0x-YuAN commited on
Commit
2ce193c
1 Parent(s): 2e47e18

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +9 -0
  2. adapter_config.json +21 -0
  3. adapter_model.bin +3 -0
  4. all_results.json +7 -0
  5. checkpoint-100/README.md +9 -0
  6. checkpoint-100/adapter_config.json +21 -0
  7. checkpoint-100/adapter_model.bin +3 -0
  8. checkpoint-100/finetuning_args.json +16 -0
  9. checkpoint-100/optimizer.pt +3 -0
  10. checkpoint-100/rng_state.pth +3 -0
  11. checkpoint-100/scaler.pt +3 -0
  12. checkpoint-100/scheduler.pt +3 -0
  13. checkpoint-100/trainer_state.json +136 -0
  14. checkpoint-100/training_args.bin +3 -0
  15. checkpoint-1000/README.md +9 -0
  16. checkpoint-1000/adapter_config.json +21 -0
  17. checkpoint-1000/adapter_model.bin +3 -0
  18. checkpoint-1000/finetuning_args.json +16 -0
  19. checkpoint-1000/optimizer.pt +3 -0
  20. checkpoint-1000/rng_state.pth +3 -0
  21. checkpoint-1000/scaler.pt +3 -0
  22. checkpoint-1000/scheduler.pt +3 -0
  23. checkpoint-1000/trainer_state.json +1216 -0
  24. checkpoint-1000/training_args.bin +3 -0
  25. checkpoint-1100/README.md +9 -0
  26. checkpoint-1100/adapter_config.json +21 -0
  27. checkpoint-1100/adapter_model.bin +3 -0
  28. checkpoint-1100/finetuning_args.json +16 -0
  29. checkpoint-1100/optimizer.pt +3 -0
  30. checkpoint-1100/rng_state.pth +3 -0
  31. checkpoint-1100/scaler.pt +3 -0
  32. checkpoint-1100/scheduler.pt +3 -0
  33. checkpoint-1100/trainer_state.json +1336 -0
  34. checkpoint-1100/training_args.bin +3 -0
  35. checkpoint-1200/README.md +9 -0
  36. checkpoint-1200/adapter_config.json +21 -0
  37. checkpoint-1200/adapter_model.bin +3 -0
  38. checkpoint-1200/finetuning_args.json +16 -0
  39. checkpoint-1200/optimizer.pt +3 -0
  40. checkpoint-1200/rng_state.pth +3 -0
  41. checkpoint-1200/scaler.pt +3 -0
  42. checkpoint-1200/scheduler.pt +3 -0
  43. checkpoint-1200/trainer_state.json +1456 -0
  44. checkpoint-1200/training_args.bin +3 -0
  45. checkpoint-1300/README.md +9 -0
  46. checkpoint-1300/adapter_config.json +21 -0
  47. checkpoint-1300/adapter_model.bin +3 -0
  48. checkpoint-1300/finetuning_args.json +16 -0
  49. checkpoint-1300/optimizer.pt +3 -0
  50. checkpoint-1300/rng_state.pth +3 -0
README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
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": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
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:c349ed41c4038ffd9cfeb2e053cd0502b7b8d8bd0983d9ddfe3860d1dc41143d
3
+ size 16821197
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.1874766861998836,
4
+ "train_runtime": 24070.4048,
5
+ "train_samples_per_second": 1.47,
6
+ "train_steps_per_second": 0.092
7
+ }
checkpoint-100/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-100/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
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": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-100/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9346a42ce9edea91942d5ae98b496aad9ddbd0eea98bf16c3412230e72109ac2
3
+ size 16821197
checkpoint-100/finetuning_args.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dpo_beta": 0.1,
3
+ "finetuning_type": "lora",
4
+ "lora_alpha": 32.0,
5
+ "lora_dropout": 0.1,
6
+ "lora_rank": 8,
7
+ "lora_target": [
8
+ "q_proj",
9
+ "v_proj"
10
+ ],
11
+ "name_module_trainable": "mlp",
12
+ "num_hidden_layers": 32,
13
+ "num_layer_trainable": 3,
14
+ "ppo_score_norm": false,
15
+ "resume_lora_training": true
16
+ }
checkpoint-100/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d66a7945c694235b16c945166db80115b80a308eeec6766145aba1cb40706d8
3
+ size 33629765
checkpoint-100/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee8c8d87baa5b2a12562dd15e1c7d92dfdce63e9eaa220006c09635666c75025
3
+ size 14575
checkpoint-100/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfa44e8523f62833816d29aa6c576eaa7783e3bbdb3e132e248b1d8aaee6132b
3
+ size 557
checkpoint-100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb3acfcdc577f60174c56ddee3e96bd7647789ad2fb245d0589b4a8acfa5b089
3
+ size 627
checkpoint-100/trainer_state.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.13559322033898305,
5
+ "global_step": 100,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.999936908557769e-05,
13
+ "loss": 1.8685,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.999747637415502e-05,
19
+ "loss": 1.698,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.999432196126308e-05,
25
+ "loss": 1.513,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.998990600611505e-05,
31
+ "loss": 1.5436,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 4.998422873159811e-05,
37
+ "loss": 1.5148,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 4.9977290424262214e-05,
43
+ "loss": 1.4708,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.996909143430562e-05,
49
+ "loss": 1.4639,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.05,
54
+ "learning_rate": 4.995963217555719e-05,
55
+ "loss": 1.4528,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.06,
60
+ "learning_rate": 4.994891312545556e-05,
61
+ "loss": 1.5201,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.07,
66
+ "learning_rate": 4.9936934825025e-05,
67
+ "loss": 1.4794,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.07,
72
+ "learning_rate": 4.992369787884809e-05,
73
+ "loss": 1.4549,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.08,
78
+ "learning_rate": 4.9909202955035264e-05,
79
+ "loss": 1.4842,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.09,
84
+ "learning_rate": 4.989345078519104e-05,
85
+ "loss": 1.4832,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.09,
90
+ "learning_rate": 4.9876442164377115e-05,
91
+ "loss": 1.4952,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.1,
96
+ "learning_rate": 4.98581779510722e-05,
97
+ "loss": 1.4078,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.11,
102
+ "learning_rate": 4.9838659067128766e-05,
103
+ "loss": 1.5192,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.12,
108
+ "learning_rate": 4.981788649772644e-05,
109
+ "loss": 1.4805,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.12,
114
+ "learning_rate": 4.97958612913223e-05,
115
+ "loss": 1.4352,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.13,
120
+ "learning_rate": 4.9772584559597976e-05,
121
+ "loss": 1.444,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.14,
126
+ "learning_rate": 4.9748057477403546e-05,
127
+ "loss": 1.4018,
128
+ "step": 100
129
+ }
130
+ ],
131
+ "max_steps": 2211,
132
+ "num_train_epochs": 3,
133
+ "total_flos": 4.894198879125504e+16,
134
+ "trial_name": null,
135
+ "trial_params": null
136
+ }
checkpoint-100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0090099d32ad29c1bd6d98560579b7c6ef7a6d711729f1926d4fd6f8220f669c
3
+ size 3361
checkpoint-1000/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-1000/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
