ "loss": 0.3373,
1304
+ "step": 2160
1305
+ },
1306
+ {
1307
+ "epoch": 0.15,
1308
+ "learning_rate": 0.0001518091861128213,
1309
+ "loss": 0.3419,
1310
+ "step": 2170
1311
+ },
1312
+ {
1313
+ "epoch": 0.15,
1314
+ "learning_rate": 0.00014987572256337336,
1315
+ "loss": 0.3493,
1316
+ "step": 2180
1317
+ },
1318
+ {
1319
+ "epoch": 0.15,
1320
+ "learning_rate": 0.0001479472446797216,
1321
+ "loss": 0.3334,
1322
+ "step": 2190
1323
+ },
1324
+ {
1325
+ "epoch": 0.15,
1326
+ "learning_rate": 0.00014602394428001712,
1327
+ "loss": 0.3384,
1328
+ "step": 2200
1329
+ },
1330
+ {
1331
+ "epoch": 0.16,
1332
+ "learning_rate": 0.00014410601266742691,
1333
+ "loss": 0.3435,
1334
+ "step": 2210
1335
+ },
1336
+ {
1337
+ "epoch": 0.16,
1338
+ "learning_rate": 0.00014219364061110565,
1339
+ "loss": 0.341,
1340
+ "step": 2220
1341
+ },
1342
+ {
1343
+ "epoch": 0.16,
1344
+ "learning_rate": 0.00014028701832722104,
1345
+ "loss": 0.3317,
1346
+ "step": 2230
1347
+ },
1348
+ {
1349
+ "epoch": 0.16,
1350
+ "learning_rate": 0.00013838633546003302,
1351
+ "loss": 0.3422,
1352
+ "step": 2240
1353
+ },
1354
+ {
1355
+ "epoch": 0.16,
1356
+ "learning_rate": 0.00013649178106303115,
1357
+ "loss": 0.3457,
1358
+ "step": 2250
1359
+ },
1360
+ {
1361
+ "epoch": 0.16,
1362
+ "learning_rate": 0.00013460354358013,
1363
+ "loss": 0.3356,
1364
+ "step": 2260
1365
+ },
1366
+ {
1367
+ "epoch": 0.16,
1368
+ "learning_rate": 0.0001327218108269255,
1369
+ "loss": 0.3377,
1370
+ "step": 2270
1371
+ },
1372
+ {
1373
+ "epoch": 0.16,
1374
+ "learning_rate": 0.00013084676997201342,
1375
+ "loss": 0.3304,
1376
+ "step": 2280
1377
+ },
1378
+ {
1379
+ "epoch": 0.16,
1380
+ "learning_rate": 0.00012897860751837263,
1381
+ "loss": 0.3413,
1382
+ "step": 2290
1383
+ },
1384
+ {
1385
+ "epoch": 0.16,
1386
+ "learning_rate": 0.00012711750928481443,
1387
+ "loss": 0.3433,
1388
+ "step": 2300
1389
+ },
1390
+ {
1391
+ "epoch": 0.16,
1392
+ "learning_rate": 0.00012526366038749956,
1393
+ "loss": 0.3419,
1394
+ "step": 2310
1395
+ },
1396
+ {
1397
+ "epoch": 0.16,
1398
+ "learning_rate": 0.000123417245221526,
1399
+ "loss": 0.3439,
1400
+ "step": 2320
1401
+ },
1402
+ {
1403
+ "epoch": 0.16,
1404
+ "learning_rate": 0.00012157844744258722,
1405
+ "loss": 0.3439,
1406
+ "step": 2330
1407
+ },
1408
+ {
1409
+ "epoch": 0.16,
1410
+ "learning_rate": 0.00011974744994870517,
1411
+ "loss": 0.3406,
1412
+ "step": 2340
1413
+ },
1414
+ {
1415
+ "epoch": 0.16,
1416
+ "learning_rate": 0.00011792443486203788,
1417
+ "loss": 0.3424,
1418
+ "step": 2350
1419
+ },
1420
+ {
1421
+ "epoch": 0.17,
1422
+ "learning_rate": 0.00011610958351076458,
1423
+ "loss": 0.34,
1424
+ "step": 2360
1425
+ },
1426
+ {
1427
+ "epoch": 0.17,