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": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-1000/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb5de60eb995a373312cfee065cca08bb4955ad654feef9e29dc187166e06411
3
+ size 16821197
checkpoint-1000/finetuning_args.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dpo_beta": 0.1,
3
+ "finetuning_type": "lora",
4
+ "lora_alpha": 32.0,
5
+ "lora_dropout": 0.1,
6
+ "lora_rank": 8,
7
+ "lora_target": [
8
+ "q_proj",
9
+ "v_proj"
10
+ ],
11
+ "name_module_trainable": "mlp",
12
+ "num_hidden_layers": 32,
13
+ "num_layer_trainable": 3,
14
+ "ppo_score_norm": false,
15
+ "resume_lora_training": true
16
+ }
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d38f819e6791b02ec0d65f5a967c1f48f9b81c80a4619fc85e5b174d27517cd
3
+ size 33629893
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfee4d9acc652db49dfc46db8f739d0594c412b1502891409ef6f8fb1dddf0f4
3
+ size 14575
checkpoint-1000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68cff80b680ddf6e7abbef98b5f336b97f9b5963e2209307f639383870e8cc71
3
+ size 557
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b512ab8a3f082e00da33b1dc112517693ea1c1c6731a2eb14c2ed1af53333b14
3
+ size 627
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,1216 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.3559322033898304,
5
+ "global_step": 1000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.999936908557769e-05,
13
+ "loss": 1.8685,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.999747637415502e-05,
19
+ "loss": 1.698,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.999432196126308e-05,
25
+ "loss": 1.513,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.998990600611505e-05,
31
+ "loss": 1.5436,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 4.998422873159811e-05,
37
+ "loss": 1.5148,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 4.9977290424262214e-05,
43
+ "loss": 1.4708,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.996909143430562e-05,
49
+ "loss": 1.4639,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.05,
54
+ "learning_rate": 4.995963217555719e-05,
55
+ "loss": 1.4528,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.06,
60
+ "learning_rate": 4.994891312545556e-05,
61
+ "loss": 1.5201,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.07,
66
+ "learning_rate": 4.9936934825025e-05,
67
+ "loss": 1.4794,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.07,
72
+ "learning_rate": 4.992369787884809e-05,
73
+ "loss": 1.4549,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.08,
78
+ "learning_rate": 4.9909202955035264e-05,
79
+ "loss": 1.4842,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.09,
84
+ "learning_rate": 4.989345078519104e-05,
85
+ "loss": 1.4832,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.09,
90
+ "learning_rate": 4.9876442164377115e-05,
91
+ "loss": 1.4952,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.1,
96
+ "learning_rate": 4.98581779510722e-05,
97
+ "loss": 1.4078,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.11,
102
+ "learning_rate": 4.9838659067128766e-05,
103
+ "loss": 1.5192,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.12,
108
+ "learning_rate": 4.981788649772644e-05,
109
+ "loss": 1.4805,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.12,
114
+ "learning_rate": 4.97958612913223e-05,
115
+ "loss": 1.4352,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.13,
120
+ "learning_rate": 4.9772584559597976e-05,
121
+ "loss": 1.444,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.14,
126
+ "learning_rate": 4.9748057477403546e-05,
127
+ "loss": 1.4018,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.14,
132
+ "learning_rate": 4.9722281282698186e-05,
133
+ "loss": 1.4472,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.15,
138
+ "learning_rate": 4.9695257276487736e-05,
139
+ "loss": 1.3842,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.16,
144
+ "learning_rate": 4.966698682275903e-05,
145
+ "loss": 1.4116,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.16,
150
+ "learning_rate": 4.9637471348411005e-05,
151
+ "loss": 1.3827,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.17,
156
+ "learning_rate": 4.960671234318276e-05,
157
+ "loss": 1.3718,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.18,
162
+ "learning_rate": 4.957471135957828e-05,
163
+ "loss": 1.3204,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.18,
168
+ "learning_rate": 4.9541470012788135e-05,
169
+ "loss": 1.3307,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 0.19,
174
+ "learning_rate": 4.9506989980607945e-05,
175
+ "loss": 1.3216,
176
+ "step": 140
177
+ },
178
+ {
179
+ "epoch": 0.2,
180
+ "learning_rate": 4.9471273003353655e-05,
181
+ "loss": 1.3268,
182
+ "step": 145
183
+ },
184
+ {
185
+ "epoch": 0.2,
186
+ "learning_rate": 4.943432088377378e-05,
187
+ "loss": 1.4072,
188
+ "step": 150
189
+ },
190
+ {
191
+ "epoch": 0.21,
192
+ "learning_rate": 4.93961354869583e-05,
193
+ "loss": 1.2871,
194
+ "step": 155
195
+ },
196
+ {
197
+ "epoch": 0.22,
198
+ "learning_rate": 4.935671874024465e-05,
199
+ "loss": 1.3451,
200
+ "step": 160
201
+ },
202
+ {
203
+ "epoch": 0.22,
204
+ "learning_rate": 4.931607263312032e-05,
205
+ "loss": 1.3447,
206
+ "step": 165
207
+ },
208
+ {
209
+ "epoch": 0.23,
210
+ "learning_rate": 4.9274199217122565e-05,
211
+ "loss": 1.3573,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 0.24,
216
+ "learning_rate": 4.923110060573472e-05,
217
+ "loss": 1.2914,
218
+ "step": 175
219
+ },
220
+ {
221
+ "epoch": 0.24,
222
+ "learning_rate": 4.9186778974279625e-05,
223
+ "loss": 1.3249,
224
+ "step": 180
225
+ },
226
+ {
227
+ "epoch": 0.25,
228
+ "learning_rate": 4.914123655980981e-05,
229
+ "loss": 1.3253,
230
+ "step": 185
231
+ },
232
+ {
233
+ "epoch": 0.26,
234
+ "learning_rate": 4.909447566099457e-05,
235
+ "loss": 1.2495,
236
+ "step": 190
237
+ },
238
+ {
239
+ "epoch": 0.26,
240
+ "learning_rate": 4.9046498638003925e-05,
241
+ "loss": 1.2333,
242
+ "step": 195
243
+ },
244
+ {
245
+ "epoch": 0.27,
246
+ "learning_rate": 4.8997307912389554e-05,
247
+ "loss": 1.3015,
248
+ "step": 200
249
+ },
250
+ {
251
+ "epoch": 0.28,
252
+ "learning_rate": 4.89469059669625e-05,
253
+ "loss": 1.3331,
254
+ "step": 205
255
+ },
256
+ {
257
+ "epoch": 0.28,
258
+ "learning_rate": 4.889529534566791e-05,
259
+ "loss": 1.2935,
260
+ "step": 210
261
+ },
262
+ {
263
+ "epoch": 0.29,
264
+ "learning_rate": 4.884247865345662e-05,
265
+ "loss": 1.3166,
266
+ "step": 215
267
+ },
268
+ {
269
+ "epoch": 0.3,
270
+ "learning_rate": 4.8788458556153635e-05,
271
+ "loss": 1.3156,
272
+ "step": 220
273
+ },
274
+ {
275
+ "epoch": 0.31,
276
+ "learning_rate": 4.873323778032366e-05,
277
+ "loss": 1.2931,
278
+ "step": 225
279
+ },
280
+ {
281
+ "epoch": 0.31,
282
+ "learning_rate": 4.8676819113133386e-05,
283
+ "loss": 1.3569,