1428
+ "learning_rate": 0.00011430307641104971,
1429
+ "loss": 0.3345,
1430
+ "step": 2370
1431
+ },
1432
+ {
1433
+ "epoch": 0.17,
1434
+ "learning_rate": 0.00011250509324908767,
1435
+ "loss": 0.3411,
1436
+ "step": 2380
1437
+ },
1438
+ {
1439
+ "epoch": 0.17,
1440
+ "learning_rate": 0.00011071581286323,
1441
+ "loss": 0.3393,
1442
+ "step": 2390
1443
+ },
1444
+ {
1445
+ "epoch": 0.17,
1446
+ "learning_rate": 0.00010893541322619732,
1447
+ "loss": 0.3337,
1448
+ "step": 2400
1449
+ },
1450
+ {
1451
+ "epoch": 0.17,
1452
+ "learning_rate": 0.00010716407142737659,
1453
+ "loss": 0.3428,
1454
+ "step": 2410
1455
+ },
1456
+ {
1457
+ "epoch": 0.17,
1458
+ "learning_rate": 0.00010540196365520754,
1459
+ "loss": 0.3406,
1460
+ "step": 2420
1461
+ },
1462
+ {
1463
+ "epoch": 0.17,
1464
+ "learning_rate": 0.00010364926517965692,
1465
+ "loss": 0.3411,
1466
+ "step": 2430
1467
+ },
1468
+ {
1469
+ "epoch": 0.17,
1470
+ "learning_rate": 0.0001019061503347858,
1471
+ "loss": 0.3356,
1472
+ "step": 2440
1473
+ },
1474
+ {
1475
+ "epoch": 0.17,
1476
+ "learning_rate": 0.00010017279250140891,
1477
+ "loss": 0.3342,
1478
+ "step": 2450
1479
+ },
1480
+ {
1481
+ "epoch": 0.17,
1482
+ "learning_rate": 9.844936408984924e-05,
1483
+ "loss": 0.3381,
1484
+ "step": 2460
1485
+ },
1486
+ {
1487
+ "epoch": 0.17,
1488
+ "learning_rate": 9.673603652278904e-05,
1489
+ "loss": 0.3324,
1490
+ "step": 2470
1491
+ },
1492
+ {
1493
+ "epoch": 0.17,
1494
+ "learning_rate": 9.503298021821905e-05,
1495
+ "loss": 0.3255,
1496
+ "step": 2480
1497
+ },
1498
+ {
1499
+ "epoch": 0.17,
1500
+ "learning_rate": 9.334036457248774e-05,
1501
+ "loss": 0.3333,
1502
+ "step": 2490
1503
+ },
1504
+ {
1505
+ "epoch": 0.18,
1506
+ "learning_rate": 9.165835794345205e-05,
1507
+ "loss": 0.33,
1508
+ "step": 2500
1509
+ },
1510
+ {
1511
+ "epoch": 0.18,
1512
+ "learning_rate": 8.99871276337315e-05,
1513
+ "loss": 0.3396,
1514
+ "step": 2510
1515
+ },
1516
+ {
1517
+ "epoch": 0.18,
1518
+ "learning_rate": 8.832683987406746e-05,
1519
+ "loss": 0.3321,
1520
+ "step": 2520
1521
+ },
1522
+ {
1523
+ "epoch": 0.18,
1524
+ "learning_rate": 8.667765980678851e-05,
1525
+ "loss": 0.3279,
1526
+ "step": 2530
1527
+ },
1528
+ {
1529
+ "epoch": 0.18,
1530
+ "learning_rate": 8.503975146938444e-05,
1531
+ "loss": 0.3337,
1532
+ "step": 2540
1533
+ },
1534
+ {
1535
+ "epoch": 0.18,
1536
+ "learning_rate": 8.341327777819035e-05,
1537
+ "loss": 0.3369,
1538
+ "step": 2550
1539
+ },
1540
+ {
1541
+ "epoch": 0.18,
1542
+ "learning_rate": 8.179840051218167e-05,
1543
+ "loss": 0.3326,
1544
+ "step": 2560
1545
+ },
1546
+ {
1547
+ "epoch": 0.18,
1548
+ "learning_rate": 8.019528029688286e-05,
1549
+ "loss": 0.3292,
1550