284
+ "step": 230
285
+ },
286
+ {
287
+ "epoch": 0.32,
288
+ "learning_rate": 4.861920540221089e-05,
289
+ "loss": 1.2342,
290
+ "step": 235
291
+ },
292
+ {
293
+ "epoch": 0.33,
294
+ "learning_rate": 4.856039955550186e-05,
295
+ "loss": 1.3015,
296
+ "step": 240
297
+ },
298
+ {
299
+ "epoch": 0.33,
300
+ "learning_rate": 4.8500404541122836e-05,
301
+ "loss": 1.3191,
302
+ "step": 245
303
+ },
304
+ {
305
+ "epoch": 0.34,
306
+ "learning_rate": 4.8439223387211406e-05,
307
+ "loss": 1.3178,
308
+ "step": 250
309
+ },
310
+ {
311
+ "epoch": 0.35,
312
+ "learning_rate": 4.837685918177337e-05,
313
+ "loss": 1.2669,
314
+ "step": 255
315
+ },
316
+ {
317
+ "epoch": 0.35,
318
+ "learning_rate": 4.8313315072526855e-05,
319
+ "loss": 1.2218,
320
+ "step": 260
321
+ },
322
+ {
323
+ "epoch": 0.36,
324
+ "learning_rate": 4.824859426674345e-05,
325
+ "loss": 1.304,
326
+ "step": 265
327
+ },
328
+ {
329
+ "epoch": 0.37,
330
+ "learning_rate": 4.818270003108635e-05,
331
+ "loss": 1.311,
332
+ "step": 270
333
+ },
334
+ {
335
+ "epoch": 0.37,
336
+ "learning_rate": 4.8115635691445436e-05,
337
+ "loss": 1.3036,
338
+ "step": 275
339
+ },
340
+ {
341
+ "epoch": 0.38,
342
+ "learning_rate": 4.8047404632769446e-05,
343
+ "loss": 1.258,
344
+ "step": 280
345
+ },
346
+ {
347
+ "epoch": 0.39,
348
+ "learning_rate": 4.797801029889508e-05,
349
+ "loss": 1.3186,
350
+ "step": 285
351
+ },
352
+ {
353
+ "epoch": 0.39,
354
+ "learning_rate": 4.7907456192373245e-05,
355
+ "loss": 1.395,
356
+ "step": 290
357
+ },
358
+ {
359
+ "epoch": 0.4,
360
+ "learning_rate": 4.783574587429219e-05,
361
+ "loss": 1.2793,
362
+ "step": 295
363
+ },
364
+ {
365
+ "epoch": 0.41,
366
+ "learning_rate": 4.776288296409785e-05,
367
+ "loss": 1.3625,
368
+ "step": 300
369
+ },
370
+ {
371
+ "epoch": 0.41,
372
+ "learning_rate": 4.768887113941107e-05,
373
+ "loss": 1.2453,
374
+ "step": 305
375
+ },
376
+ {
377
+ "epoch": 0.42,
378
+ "learning_rate": 4.761371413584209e-05,
379
+ "loss": 1.3312,
380
+ "step": 310
381
+ },
382
+ {
383
+ "epoch": 0.43,
384
+ "learning_rate": 4.7537415746801876e-05,
385
+ "loss": 1.2627,
386
+ "step": 315
387
+ },
388
+ {
389
+ "epoch": 0.43,
390
+ "learning_rate": 4.745997982331077e-05,
391
+ "loss": 1.2231,
392
+ "step": 320
393
+ },
394
+ {
395
+ "epoch": 0.44,
396
+ "learning_rate": 4.738141027380405e-05,
397
+ "loss": 1.2063,
398
+ "step": 325
399
+ },
400
+ {
401
+ "epoch": 0.45,
402
+ "learning_rate": 4.730171106393466e-05,
403
+ "loss": 1.3148,
404
+ "step": 330
405
+ },
406
+ {
407
+ "epoch": 0.45,
408
+ "learning_rate": 4.722088621637309e-05,
409
+ "loss": 1.2787,
410
+ "step": 335
411
+ },
412
+ {
413
+ "epoch": 0.46,
414
+ "learning_rate": 4.7138939810604286e-05,
415
+ "loss": 1.2111,
416
+ "step": 340
417
+ },
418
+ {
419
+ "epoch": 0.47,
420
+ "learning_rate": 4.7055875982721794e-05,
421
+ "loss": 1.3034,
422
+ "step": 345
423
+ },
424
+ {
425
+ "epoch": 0.47,
426
+ "learning_rate": 4.697169892521898e-05,
427
+ "loss": 1.288,
428
+ "step": 350
429
+ },
430
+ {
431
+ "epoch": 0.48,
432
+ "learning_rate": 4.688641288677739e-05,
433
+ "loss": 1.2935,
434
+ "step": 355
435
+ },
436
+ {
437
+ "epoch": 0.49,
438
+ "learning_rate": 4.680002217205238e-05,
439
+ "loss": 1.3183,
440
+ "step": 360
441
+ },
442
+ {
443
+ "epoch": 0.49,
444
+ "learning_rate": 4.6712531141455785e-05,
445
+ "loss": 1.1688,
446
+ "step": 365
447
+ },
448
+ {
449
+ "epoch": 0.5,
450
+ "learning_rate": 4.662394421093582e-05,
451
+ "loss": 1.2734,
452
+ "step": 370
453
+ },
454
+ {
455
+ "epoch": 0.51,
456
+ "learning_rate": 4.653426585175428e-05,
457
+ "loss": 1.2155,
458
+ "step": 375
459
+ },
460
+ {
461
+ "epoch": 0.52,
462
+ "learning_rate": 4.644350059026076e-05,
463
+ "loss": 1.2641,
464
+ "step": 380
465
+ },
466
+ {
467
+ "epoch": 0.52,
468
+ "learning_rate": 4.6351653007664276e-05,
469
+ "loss": 1.2819,
470
+ "step": 385
471
+ },
472
+ {
473
+ "epoch": 0.53,
474
+ "learning_rate": 4.625872773980197e-05,
475
+ "loss": 1.2435,
476
+ "step": 390
477
+ },
478
+ {
479
+ "epoch": 0.54,
480
+ "learning_rate": 4.61647294769052e-05,
481
+ "loss": 1.2541,
482
+ "step": 395
483
+ },
484
+ {
485
+ "epoch": 0.54,
486
+ "learning_rate": 4.6069662963362724e-05,
487
+ "loss": 1.2464,
488
+ "step": 400
489
+ },
490
+ {
491
+ "epoch": 0.55,
492
+ "learning_rate": 4.597353299748132e-05,
493
+ "loss": 1.2843,
494
+ "step": 405
495
+ },
496
+ {
497
+ "epoch": 0.56,
498
+ "learning_rate": 4.587634443124352e-05,
499
+ "loss": 1.3026,
500
+ "step": 410
501
+ },
502
+ {
503
+ "epoch": 0.56,
504
+ "learning_rate": 4.577810217006279e-05,
505
+ "loss": 1.2905,
506
+ "step": 415
507
+ },
508
+ {
509
+ "epoch": 0.57,
510
+ "learning_rate": 4.567881117253587e-05,
511
+ "loss": 1.2689,
512
+ "step": 420
513
+ },
514
+ {
515
+ "epoch": 0.58,
516
+ "learning_rate": 4.557847645019257e-05,
517
+ "loss": 1.2959,
518
+ "step": 425
519
+ },
520
+ {
521
+ "epoch": 0.58,
522
+ "learning_rate": 4.547710306724274e-05,
523
+ "loss": 1.2237,
524
+ "step": 430
525
+ },
526
+ {
527
+ "epoch": 0.59,
528
+ "learning_rate": 4.537469614032074e-05,
529
+ "loss": 1.2885,
530
+ "step": 435
531
+ },
532
+ {
533
+ "epoch": 0.6,
534
+ "learning_rate": 4.5271260838227124e-05,
535
+ "loss": 1.2622,
536
+ "step": 440
537
+ },
538
+ {
539
+ "epoch": 0.6,
540
+ "learning_rate": 4.516680238166783e-05,
541
+ "loss": 1.2567,
542
+ "step": 445
543
+ },
544
+ {
545
+ "epoch": 0.61,
546
+ "learning_rate": 4.506132604299057e-05,
547
+ "loss": 1.2618,
548
+ "step": 450
549
+ },
550
+ {
551
+ "epoch": 0.62,
552
+ "learning_rate": 4.4954837145918826e-05,
553
+ "loss": 1.249,
554
+ "step": 455
555
+ },
556
+ {
557
+ "epoch": 0.62,
558
+ "learning_rate": 4.484734106528307e-05,
559
+ "loss": 1.2891,
560
+ "step": 460
561
+ },
562
+ {
563
+ "epoch": 0.63,
564
+ "learning_rate": 4.473884322674952e-05,
565
+ "loss": 1.2423,
566
+ "step": 465
567
+ },
568
+ {
569
+ "epoch": 0.64,
570
+ "learning_rate": 4.462934910654624e-05,
571
+ "loss": 1.1933,
572
+ "step": 470
573
+ },
574
+ {
575
+ "epoch": 0.64,
576
+ "learning_rate": 4.451886423118682e-05,
577
+ "loss": 1.1707,
578
+ "step": 475
579
+ },
580
+ {
581
+ "epoch": 0.65,
582
+ "learning_rate": 4.4407394177191366e-05,
583
+ "loss": 1.2059,
584
+ "step": 480
585
+ },
586
+ {
587
+ "epoch": 0.66,
588
+ "learning_rate": 4.429494457080505e-05,
589
+ "loss": 1.2385,
590
+ "step": 485
591
+ },
592
+ {
593
+ "epoch": 0.66,
594
+ "learning_rate": 4.418152108771414e-05,
595
+ "loss": 1.2545,
596
+ "step": 490
597
+ },
598
+ {
599
+ "epoch": 0.67,
600
+ "learning_rate": 4.406712945275955e-05,
601
+ "loss": 1.2488,
602
+ "step": 495
603
+ },
604
+ {
605
+ "epoch": 0.68,
606
+ "learning_rate": 4.395177543964787e-05,
607
+ "loss": 1.2613,
608
+ "step": 500
609
+ },
610
+ {
611
+ "epoch": 0.68,
612
+ "learning_rate": 4.383546487065993e-05,
613
+ "loss": 1.1931,
614
+ "step": 505
615
+ },
616
+ {
617
+ "epoch": 0.69,
618
+ "learning_rate": 4.3718203616356954e-05,
619
+ "loss": 1.231,
620
+ "step": 510
621
+ },
622
+ {
623
+ "epoch": 0.7,
624
+ "learning_rate": 4.35999975952843e-05,
625
+ "loss": 1.1383,
626
+ "step": 515
627
+ },
628
+ {
629
+ "epoch": 0.71,
630
+ "learning_rate": 4.348085277367262e-05,
631
+ "loss": 1.1776,
632
+ "step": 520
633
+ },
634
+ {
635
+ "epoch": 0.71,
636
+ "learning_rate": 4.336077516513682e-05,
637
+ "loss": 1.2254,
638
+ "step": 525
639
+ },
640
+ {
641
+ "epoch": 0.72,
642
+ "learning_rate": 4.323977083037252e-05,
643
+ "loss": 1.2157,
644
+ "step": 530
645
+ },
646
+ {
647
+ "epoch": 0.73,
648
+ "learning_rate": 4.3117845876850085e-05,
649
+ "loss": 1.1688,
650
+ "step": 535
651
+ },
652
+ {
653
+ "epoch": 0.73,
654
+ "learning_rate": 4.299500645850647e-05,
655
+ "loss": 1.2387,
656
+ "step": 540
657
+ },
658
+ {
659
+ "epoch": 0.74,
660
+ "learning_rate": 4.287125877543453e-05,
661
+ "loss": 1.2281,
662
+ "step": 545
663
+ },
664
+ {
665
+ "epoch": 0.75,
666
+ "learning_rate": 4.274660907357009e-05,
667
+ "loss": 1.0799,
668
+ "step": 550
669
+ },
670
+ {
671
+ "epoch": 0.75,
672
+ "learning_rate": 4.262106364437671e-05,
673
+ "loss": 1.2725,
674
+ "step": 555
675
+ },
676
+ {
677
+ "epoch": 0.76,
678
+ "learning_rate": 4.2494628824528184e-05,
679
+ "loss": 1.2027,
680
+ "step": 560
681
+ },
682
+ {
683
+ "epoch": 0.77,
684
+ "learning_rate": 4.2367310995588585e-05,
685
+ "loss": 1.3018,
686
+ "step": 565
687
+ },
688
+ {
689
+ "epoch": 0.77,
690
+ "learning_rate": 4.223911658369027e-05,
691
+ "loss": 1.2029,
692
+ "step": 570
693
+ },
694
+ {
695
+ "epoch": 0.78,
696
+ "learning_rate": 4.211005205920953e-05,
697
+ "loss": 1.2697,
698
+ "step": 575
699
+ },
700
+ {
701
+ "epoch": 0.79,
702
+ "learning_rate": 4.198012393643994e-05,
703
+ "loss": 1.2548,
704
+ "step": 580
705
+ },
706
+ {
707
+ "epoch": 0.79,
708
+ "learning_rate": 4.1849338773263625e-05,
709
+ "loss": 1.2451,
710
+ "step": 585
711
+ },
712
+ {
713
+ "epoch": 0.8,
714
+ "learning_rate": 4.171770317082024e-05,
715
+ "loss": 1.2411,
716
+ "step": 590
717
+ },
718
+ {
719
+ "epoch": 0.81,
720
+ "learning_rate": 4.158522377317379e-05,
721
+ "loss": 1.2933,
722
+ "step": 595
723
+ },
724
+ {
725
+ "epoch": 0.81,
726
+ "learning_rate": 4.1451907266977285e-05,
727
+ "loss": 1.2684,
728
+ "step": 600
729
+ },
730
+ {
731
+ "epoch": 0.82,
732
+ "learning_rate": 4.131776038113524e-05,
733
+ "loss": 1.3233,
734
+ "step": 605
735
+ },
736
+ {
737
+ "epoch": 0.83,
738
+ "learning_rate": 4.118278988646407e-05,
739
+ "loss": 1.1725,
740
+ "step": 610
741
+ },
742
+ {
743
+ "epoch": 0.83,
744
+ "learning_rate": 4.104700259535029e-05,
745
+ "loss": 1.1811,
746
+ "step": 615
747
+ },
748
+ {
749
+ "epoch": 0.84,
750
+ "learning_rate": 4.091040536140673e-05,
751
+ "loss": 1.1264,
752
+ "step": 620
753
+ },
754
+ {
755
+ "epoch": 0.85,
756
+ "learning_rate": 4.07730050791266e-05,
757
+ "loss": 1.2772,
758
+ "step": 625
759
+ },
760
+ {
761
+ "epoch": 0.85,
762
+ "learning_rate": 4.063480868353546e-05,
763
+ "loss": 1.2536,
764
+ "step": 630
765
+ },
766
+ {
767
+ "epoch": 0.86,
768
+ "learning_rate": 4.049582314984124e-05,
769
+ "loss": 1.2617,
770
+ "step": 635
771
+ },
772
+ {
773
+ "epoch": 0.87,
774
+ "learning_rate": 4.0356055493082166e-05,
775
+ "loss": 1.2038,
776
+ "step": 640
777
+ },
778
+ {
779
+ "epoch": 0.87,
780
+ "learning_rate": 4.021551276777266e-05,
781
+ "loss": 1.2328,
782
+ "step": 645
783
+ },
784
+ {
785
+ "epoch": 0.88,
786
+ "learning_rate": 4.007420206754732e-05,
787
+ "loss": 1.244,
788
+ "step": 650
789
+ },
790
+ {
791
+ "epoch": 0.89,
792
+ "learning_rate": 3.993213052480285e-05,
793
+ "loss": 1.1877,
794
+ "step": 655
795
+ },
796
+ {
797
+ "epoch": 0.89,
798
+ "learning_rate": 3.978930531033807e-05,
799
+ "loss": 1.1767,
800
+ "step": 660
801
+ },
802
+ {
803
+ "epoch": 0.9,
804
+ "learning_rate": 3.964573363299198e-05,
805
+ "loss": 1.1808,
806
+ "step": 665
807
+ },
808
+ {
809
+ "epoch": 0.91,
810
+ "learning_rate": 3.9501422739279956e-05,
811
+ "loss": 1.1984,
812
+ "step": 670
813
+ },
814
+ {
815
+ "epoch": 0.92,
816
+ "learning_rate": 3.935637991302792e-05,
817
+ "loss": 1.234,
818
+ "step": 675
819
+ },
820
+ {
821
+ "epoch": 0.92,
822
+ "learning_rate": 3.921061247500474e-05,
823
+ "loss": 1.2115,
824
+ "step": 680
825
+ },
826
+ {
827
+ "epoch": 0.93,
828
+ "learning_rate": 3.906412778255274e-05,
829
+ "loss": 1.2645,
830
+ "step": 685
831
+ },
832
+ {
833
+ "epoch": 0.94,
834
+ "learning_rate": 3.8916933229216315e-05,
835
+ "loss": 1.1742,
836
+ "step": 690
837
+ },
838
+ {
839
+ "epoch": 0.94,
840
+ "learning_rate": 3.8769036244368815e-05,
841
+ "loss": 1.1966,
842
+ "step": 695
843
+ },
844
+ {
845
+ "epoch": 0.95,
846
+ "learning_rate": 3.8620444292837485e-05,
847
+ "loss": 1.1635,
848
+ "step": 700
849
+ },
850
+ {
851
+ "epoch": 0.96,
852
+ "learning_rate": 3.847116487452674e-05,
853
+ "loss": 1.1874,
854
+ "step": 705
855
+ },
856
+ {
857
+ "epoch": 0.96,
858
+ "learning_rate": 3.8321205524039635e-05,
859
+ "loss": 1.166,
860
+ "step": 710
861
+ },
862
+ {
863
+ "epoch": 0.97,
864
+ "learning_rate": 3.817057381029752e-05,
865
+ "loss": 1.1213,
866
+ "step": 715
867
+ },
868
+ {
869
+ "epoch": 0.98,
870
+ "learning_rate": 3.801927733615804e-05,
871
+ "loss": 1.0497,
872
+ "step": 720
873
+ },
874
+ {
875
+ "epoch": 0.98,
876
+ "learning_rate": 3.786732373803141e-05,
877
+ "loss": 1.1248,
878
+ "step": 725
879
+ },
880
+ {
881
+ "epoch": 0.99,
882
+ "learning_rate": 3.771472068549496e-05,
883
+ "loss": 1.2836,
884
+ "step": 730
885
+ },
886
+ {
887
+ "epoch": 1.0,
888
+ "learning_rate": 3.7561475880906026e-05,
889
+ "loss": 1.1569,
890
+ "step": 735
891
+ },
892
+ {
893
+ "epoch": 1.0,
894
+ "learning_rate": 3.7407597059013195e-05,
895
+ "loss": 1.1206,
896
+ "step": 740
897
+ },
898
+ {
899
+ "epoch": 1.01,
900
+ "learning_rate": 3.72530919865659e-05,
901
+ "loss": 1.175,
902
+ "step": 745
903
+ },
904
+ {
905
+ "epoch": 1.02,
906
+ "learning_rate": 3.709796846192243e-05,
907
+ "loss": 1.1392,
908
+ "step": 750
909
+ },
910
+ {
911
+ "epoch": 1.02,
912
+ "learning_rate": 3.694223431465631e-05,
913
+ "loss": 1.1942,
914
+ "step": 755
915
+ },
916
+ {
917
+ "epoch": 1.03,
918
+ "learning_rate": 3.678589740516109e-05,
919
+ "loss": 1.2052,
920
+ "step": 760
921
+ },
922
+ {
923
+ "epoch": 1.04,
924
+ "learning_rate": 3.662896562425364e-05,
925
+ "loss": 1.2162,
926
+ "step": 765
927
+ },
928
+ {
929
+ "epoch": 1.04,
930
+ "learning_rate": 3.6471446892775895e-05,
931
+ "loss": 1.2136,
932
+ "step": 770
933
+ },
934
+ {
935
+ "epoch": 1.05,
936
+ "learning_rate": 3.6313349161194995e-05,
937
+ "loss": 1.146,
938
+ "step": 775
939
+ },
940
+ {
941
+ "epoch": 1.06,
942
+ "learning_rate": 3.6154680409202064e-05,
943
+ "loss": 1.187,
944
+ "step": 780
945
+ },
946
+ {
947
+ "epoch": 1.06,
948
+ "learning_rate": 3.5995448645309424e-05,
949
+ "loss": 1.2493,
950
+ "step": 785
951
+ },
952
+ {
953
+ "epoch": 1.07,
954
+ "learning_rate": 3.583566190644638e-05,