+ "step": 2570
1551
+ },
1552
+ {
1553
+ "epoch": 0.18,
1554
+ "learning_rate": 7.860407658839049e-05,
1555
+ "loss": 0.3377,
1556
+ "step": 2580
1557
+ },
1558
+ {
1559
+ "epoch": 0.18,
1560
+ "learning_rate": 7.702494765751285e-05,
1561
+ "loss": 0.333,
1562
+ "step": 2590
1563
+ },
1564
+ {
1565
+ "epoch": 0.18,
1566
+ "learning_rate": 7.545805057402733e-05,
1567
+ "loss": 0.3248,
1568
+ "step": 2600
1569
+ },
1570
+ {
1571
+ "epoch": 0.18,
1572
+ "learning_rate": 7.390354119105722e-05,
1573
+ "loss": 0.3357,
1574
+ "step": 2610
1575
+ },
1576
+ {
1577
+ "epoch": 0.18,
1578
+ "learning_rate": 7.236157412956994e-05,
1579
+ "loss": 0.3326,
1580
+ "step": 2620
1581
+ },
1582
+ {
1583
+ "epoch": 0.18,
1584
+ "learning_rate": 7.0832302762997e-05,
1585
+ "loss": 0.328,
1586
+ "step": 2630
1587
+ },
1588
+ {
1589
+ "epoch": 0.19,
1590
+ "learning_rate": 6.93158792019789e-05,
1591
+ "loss": 0.3235,
1592
+ "step": 2640
1593
+ },
1594
+ {
1595
+ "epoch": 0.19,
1596
+ "learning_rate": 6.781245427923522e-05,
1597
+ "loss": 0.3293,
1598
+ "step": 2650
1599
+ },
1600
+ {
1601
+ "epoch": 0.19,
1602
+ "learning_rate": 6.632217753456174e-05,
1603
+ "loss": 0.3238,
1604
+ "step": 2660
1605
+ },
1606
+ {
1607
+ "epoch": 0.19,
1608
+ "learning_rate": 6.484519719995647e-05,
1609
+ "loss": 0.3301,
1610
+ "step": 2670
1611
+ },
1612
+ {
1613
+ "epoch": 0.19,
1614
+ "learning_rate": 6.338166018487555e-05,
1615
+ "loss": 0.3273,
1616
+ "step": 2680
1617
+ },
1618
+ {
1619
+ "epoch": 0.19,
1620
+ "learning_rate": 6.193171206162065e-05,
1621
+ "loss": 0.3305,
1622
+ "step": 2690
1623
+ },
1624
+ {
1625
+ "epoch": 0.19,
1626
+ "learning_rate": 6.0495497050859574e-05,
1627
+ "loss": 0.3335,
1628
+ "step": 2700
1629
+ },
1630
+ {
1631
+ "epoch": 0.19,
1632
+ "learning_rate": 5.907315800728106e-05,
1633
+ "loss": 0.3396,
1634
+ "step": 2710
1635
+ },
1636
+ {
1637
+ "epoch": 0.19,
1638
+ "learning_rate": 5.766483640538587e-05,
1639
+ "loss": 0.3281,
1640
+ "step": 2720
1641
+ },
1642
+ {
1643
+ "epoch": 0.19,
1644
+ "learning_rate": 5.62706723254145e-05,
1645
+ "loss": 0.3255,
1646
+ "step": 2730
1647
+ },
1648
+ {
1649
+ "epoch": 0.19,
1650
+ "learning_rate": 5.489080443941415e-05,
1651
+ "loss": 0.3336,
1652
+ "step": 2740
1653
+ },
1654
+ {
1655
+ "epoch": 0.19,
1656
+ "learning_rate": 5.352536999744557e-05,
1657
+ "loss": 0.3233,
1658
+ "step": 2750
1659
+ },
1660
+ {
1661
+ "epoch": 0.19,
1662
+ "learning_rate": 5.217450481393129e-05,
1663
+ "loss": 0.3273,
1664
+ "step": 2760
1665
+ },
1666
+ {
1667
+ "epoch": 0.19,
1668
+ "learning_rate": 5.083834325414667e-05,
1669
+ "loss": 0.3329,
1670
+ "step": 2770
1671
+ },
1672
+ {
1673
+ "epoch": 0.2,
1674
+ "learning_rate": 4.951701822085515e-05,
1675
+ "loss": 0.3241,
1676
+ "step": 2780
1677
+ },
1678
+ {
1679
+ "epoch": 0.2,
1680
+ "learning_rate": 4.821066114108892e-05,
1681
+ "loss": 0.3346,
1682
+ "step": 2790
1683
+ },
1684
+ {
1685
+ "epoch": 0.2,
1686
+ "learning_rate": 4.69194019530764e-05,
1687
+ "loss": 0.3358,
1688
+ "step": 2800
1689
+ },
1690
+ {
1691
+ "epoch": 0.2,
1692
+ "learning_rate": 4.564336909331768e-05,
1693
+ "loss": 0.3275,
1694
+ "step": 2810
1695
+ },
1696
+ {
1697
+ "epoch": 0.2,
1698
+ "learning_rate": 4.438268948380972e-05,
1699
+ "loss": 0.3245,
1700
+ "step": 2820
1701
+ },
1702
+ {
1703
+ "epoch": 0.2,
1704
+ "learning_rate": 4.3137488519421656e-05,
1705
+ "loss": 0.33,
1706
+ "step": 2830
1707
+ },
1708
+ {
1709
+ "epoch": 0.2,
1710
+ "learning_rate": 4.1907890055422286e-05,
1711
+ "loss": 0.326,
1712
+ "step": 2840
1713
+ },
1714
+ {
1715
+ "epoch": 0.2,
1716
+ "learning_rate": 4.069401639516075e-05,
1717
+ "loss": 0.3262,
1718
+ "step": 2850
1719
+ },
1720
+ {
1721
+ "epoch": 0.2,
1722
+ "learning_rate": 3.949598827790155e-05,
1723
+ "loss": 0.3266,
1724
+ "step": 2860
1725
+ },
1726
+ {
1727
+ "epoch": 0.2,
1728
+ "learning_rate": 3.831392486681495e-05,
1729
+ "loss": 0.3228,
1730
+ "step": 2870
1731
+ },
1732
+ {
1733
+ "epoch": 0.2,
1734
+ "learning_rate": 3.714794373712431e-05,
1735
+ "loss": 0.3327,
1736
+ "step": 2880
1737
+ },
1738
+ {
1739
+ "epoch": 0.2,
1740
+ "learning_rate": 3.5998160864411476e-05,
1741
+ "loss": 0.3171,
1742
+ "step": 2890
1743
+ },
1744
+ {
1745
+ "epoch": 0.2,
1746
+ "learning_rate": 3.486469061308093e-05,
1747
+ "loss": 0.3197,
1748
+ "step": 2900
1749
+ },
1750
+ {
1751
+ "epoch": 0.2,
1752
+ "learning_rate": 3.3747645724984544e-05,
1753
+ "loss": 0.3268,
1754
+ "step": 2910
1755
+ },
1756
+ {
1757
+ "epoch": 0.2,
1758
+ "learning_rate": 3.264713730820768e-05,
1759
+ "loss": 0.3254,
1760
+ "step": 2920
1761
+ },
1762
+ {
1763
+ "epoch": 0.21,
1764
+ "learning_rate": 3.156327482601742e-05,
1765
+ "loss": 0.3251,
1766
+ "step": 2930
1767
+ },
1768
+ {
1769
+ "epoch": 0.21,
1770
+ "learning_rate": 3.0496166085974943e-05,
1771
+ "loss": 0.3302,
1772
+ "step": 2940
1773
+ },
1774
+ {
1775
+ "epoch": 0.21,
1776
+ "learning_rate": 2.9445917229212193e-05,
1777
+ "loss": 0.3176,
1778
+ "step": 2950
1779
+ },
1780
+ {
1781
+ "epoch": 0.21,
1782
+ "learning_rate": 2.8412632719874532e-05,
1783
+ "loss": 0.3203,
1784
+ "step": 2960
1785
+ },
1786
+ {
1787
+ "epoch": 0.21,
1788
+ "learning_rate": 2.7396415334729964e-05,
1789
+ "loss": 0.3183,
1790
+ "step": 2970
1791
+ },
1792
+ {
1793
+ "epoch": 0.21,
1794
+ "learning_rate": 2.6397366152946523e-05,
1795
+ "loss": 0.3285,
1796
+ "step": 2980
1797
+ },
1798
+ {
1799
+ "epoch": 0.21,