955
+ "loss": 1.2171,
956
+ "step": 790
957
+ },
958
+ {
959
+ "epoch": 1.08,
960
+ "learning_rate": 3.567532825755358e-05,
961
+ "loss": 1.214,
962
+ "step": 795
963
+ },
964
+ {
965
+ "epoch": 1.08,
966
+ "learning_rate": 3.551445579117592e-05,
967
+ "loss": 1.1566,
968
+ "step": 800
969
+ },
970
+ {
971
+ "epoch": 1.09,
972
+ "learning_rate": 3.5353052627054155e-05,
973
+ "loss": 1.1658,
974
+ "step": 805
975
+ },
976
+ {
977
+ "epoch": 1.1,
978
+ "learning_rate": 3.519112691171501e-05,
979
+ "loss": 1.182,
980
+ "step": 810
981
+ },
982
+ {
983
+ "epoch": 1.11,
984
+ "learning_rate": 3.5028686818060016e-05,
985
+ "loss": 1.1485,
986
+ "step": 815
987
+ },
988
+ {
989
+ "epoch": 1.11,
990
+ "learning_rate": 3.4865740544952995e-05,
991
+ "loss": 1.1847,
992
+ "step": 820
993
+ },
994
+ {
995
+ "epoch": 1.12,
996
+ "learning_rate": 3.4702296316806244e-05,
997
+ "loss": 1.134,
998
+ "step": 825
999
+ },
1000
+ {
1001
+ "epoch": 1.13,
1002
+ "learning_rate": 3.4538362383165434e-05,
1003
+ "loss": 1.1939,
1004
+ "step": 830
1005
+ },
1006
+ {
1007
+ "epoch": 1.13,
1008
+ "learning_rate": 3.437394701829321e-05,
1009
+ "loss": 1.2424,
1010
+ "step": 835
1011
+ },
1012
+ {
1013
+ "epoch": 1.14,
1014
+ "learning_rate": 3.420905852075156e-05,
1015
+ "loss": 1.1147,
1016
+ "step": 840
1017
+ },
1018
+ {
1019
+ "epoch": 1.15,
1020
+ "learning_rate": 3.4043705212982987e-05,
1021
+ "loss": 1.1557,
1022
+ "step": 845
1023
+ },
1024
+ {
1025
+ "epoch": 1.15,
1026
+ "learning_rate": 3.38778954408904e-05,
1027
+ "loss": 1.2858,
1028
+ "step": 850
1029
+ },
1030
+ {
1031
+ "epoch": 1.16,
1032
+ "learning_rate": 3.371163757341597e-05,
1033
+ "loss": 1.1724,
1034
+ "step": 855
1035
+ },
1036
+ {
1037
+ "epoch": 1.17,
1038
+ "learning_rate": 3.354494000211856e-05,
1039
+ "loss": 1.2191,
1040
+ "step": 860
1041
+ },
1042
+ {
1043
+ "epoch": 1.17,
1044
+ "learning_rate": 3.337781114075035e-05,
1045
+ "loss": 1.1749,
1046
+ "step": 865
1047
+ },
1048
+ {
1049
+ "epoch": 1.18,
1050
+ "learning_rate": 3.321025942483205e-05,
1051
+ "loss": 1.1776,
1052
+ "step": 870
1053
+ },
1054
+ {
1055
+ "epoch": 1.19,
1056
+ "learning_rate": 3.3042293311227194e-05,
1057
+ "loss": 1.2484,
1058
+ "step": 875
1059
+ },
1060
+ {
1061
+ "epoch": 1.19,
1062
+ "learning_rate": 3.287392127771526e-05,
1063
+ "loss": 1.1541,
1064
+ "step": 880
1065
+ },
1066
+ {
1067
+ "epoch": 1.2,
1068
+ "learning_rate": 3.2705151822563784e-05,
1069
+ "loss": 1.2192,
1070
+ "step": 885
1071
+ },
1072
+ {
1073
+ "epoch": 1.21,
1074
+ "learning_rate": 3.2535993464099444e-05,
1075
+ "loss": 1.1494,
1076
+ "step": 890
1077
+ },
1078
+ {
1079
+ "epoch": 1.21,
1080
+ "learning_rate": 3.236645474027807e-05,
1081
+ "loss": 1.2086,
1082
+ "step": 895
1083
+ },
1084
+ {
1085
+ "epoch": 1.22,
1086
+ "learning_rate": 3.219654420825373e-05,
1087
+ "loss": 1.1576,
1088
+ "step": 900
1089
+ },
1090
+ {
1091
+ "epoch": 1.23,
1092
+ "learning_rate": 3.202627044394686e-05,
1093
+ "loss": 1.1459,
1094
+ "step": 905
1095
+ },
1096
+ {
1097
+ "epoch": 1.23,
1098
+ "learning_rate": 3.185564204161134e-05,
1099
+ "loss": 1.1342,
1100
+ "step": 910
1101
+ },
1102
+ {
1103
+ "epoch": 1.24,
1104
+ "learning_rate": 3.168466761340076e-05,
1105
+ "loss": 1.1235,
1106
+ "step": 915
1107
+ },
1108
+ {
1109
+ "epoch": 1.25,
1110
+ "learning_rate": 3.151335578893374e-05,
1111
+ "loss": 1.1784,
1112
+ "step": 920
1113
+ },
1114
+ {
1115
+ "epoch": 1.25,
1116
+ "learning_rate": 3.1341715214858316e-05,
1117
+ "loss": 1.178,
1118
+ "step": 925
1119
+ },
1120
+ {
1121
+ "epoch": 1.26,
1122
+ "learning_rate": 3.1169754554415605e-05,
1123
+ "loss": 1.1898,
1124
+ "step": 930
1125
+ },
1126
+ {
1127
+ "epoch": 1.27,
1128
+ "learning_rate": 3.099748248700245e-05,
1129
+ "loss": 1.1082,
1130
+ "step": 935
1131
+ },
1132
+ {
1133
+ "epoch": 1.27,
1134
+ "learning_rate": 3.082490770773341e-05,
1135
+ "loss": 1.2071,
1136
+ "step": 940
1137
+ },
1138
+ {
1139
+ "epoch": 1.28,
1140
+ "learning_rate": 3.065203892700186e-05,
1141
+ "loss": 1.1081,
1142
+ "step": 945
1143
+ },
1144
+ {
1145
+ "epoch": 1.29,
1146
+ "learning_rate": 3.0478884870040348e-05,
1147
+ "loss": 1.1357,
1148
+ "step": 950
1149
+ },
1150
+ {
1151
+ "epoch": 1.29,
1152
+ "learning_rate": 3.030545427648022e-05,
1153
+ "loss": 1.1597,
1154
+ "step": 955
1155
+ },
1156
+ {
1157
+ "epoch": 1.3,
1158
+ "learning_rate": 3.0131755899910502e-05,
1159
+ "loss": 1.1475,
1160
+ "step": 960
1161
+ },
1162
+ {
1163
+ "epoch": 1.31,
1164
+ "learning_rate": 2.9957798507436064e-05,
1165
+ "loss": 1.1334,
1166
+ "step": 965
1167
+ },
1168
+ {
1169
+ "epoch": 1.32,
1170
+ "learning_rate": 2.9783590879235124e-05,
1171
+ "loss": 1.169,
1172
+ "step": 970
1173
+ },
1174
+ {
1175
+ "epoch": 1.32,
1176
+ "learning_rate": 2.9609141808116103e-05,
1177
+ "loss": 1.1756,
1178
+ "step": 975
1179
+ },
1180
+ {
1181
+ "epoch": 1.33,
1182
+ "learning_rate": 2.9434460099073773e-05,
1183
+ "loss": 1.1572,
1184
+ "step": 980
1185
+ },
1186
+ {
1187
+ "epoch": 1.34,
1188
+ "learning_rate": 2.9259554568844916e-05,
1189
+ "loss": 1.1201,
1190
+ "step": 985
1191
+ },
1192
+ {
1193
+ "epoch": 1.34,
1194
+ "learning_rate": 2.9084434045463255e-05,
1195
+ "loss": 1.0908,
1196
+ "step": 990
1197
+ },
1198
+ {
1199
+ "epoch": 1.35,
1200
+ "learning_rate": 2.8909107367813893e-05,
1201
+ "loss": 1.1573,
1202
+ "step": 995
1203
+ },
1204
+ {
1205
+ "epoch": 1.36,
1206
+ "learning_rate": 2.8733583385187195e-05,
1207
+ "loss": 1.1656,
1208
+ "step": 1000
1209
+ }
1210
+ ],
1211
+ "max_steps": 2211,
1212
+ "num_train_epochs": 3,
1213
+ "total_flos": 4.81927560910848e+17,
1214
+ "trial_name": null,
1215
+ "trial_params": null
1216
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0090099d32ad29c1bd6d98560579b7c6ef7a6d711729f1926d4fd6f8220f669c
3
+ size 3361
checkpoint-1100/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-1100/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
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": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-1100/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e26f95cea2408d7697dbb1d07117174955fecf8c42a96c0539204b6a87f582a9
3
+ size 16821197
checkpoint-1100/finetuning_args.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dpo_beta": 0.1,
3
+ "finetuning_type": "lora",
4
+ "lora_alpha": 32.0,
5
+ "lora_dropout": 0.1,
6
+ "lora_rank": 8,
7
+ "lora_target": [
8
+ "q_proj",
9
+ "v_proj"
10
+ ],
11
+ "name_module_trainable": "mlp",
12
+ "num_hidden_layers": 32,
13
+ "num_layer_trainable": 3,
14
+ "ppo_score_norm": false,
15
+ "resume_lora_training": true
16
+ }
checkpoint-1100/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c417ef91b2f0d5523f9c1c2ed86f717dc78e0a52c0113797d9e22451dcb2ac89
3
+ size 33629893
checkpoint-1100/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:520767e21a3842691dc578f18f602e55060e97b3ecb76e9ca72cb50909c8f1ea
3
+ size 14575