1800
+ "learning_rate": 2.5415584546038096e-05,
1801
+ "loss": 0.3279,
1802
+ "step": 2990
1803
+ },
1804
+ {
1805
+ "epoch": 0.21,
1806
+ "learning_rate": 2.4451168167980497e-05,
1807
+ "loss": 0.3186,
1808
+ "step": 3000
1809
+ },
1810
+ {
1811
+ "epoch": 0.21,
1812
+ "learning_rate": 2.350421294549825e-05,
1813
+ "loss": 0.3222,
1814
+ "step": 3010
1815
+ },
1816
+ {
1817
+ "epoch": 0.21,
1818
+ "learning_rate": 2.2574813068522894e-05,
1819
+ "loss": 0.3271,
1820
+ "step": 3020
1821
+ },
1822
+ {
1823
+ "epoch": 0.21,
1824
+ "learning_rate": 2.166306098082451e-05,
1825
+ "loss": 0.3253,
1826
+ "step": 3030
1827
+ },
1828
+ {
1829
+ "epoch": 0.21,
1830
+ "learning_rate": 2.0769047370816553e-05,
1831
+ "loss": 0.3234,
1832
+ "step": 3040
1833
+ },
1834
+ {
1835
+ "epoch": 0.21,
1836
+ "learning_rate": 1.989286116253557e-05,
1837
+ "loss": 0.3256,
1838
+ "step": 3050
1839
+ },
1840
+ {
1841
+ "epoch": 0.21,
1842
+ "learning_rate": 1.903458950679613e-05,
1843
+ "loss": 0.3234,
1844
+ "step": 3060
1845
+ },
1846
+ {
1847
+ "epoch": 0.22,
1848
+ "learning_rate": 1.8194317772522362e-05,
1849
+ "loss": 0.3228,
1850
+ "step": 3070
1851
+ },
1852
+ {
1853
+ "epoch": 0.22,
1854
+ "learning_rate": 1.7372129538256667e-05,
1855
+ "loss": 0.32,
1856
+ "step": 3080
1857
+ },
1858
+ {
1859
+ "epoch": 0.22,
1860
+ "learning_rate": 1.6568106583846378e-05,
1861
+ "loss": 0.3198,
1862
+ "step": 3090
1863
+ },
1864
+ {
1865
+ "epoch": 0.22,
1866
+ "learning_rate": 1.5782328882309484e-05,
1867
+ "loss": 0.331,
1868
+ "step": 3100
1869
+ },
1870
+ {
1871
+ "epoch": 0.22,
1872
+ "learning_rate": 1.5014874591880157e-05,
1873
+ "loss": 0.3264,
1874
+ "step": 3110
1875
+ },
1876
+ {
1877
+ "epoch": 0.22,
1878
+ "learning_rate": 1.4265820048234447e-05,
1879
+ "loss": 0.3253,
1880
+ "step": 3120
1881
+ },
1882
+ {
1883
+ "epoch": 0.22,
1884
+ "learning_rate": 1.3535239756897566e-05,
1885
+ "loss": 0.3187,
1886
+ "step": 3130
1887
+ },
1888
+ {
1889
+ "epoch": 0.22,
1890
+ "learning_rate": 1.2823206385833187e-05,
1891
+ "loss": 0.3199,
1892
+ "step": 3140
1893
+ },
1894
+ {
1895
+ "epoch": 0.22,
1896
+ "learning_rate": 1.212979075821532e-05,
1897
+ "loss": 0.3225,
1898
+ "step": 3150
1899
+ },
1900
+ {
1901
+ "epoch": 0.22,
1902
+ "learning_rate": 1.1455061845383852e-05,
1903
+ "loss": 0.3195,
1904
+ "step": 3160
1905
+ },
1906
+ {
1907
+ "epoch": 0.22,
1908
+ "learning_rate": 1.0799086759984333e-05,
1909
+ "loss": 0.3163,
1910
+ "step": 3170
1911
+ },
1912
+ {
1913
+ "epoch": 0.22,
1914
+ "learning_rate": 1.016193074929237e-05,
1915
+ "loss": 0.3217,
1916
+ "step": 3180
1917
+ },
1918
+ {
1919
+ "epoch": 0.22,
1920
+ "learning_rate": 9.543657188723876e-06,
1921
+ "loss": 0.3204,
1922
+ "step": 3190
1923
+ },
1924
+ {
1925
+ "epoch": 0.22,
1926
+ "learning_rate": 8.944327575531275e-06,
1927
+ "loss": 0.3196,
1928
+ "step": 3200
1929
+ },
1930
+ {
1931
+ "epoch": 0.23,
1932
+ "learning_rate": 8.364001522686726e-06,
1933
+ "loss": 0.3326,
1934
+ "step": 3210
1935
+ },
1936
+ {
1937
+ "epoch": 0.23,
1938
+ "learning_rate": 7.802736752952533e-06,
1939
+ "loss": 0.3153,
1940
+ "step": 3220
1941
+ },
1942
+ {
1943
+ "epoch": 0.23,
1944
+ "learning_rate": 7.260589093139736e-06,
1945
+ "loss": 0.323,
1946
+ "step": 3230
1947
+ },
1948
+ {
1949
+ "epoch": 0.23,
1950
+ "learning_rate": 6.737612468555221e-06,
1951
+ "loss": 0.3192,
1952
+ "step": 3240
1953
+ },
1954
+ {
1955
+ "epoch": 0.23,
1956
+ "learning_rate": 6.2338588976380115e-06,
1957
+ "loss": 0.3166,
1958
+ "step": 3250
1959
+ },
1960
+ {
1961
+ "epoch": 0.23,
1962
+ "learning_rate": 5.749378486785162e-06,
1963
+ "loss": 0.3134,
1964
+ "step": 3260
1965
+ },
1966
+ {
1967
+ "epoch": 0.23,
1968
+ "learning_rate": 5.2842194253679424e-06,
1969
+ "loss": 0.3225,
1970
+ "step": 3270
1971
+ },
1972
+ {
1973
+ "epoch": 0.23,
1974
+ "learning_rate": 4.8384279809385426e-06,
1975
+ "loss": 0.3251,
1976
+ "step": 3280
1977
+ },
1978
+ {
1979
+ "epoch": 0.23,
1980
+ "learning_rate": 4.41204849462804e-06,
1981
+ "loss": 0.3228,
1982
+ "step": 3290
1983
+ },
1984
+ {
1985
+ "epoch": 0.23,
1986
+ "learning_rate": 4.005123376735997e-06,
1987
+ "loss": 0.3181,
1988
+ "step": 3300
1989
+ },
1990
+ {
1991
+ "epoch": 0.23,
1992
+ "learning_rate": 3.617693102512032e-06,
1993
+ "loss": 0.3135,
1994
+ "step": 3310
1995
+ },
1996
+ {
1997
+ "epoch": 0.23,
1998
+ "learning_rate": 3.2497962081299514e-06,
1999
+ "loss": 0.3196,
2000
+ "step": 3320
2001
+ },
2002
+ {
2003
+ "epoch": 0.23,
2004
+ "learning_rate": 2.9014692868546633e-06,
2005
+ "loss": 0.3143,
2006
+ "step": 3330
2007
+ },
2008
+ {
2009
+ "epoch": 0.23,
2010
+ "learning_rate": 2.572746985402419e-06,
2011
+ "loss": 0.3124,
2012
+ "step": 3340
2013
+ },
2014
+ {
2015
+ "epoch": 0.24,
2016
+ "learning_rate": 2.2636620004946154e-06,
2017
+ "loss": 0.3157,
2018
+ "step": 3350
2019
+ },
2020
+ {
2021
+ "epoch": 0.24,
2022
+ "learning_rate": 1.974245075605574e-06,
2023
+ "loss": 0.3225,
2024
+ "step": 3360
2025
+ },
2026
+ {
2027
+ "epoch": 0.24,
2028
+ "learning_rate": 1.7045249979046995e-06,
2029
+ "loss": 0.3165,
2030
+ "step": 3370
2031
+ },
2032
+ {
2033
+ "epoch": 0.24,
2034
+ "learning_rate": 1.4545285953929677e-06,
2035
+ "loss": 0.3178,
2036
+ "step": 3380
2037
+ },
2038
+ {
2039
+ "epoch": 0.24,
2040
+ "learning_rate": 1.224280734234573e-06,
2041
+ "loss": 0.3129,
2042
+ "step": 3390
2043
+ },
2044
+ {
2045
+ "epoch": 0.24,
2046