checkpoint-1100/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:677eca5ad6580d3f557bbfda51765525add80f9acfd6c5ca0edf4a229f0c70eb
3
+ size 557
checkpoint-1100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:beaea01075236d214363b35c1fad1838327577ba8c25a6c6142be5f449bf0533
3
+ size 627
checkpoint-1100/trainer_state.json ADDED
@@ -0,0 +1,1336 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.4915254237288136,
5
+ "global_step": 1100,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.999936908557769e-05,
13
+ "loss": 1.8685,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.999747637415502e-05,
19
+ "loss": 1.698,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.999432196126308e-05,
25
+ "loss": 1.513,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.998990600611505e-05,
31
+ "loss": 1.5436,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 4.998422873159811e-05,
37
+ "loss": 1.5148,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 4.9977290424262214e-05,
43
+ "loss": 1.4708,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.996909143430562e-05,
49
+ "loss": 1.4639,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.05,
54
+ "learning_rate": 4.995963217555719e-05,
55
+ "loss": 1.4528,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.06,
60
+ "learning_rate": 4.994891312545556e-05,
61
+ "loss": 1.5201,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.07,
66
+ "learning_rate": 4.9936934825025e-05,
67
+ "loss": 1.4794,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.07,
72
+ "learning_rate": 4.992369787884809e-05,
73
+ "loss": 1.4549,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.08,
78
+ "learning_rate": 4.9909202955035264e-05,
79
+ "loss": 1.4842,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.09,
84
+ "learning_rate": 4.989345078519104e-05,
85
+ "loss": 1.4832,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.09,
90
+ "learning_rate": 4.9876442164377115e-05,
91
+ "loss": 1.4952,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.1,
96
+ "learning_rate": 4.98581779510722e-05,
97
+ "loss": 1.4078,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.11,
102
+ "learning_rate": 4.9838659067128766e-05,
103
+ "loss": 1.5192,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.12,
108
+ "learning_rate": 4.981788649772644e-05,
109
+ "loss": 1.4805,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.12,
114
+ "learning_rate": 4.97958612913223e-05,
115
+ "loss": 1.4352,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.13,
120
+ "learning_rate": 4.9772584559597976e-05,
121
+ "loss": 1.444,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.14,
126
+ "learning_rate": 4.9748057477403546e-05,
127
+ "loss": 1.4018,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.14,
132
+ "learning_rate": 4.9722281282698186e-05,
133
+ "loss": 1.4472,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.15,
138
+ "learning_rate": 4.9695257276487736e-05,
139
+ "loss": 1.3842,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.16,
144
+ "learning_rate": 4.966698682275903e-05,
145
+ "loss": 1.4116,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.16,
150
+ "learning_rate": 4.9637471348411005e-05,
151
+ "loss": 1.3827,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.17,
156
+ "learning_rate": 4.960671234318276e-05,
157
+ "loss": 1.3718,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.18,
162
+ "learning_rate": 4.957471135957828e-05,
163
+ "loss": 1.3204,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.18,
168
+ "learning_rate": 4.9541470012788135e-05,
169
+ "loss": 1.3307,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 0.19,
174
+ "learning_rate": 4.9506989980607945e-05,
175
+ "loss": 1.3216,
176
+ "step": 140
177
+ },
178
+ {
179
+ "epoch": 0.2,
180
+ "learning_rate": 4.9471273003353655e-05,
181
+ "loss": 1.3268,
182
+ "step": 145
183
+ },
184
+ {
185
+ "epoch": 0.2,
186
+ "learning_rate": 4.943432088377378e-05,
187
+ "loss": 1.4072,
188
+ "step": 150
189
+ },
190
+ {
191
+ "epoch": 0.21,
192
+ "learning_rate": 4.93961354869583e-05,
193
+ "loss": 1.2871,
194
+ "step": 155
195
+ },
196
+ {
197
+ "epoch": 0.22,
198
+ "learning_rate": 4.935671874024465e-05,
199
+ "loss": 1.3451,
200
+ "step": 160
201
+ },
202
+ {
203
+ "epoch": 0.22,
204
+ "learning_rate": 4.931607263312032e-05,
205
+ "loss": 1.3447,
206
+ "step": 165
207
+ },
208
+ {
209
+ "epoch": 0.23,
210
+ "learning_rate": 4.9274199217122565e-05,
211
+ "loss": 1.3573,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 0.24,
216
+ "learning_rate": 4.923110060573472e-05,
217
+ "loss": 1.2914,
218
+ "step": 175
219
+ },
220
+ {
221
+ "epoch": 0.24,
222
+ "learning_rate": 4.9186778974279625e-05,
223
+ "loss": 1.3249,
224
+ "step": 180
225
+ },
226
+ {
227
+ "epoch": 0.25,
228
+ "learning_rate": 4.914123655980981e-05,
229
+ "loss": 1.3253,
230
+ "step": 185
231
+ },
232
+ {
233
+ "epoch": 0.26,
234
+ "learning_rate": 4.909447566099457e-05,
235
+ "loss": 1.2495,
236
+ "step": 190
237
+ },
238
+ {
239
+ "epoch": 0.26,
240
+ "learning_rate": 4.9046498638003925e-05,
241
+ "loss": 1.2333,
242
+ "step": 195
243
+ },
244
+ {
245
+ "epoch": 0.27,
246
+ "learning_rate": 4.8997307912389554e-05,
247
+ "loss": 1.3015,
248
+ "step": 200
249
+ },
250
+ {
251
+ "epoch": 0.28,
252
+ "learning_rate": 4.89469059669625e-05,
253
+ "loss": 1.3331,
254
+ "step": 205
255
+ },
256
+ {
257
+ "epoch": 0.28,
258
+ "learning_rate": 4.889529534566791e-05,
259
+ "loss": 1.2935,
260
+ "step": 210
261
+ },
262
+ {
263
+ "epoch": 0.29,
264
+ "learning_rate": 4.884247865345662e-05,
265
+ "loss": 1.3166,
266
+ "step": 215
267
+ },
268
+ {
269
+ "epoch": 0.3,
270
+ "learning_rate": 4.8788458556153635e-05,
271
+ "loss": 1.3156,
272
+ "step": 220
273
+ },
274
+ {
275
+ "epoch": 0.31,
276
+ "learning_rate": 4.873323778032366e-05,
277
+ "loss": 1.2931,
278
+ "step": 225
279
+ },
280
+ {
281
+ "epoch": 0.31,
282
+ "learning_rate": 4.8676819113133386e-05,
283
+ "loss": 1.3569,
284
+ "step": 230
285
+ },
286
+ {
287
+ "epoch": 0.32,
288
+ "learning_rate": 4.861920540221089e-05,
289
+ "loss": 1.2342,
290
+ "step": 235
291
+ },
292
+ {
293
+ "epoch": 0.33,
294
+ "learning_rate": 4.856039955550186e-05,
295
+ "loss": 1.3015,
296
+ "step": 240
297
+ },
298
+ {
299
+ "epoch": 0.33,
300
+ "learning_rate": 4.8500404541122836e-05,
301
+ "loss": 1.3191,
302
+ "step": 245
303
+ },
304
+ {
305
+ "epoch": 0.34,
306
+ "learning_rate": 4.8439223387211406e-05,
307
+ "loss": 1.3178,
308
+ "step": 250
309
+ },
310
+ {
311
+ "epoch": 0.35,
312
+ "learning_rate": 4.837685918177337e-05,
313
+ "loss": 1.2669,
314
+ "step": 255
315
+ },
316
+ {
317
+ "epoch": 0.35,
318
+ "learning_rate": 4.8313315072526855e-05,
319
+ "loss": 1.2218,
320
+ "step": 260
321
+ },
322
+ {
323
+ "epoch": 0.36,
324
+ "learning_rate": 4.824859426674345e-05,
325
+ "loss": 1.304,
326
+ "step": 265
327
+ },
328
+ {
329
+ "epoch": 0.37,
330
+ "learning_rate": 4.818270003108635e-05,
331
+ "loss": 1.311,