+ "learning_rate": 1.013804316283573e-06,
2047
+ "loss": 0.3187,
2048
+ "step": 3400
2049
+ },
2050
+ {
2051
+ "epoch": 0.24,
2052
+ "learning_rate": 8.231202768059332e-07,
2053
+ "loss": 0.3178,
2054
+ "step": 3410
2055
+ },
2056
+ {
2057
+ "epoch": 0.24,
2058
+ "learning_rate": 6.522475823970808e-07,
2059
+ "loss": 0.3252,
2060
+ "step": 3420
2061
+ },
2062
+ {
2063
+ "epoch": 0.24,
2064
+ "learning_rate": 5.012032290955037e-07,
2065
+ "loss": 0.311,
2066
+ "step": 3430
2067
+ },
2068
+ {
2069
+ "epoch": 0.24,
2070
+ "learning_rate": 3.7000224069216883e-07,
2071
+ "loss": 0.3253,
2072
+ "step": 3440
2073
+ },
2074
+ {
2075
+ "epoch": 0.24,
2076
+ "learning_rate": 2.586576672361396e-07,
2077
+ "loss": 0.3205,
2078
+ "step": 3450
2079
+ },
2080
+ {
2081
+ "epoch": 0.24,
2082
+ "learning_rate": 1.671805837365703e-07,
2083
+ "loss": 0.3196,
2084
+ "step": 3460
2085
+ },
2086
+ {
2087
+ "epoch": 0.24,
2088
+ "learning_rate": 9.558008906112026e-08,
2089
+ "loss": 0.3166,
2090
+ "step": 3470
2091
+ },
2092
+ {
2093
+ "epoch": 0.24,
2094
+ "learning_rate": 4.386330503090008e-08,
2095
+ "loss": 0.3255,
2096
+ "step": 3480
2097
+ },
2098
+ {
2099
+ "epoch": 0.24,
2100
+ "learning_rate": 1.2035375712105001e-08,
2101
+ "loss": 0.3181,
2102
+ "step": 3490
2103
+ },
2104
+ {
2105
+ "epoch": 0.25,
2106
+ "learning_rate": 9.946690433526584e-11,
2107
+ "loss": 0.3212,
2108
+ "step": 3500
2109
+ }
2110
+ ],
2111
+ "max_steps": 3500,
2112
+ "num_train_epochs": 1,
2113
+ "total_flos": 2.21324249137152e+18,
2114
+ "trial_name": null,
2115
+ "trial_params": null
2116
+ }
checkpoint-3500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a9ed45480b6d3ebc301a155cd6615284a1e4d59a14aa91af42a3b61f071cfb6
3
+ size 4347
runs/Nov17_16-10-29_18e7f5055510/events.out.tfevents.1700237477.18e7f5055510.7008.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ee7caee4971de8dd5e46f68ee6f3f7afa9d9067f118d6af102022dcdc7d5243
3
+ size 59357
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "__type": "AddedToken",
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "clean_up_tokenization_spaces": false,
11
+ "eos_token": {
12
+ "__type": "AddedToken",
13
+ "content": "</s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "legacy": false,
20
+ "model_max_length": 1000000000000000019884624838656,
21
+ "pad_token": null,
22
+ "sp_model_kwargs": {},
23
+ "tokenizer_class": "LlamaTokenizer",
24
+ "unk_token": {
25
+ "__type": "AddedToken",
26
+ "content": "<unk>",
27
+ "lstrip": false,
28
+ "normalized": true,
29
+ "rstrip": false,
30
+ "single_word": false
31
+ }
32
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a9ed45480b6d3ebc301a155cd6615284a1e4d59a14aa91af42a3b61f071cfb6
3
+ size 4347