332
+ "step": 270
333
+ },
334
+ {
335
+ "epoch": 0.37,
336
+ "learning_rate": 4.8115635691445436e-05,
337
+ "loss": 1.3036,
338
+ "step": 275
339
+ },
340
+ {
341
+ "epoch": 0.38,
342
+ "learning_rate": 4.8047404632769446e-05,
343
+ "loss": 1.258,
344
+ "step": 280
345
+ },
346
+ {
347
+ "epoch": 0.39,
348
+ "learning_rate": 4.797801029889508e-05,
349
+ "loss": 1.3186,
350
+ "step": 285
351
+ },
352
+ {
353
+ "epoch": 0.39,
354
+ "learning_rate": 4.7907456192373245e-05,
355
+ "loss": 1.395,
356
+ "step": 290
357
+ },
358
+ {
359
+ "epoch": 0.4,
360
+ "learning_rate": 4.783574587429219e-05,
361
+ "loss": 1.2793,
362
+ "step": 295
363
+ },
364
+ {
365
+ "epoch": 0.41,
366
+ "learning_rate": 4.776288296409785e-05,
367
+ "loss": 1.3625,
368
+ "step": 300
369
+ },
370
+ {
371
+ "epoch": 0.41,
372
+ "learning_rate": 4.768887113941107e-05,
373
+ "loss": 1.2453,
374
+ "step": 305
375
+ },
376
+ {
377
+ "epoch": 0.42,
378
+ "learning_rate": 4.761371413584209e-05,
379
+ "loss": 1.3312,
380
+ "step": 310
381
+ },
382
+ {
383
+ "epoch": 0.43,
384
+ "learning_rate": 4.7537415746801876e-05,
385
+ "loss": 1.2627,
386
+ "step": 315
387
+ },
388
+ {
389
+ "epoch": 0.43,
390
+ "learning_rate": 4.745997982331077e-05,
391
+ "loss": 1.2231,
392
+ "step": 320
393
+ },
394
+ {
395
+ "epoch": 0.44,
396
+ "learning_rate": 4.738141027380405e-05,
397
+ "loss": 1.2063,
398
+ "step": 325
399
+ },
400
+ {
401
+ "epoch": 0.45,
402
+ "learning_rate": 4.730171106393466e-05,
403
+ "loss": 1.3148,
404
+ "step": 330
405
+ },
406
+ {
407
+ "epoch": 0.45,
408
+ "learning_rate": 4.722088621637309e-05,
409
+ "loss": 1.2787,
410
+ "step": 335
411
+ },
412
+ {
413
+ "epoch": 0.46,
414
+ "learning_rate": 4.7138939810604286e-05,
415
+ "loss": 1.2111,
416
+ "step": 340
417
+ },
418
+ {
419
+ "epoch": 0.47,
420
+ "learning_rate": 4.7055875982721794e-05,
421
+ "loss": 1.3034,
422
+ "step": 345
423
+ },
424
+ {
425
+ "epoch": 0.47,
426
+ "learning_rate": 4.697169892521898e-05,
427
+ "loss": 1.288,
428
+ "step": 350
429
+ },
430
+ {
431
+ "epoch": 0.48,
432
+ "learning_rate": 4.688641288677739e-05,
433
+ "loss": 1.2935,
434
+ "step": 355
435
+ },
436
+ {
437
+ "epoch": 0.49,
438
+ "learning_rate": 4.680002217205238e-05,
439
+ "loss": 1.3183,
440
+ "step": 360
441
+ },
442
+ {
443
+ "epoch": 0.49,
444
+ "learning_rate": 4.6712531141455785e-05,
445
+ "loss": 1.1688,
446
+ "step": 365
447
+ },
448
+ {
449
+ "epoch": 0.5,
450
+ "learning_rate": 4.662394421093582e-05,
451
+ "loss": 1.2734,
452
+ "step": 370
453
+ },
454
+ {
455
+ "epoch": 0.51,
456
+ "learning_rate": 4.653426585175428e-05,
457
+ "loss": 1.2155,
458
+ "step": 375
459
+ },
460
+ {
461
+ "epoch": 0.52,
462
+ "learning_rate": 4.644350059026076e-05,
463
+ "loss": 1.2641,
464
+ "step": 380
465
+ },
466
+ {
467
+ "epoch": 0.52,
468
+ "learning_rate": 4.6351653007664276e-05,
469
+ "loss": 1.2819,
470
+ "step": 385
471
+ },
472
+ {
473
+ "epoch": 0.53,
474
+ "learning_rate": 4.625872773980197e-05,
475
+ "loss": 1.2435,
476
+ "step": 390
477
+ },
478
+ {
479
+ "epoch": 0.54,
480
+ "learning_rate": 4.61647294769052e-05,
481
+ "loss": 1.2541,
482
+ "step": 395
483
+ },
484
+ {
485
+ "epoch": 0.54,
486
+ "learning_rate": 4.6069662963362724e-05,
487
+ "loss": 1.2464,
488
+ "step": 400
489
+ },
490
+ {
491
+ "epoch": 0.55,
492
+ "learning_rate": 4.597353299748132e-05,
493
+ "loss": 1.2843,
494
+ "step": 405
495
+ },
496
+ {
497
+ "epoch": 0.56,
498
+ "learning_rate": 4.587634443124352e-05,
499
+ "loss": 1.3026,
500
+ "step": 410
501
+ },
502
+ {
503
+ "epoch": 0.56,
504
+ "learning_rate": 4.577810217006279e-05,
505
+ "loss": 1.2905,
506
+ "step": 415
507
+ },
508
+ {
509
+ "epoch": 0.57,
510
+ "learning_rate": 4.567881117253587e-05,
511
+ "loss": 1.2689,
512
+ "step": 420
513
+ },
514
+ {
515
+ "epoch": 0.58,
516
+ "learning_rate": 4.557847645019257e-05,
517
+ "loss": 1.2959,
518
+ "step": 425
519
+ },
520
+ {
521
+ "epoch": 0.58,
522
+ "learning_rate": 4.547710306724274e-05,
523
+ "loss": 1.2237,
524
+ "step": 430
525
+ },
526
+ {
527
+ "epoch": 0.59,
528
+ "learning_rate": 4.537469614032074e-05,
529
+ "loss": 1.2885,
530
+ "step": 435
531
+ },
532
+ {
533
+ "epoch": 0.6,
534
+ "learning_rate": 4.5271260838227124e-05,
535
+ "loss": 1.2622,
536
+ "step": 440
537
+ },
538
+ {
539
+ "epoch": 0.6,
540
+ "learning_rate": 4.516680238166783e-05,
541
+ "loss": 1.2567,
542
+ "step": 445
543
+ },
544
+ {
545
+ "epoch": 0.61,
546
+ "learning_rate": 4.506132604299057e-05,
547
+ "loss": 1.2618,
548
+ "step": 450
549
+ },
550
+ {
551
+ "epoch": 0.62,
552
+ "learning_rate": 4.4954837145918826e-05,
553
+ "loss": 1.249,
554
+ "step": 455
555
+ },
556
+ {
557
+ "epoch": 0.62,
558
+ "learning_rate": 4.484734106528307e-05,
559
+ "loss": 1.2891,
560
+ "step": 460
561
+ },
562
+ {
563
+ "epoch": 0.63,
564
+ "learning_rate": 4.473884322674952e-05,
565
+ "loss": 1.2423,
566
+ "step": 465
567
+ },
568
+ {
569
+ "epoch": 0.64,
570
+ "learning_rate": 4.462934910654624e-05,
571
+ "loss": 1.1933,
572
+ "step": 470
573
+ },
574
+ {
575
+ "epoch": 0.64,
576
+ "learning_rate": 4.451886423118682e-05,
577
+ "loss": 1.1707,
578
+ "step": 475
579
+ },
580
+ {
581
+ "epoch": 0.65,
582
+ "learning_rate": 4.4407394177191366e-05,
583
+ "loss": 1.2059,
584
+ "step": 480
585
+ },
586
+ {
587
+ "epoch": 0.66,
588
+ "learning_rate": 4.429494457080505e-05,
589
+ "loss": 1.2385,
590
+ "step": 485
591
+ },
592
+ {
593
+ "epoch": 0.66,
594
+ "learning_rate": 4.418152108771414e-05,
595
+ "loss": 1.2545,
596
+ "step": 490
597
+ },
598
+ {
599
+ "epoch": 0.67,
600
+ "learning_rate": 4.406712945275955e-05,
601
+ "loss": 1.2488,
602
+ "step": 495
603
+ },
604
+ {
605
+ "epoch": 0.68,
606
+ "learning_rate": 4.395177543964787e-05,
607
+ "loss": 1.2613,
608
+ "step": 500
609
+ },
610
+ {
611
+ "epoch": 0.68,
612
+ "learning_rate": 4.383546487065993e-05,
613
+ "loss": 1.1931,
614
+ "step": 505
615
+ },
616
+ {
617
+ "epoch": 0.69,
618
+ "learning_rate": 4.3718203616356954e-05,
619
+ "loss": 1.231,
620
+ "step": 510
621
+ },
622
+ {
623
+ "epoch": 0.7,
624
+ "learning_rate": 4.35999975952843e-05,
625
+ "loss": 1.1383,
626
+ "step": 515
627
+ },
628
+ {
629
+ "epoch": 0.71,
630
+ "learning_rate": 4.348085277367262e-05,
631
+ "loss": 1.1776,
632
+ "step": 520
633
+ },
634
+ {
635
+ "epoch": 0.71,
636
+ "learning_rate": 4.336077516513682e-05,
637
+ "loss": 1.2254,
638
+ "step": 525
639
+ },
640
+ {
641
+ "epoch": 0.72,
642
+ "learning_rate": 4.323977083037252e-05,
643
+ "loss": 1.2157,
644
+ "step": 530
645
+ },
646
+ {
647
+ "epoch": 0.73,
648
+ "learning_rate": 4.3117845876850085e-05,
649
+ "loss": 1.1688,
650
+ "step": 535
651
+ },
652
+ {
653
+ "epoch": 0.73,
654
+ "learning_rate": 4.299500645850647e-05,
655
+ "loss": 1.2387,
656
+ "step": 540
657
+ },
658
+ {
659
+ "epoch": 0.74,
660
+ "learning_rate": 4.287125877543453e-05,
661
+ "loss": 1.2281,
662
+ "step": 545
663
+ },
664
+ {
665
+ "epoch": 0.75,
666
+ "learning_rate": 4.274660907357009e-05,
667
+ "loss": 1.0799,
668
+ "step": 550
669
+ },
670
+ {
671
+ "epoch": 0.75,
672
+ "learning_rate": 4.262106364437671e-05,
673
+ "loss": 1.2725,
674
+ "step": 555
675
+ },
676
+ {
677
+ "epoch": 0.76,
678
+ "learning_rate": 4.2494628824528184e-05,
679
+ "loss": 1.2027,
680
+ "step": 560
681
+ },
682
+ {
683
+ "epoch": 0.77,
684
+ "learning_rate": 4.2367310995588585e-05,
685
+ "loss": 1.3018,
686
+ "step": 565
687
+ },
688
+ {
689
+ "epoch": 0.77,
690
+ "learning_rate": 4.223911658369027e-05,
691
+ "loss": 1.2029,
692
+ "step": 570
693
+ },
694
+ {
695
+ "epoch": 0.78,
696
+ "learning_rate": 4.211005205920953e-05,
697
+ "loss": 1.2697,
698
+ "step": 575
699
+ },
700
+ {
701
+ "epoch": 0.79,
702
+ "learning_rate": 4.198012393643994e-05,
703
+ "loss": 1.2548,
704
+ "step": 580
705
+ },
706
+ {
707
+ "epoch": 0.79,
708
+ "learning_rate": 4.1849338773263625e-05,
709
+ "loss": 1.2451,
710
+ "step": 585
711
+ },
712
+ {
713
+ "epoch": 0.8,
714
+ "learning_rate": 4.171770317082024e-05,
715
+ "loss": 1.2411,
716
+ "step": 590
717
+ },
718
+ {
719
+ "epoch": 0.81,
720
+ "learning_rate": 4.158522377317379e-05,
721
+ "loss": 1.2933,
722
+ "step": 595
723
+ },
724
+ {
725
+ "epoch": 0.81,
726
+ "learning_rate": 4.1451907266977285e-05,
727
+ "loss": 1.2684,
728
+ "step": 600
729
+ },
730
+ {
731
+ "epoch": 0.82,
732
+ "learning_rate": 4.131776038113524e-05,
733
+ "loss": 1.3233,
734
+ "step": 605
735
+ },
736
+ {
737
+ "epoch": 0.83,
738
+ "learning_rate": 4.118278988646407e-05,
739
+ "loss": 1.1725,
740
+ "step": 610
741
+ },
742
+ {
743
+ "epoch": 0.83,
744
+ "learning_rate": 4.104700259535029e-05,
745
+ "loss": 1.1811,
746
+ "step": 615
747
+ },
748
+ {
749
+ "epoch": 0.84,
750
+ "learning_rate": 4.091040536140673e-05,
751
+ "loss": 1.1264,
752
+ "step": 620
753
+ },
754
+ {
755
+ "epoch": 0.85,
756
+ "learning_rate": 4.07730050791266e-05,
757
+ "loss": 1.2772,
758
+ "step": 625
759
+ },
760
+ {
761
+ "epoch": 0.85,
762
+ "learning_rate": 4.063480868353546e-05,
763
+ "loss": 1.2536,
764
+ "step": 630
765
+ },
766
+ {
767
+ "epoch": 0.86,
768
+ "learning_rate": 4.049582314984124e-05,
769
+ "loss": 1.2617,
770
+ "step": 635
771
+ },
772
+ {
773
+ "epoch": 0.87,
774
+ "learning_rate": 4.0356055493082166e-05,
775
+ "loss": 1.2038,
776
+ "step": 640
777
+ },
778
+ {
779
+ "epoch": 0.87,
780
+ "learning_rate": 4.021551276777266e-05,
781
+ "loss": 1.2328,
782
+ "step": 645
783
+ },
784
+ {
785
+ "epoch": 0.88,
786
+ "learning_rate": 4.007420206754732e-05,
787
+ "loss": 1.244,
788
+ "step": 650
789
+ },
790
+ {
791
+ "epoch": 0.89,
792
+ "learning_rate": 3.993213052480285e-05,
793
+ "loss": 1.1877,
794
+ "step": 655
795
+ },
796
+ {
797
+ "epoch": 0.89,
798
+ "learning_rate": 3.978930531033807e-05,
799
+ "loss": 1.1767,
800
+ "step": 660
801
+ },
802
+ {
803
+ "epoch": 0.9,
804
+ "learning_rate": 3.964573363299198e-05,
805
+ "loss": 1.1808,
806
+ "step": 665
807
+ },
808
+ {
809
+ "epoch": 0.91,
810
+ "learning_rate": 3.9501422739279956e-05,
811
+ "loss": 1.1984,
812
+ "step": 670
813
+ },
814
+ {
815
+ "epoch": 0.92,
816
+ "learning_rate": 3.935637991302792e-05,
817
+ "loss": 1.234,
818
+ "step": 675
819
+ },
820
+ {
821
+ "epoch": 0.92,
822
+ "learning_rate": 3.921061247500474e-05,
823
+ "loss": 1.2115,
824
+ "step": 680
825
+ },
826
+ {
827
+ "epoch": 0.93,
828
+ "learning_rate": 3.906412778255274e-05,
829
+ "loss": 1.2645,
830
+ "step": 685
831
+ },
832
+ {
833
+ "epoch": 0.94,
834
+ "learning_rate": 3.8916933229216315e-05,
835
+ "loss": 1.1742,
836
+ "step": 690
837
+ },
838
+ {
839
+ "epoch": 0.94,
840
+ "learning_rate": 3.8769036244368815e-05,
841
+ "loss": 1.1966,
842
+ "step": 695
843
+ },
844
+ {
845
+ "epoch": 0.95,
846
+ "learning_rate": 3.8620444292837485e-05,
847
+ "loss": 1.1635,
848
+ "step": 700
849
+ },
850
+ {
851
+ "epoch": 0.96,
852
+ "learning_rate": 3.847116487452674e-05,
853
+ "loss": 1.1874,
854
+ "step": 705
855
+ },
856
+ {
857
+ "epoch": 0.96,
858
+ "learning_rate": 3.8321205524039635e-05,
859
+ "loss": 1.166,
860
+ "step": 710
861
+ },
862
+ {
863
+ "epoch": 0.97,
864
+ "learning_rate": 3.817057381029752e-05,
865
+ "loss": 1.1213,
866
+ "step": 715
867
+ },
868
+ {
869
+ "epoch": 0.98,
870
+ "learning_rate": 3.801927733615804e-05,
871
+ "loss": 1.0497,
872
+ "step": 720
873
+ },
874
+ {
875
+ "epoch": 0.98,
876
+ "learning_rate": 3.786732373803141e-05,
877
+ "loss": 1.1248,
878
+ "step": 725
879
+ },
880
+ {
881
+ "epoch": 0.99,
882
+ "learning_rate": 3.771472068549496e-05,
883
+ "loss": 1.2836,
884
+ "step": 730
885
+ },
886
+ {
887
+ "epoch": 1.0,
888
+ "learning_rate": 3.7561475880906026e-05,
889
+ "loss": 1.1569,
890
+ "step": 735
891
+ },
892
+ {
893
+ "epoch": 1.0,
894
+ "learning_rate": 3.7407597059013195e-05,
895
+ "loss": 1.1206,
896
+ "step": 740
897
+ },
898
+ {
899
+ "epoch": 1.01,
900
+ "learning_rate": 3.72530919865659e-05,
901
+ "loss": 1.175,
902
+ "step": 745
903
+ },
904
+ {
905
+ "epoch": 1.02,
906
+ "learning_rate": 3.709796846192243e-05,
907
+ "loss": 1.1392,
908
+ "step": 750
909
+ },
910
+ {
911
+ "epoch": 1.02,
912
+ "learning_rate": 3.694223431465631e-05,
913
+ "loss": 1.1942,
914
+ "step": 755
915
+ },
916
+ {
917
+ "epoch": 1.03,
918
+ "learning_rate": 3.678589740516109e-05,
919
+ "loss": 1.2052,
920
+ "step": 760
921
+ },
922
+ {
923
+ "epoch": 1.04,
924
+ "learning_rate": 3.662896562425364e-05,
925
+ "loss": 1.2162,
926
+ "step": 765
927
+ },
928
+ {
929
+ "epoch": 1.04,
930
+ "learning_rate": 3.6471446892775895e-05,
931
+ "loss": 1.2136,
932
+ "step": 770
933
+ },
934
+ {
935
+ "epoch": 1.05,
936
+ "learning_rate": 3.6313349161194995e-05,
937
+ "loss": 1.146,
938
+ "step": 775
939
+ },
940
+ {
941
+ "epoch": 1.06,
942
+ "learning_rate": 3.6154680409202064e-05,
943
+ "loss": 1.187,
944
+ "step": 780
945
+ },
946
+ {
947
+ "epoch": 1.06,
948
+ "learning_rate": 3.5995448645309424e-05,
949
+ "loss": 1.2493,
950
+ "step": 785
951
+ },
952
+ {
953
+ "epoch": 1.07,
954
+ "learning_rate": 3.583566190644638e-05,
955
+ "loss": 1.2171,
956
+ "step": 790
957
+ },
958
+ {
959
+ "epoch": 1.08,
960
+ "learning_rate": 3.567532825755358e-05,
961
+ "loss": 1.214,
962
+ "step": 795
963
+ },
964
+ {
965
+ "epoch": 1.08,
966
+ "learning_rate": 3.551445579117592e-05,
967
+ "loss": 1.1566,
968
+ "step": 800
969
+ },
970
+ {
971
+ "epoch": 1.09,
972
+ "learning_rate": 3.5353052627054155e-05,
973
+ "loss": 1.1658,
974
+ "step": 805
975
+ },
976
+ {
977
+ "epoch": 1.1,
978
+ "learning_rate": 3.519112691171501e-05,
979
+ "loss": 1.182,
980
+ "step": 810
981
+ },
982
+ {
983
+ "epoch": 1.11,
984
+ "learning_rate": 3.5028686818060016e-05,
985
+ "loss": 1.1485,
986
+ "step": 815
987
+ },
988
+ {
989
+ "epoch": 1.11,
990
+ "learning_rate": 3.4865740544952995e-05,
991
+ "loss": 1.1847,
992
+ "step": 820
993
+ },
994
+ {
995
+ "epoch": 1.12,
996
+ "learning_rate": 3.4702296316806244e-05,
997
+ "loss": 1.134,
998
+ "step": 825
999
+ },
1000
+ {
1001
+ "epoch": 1.13,
1002
+ "learning_rate": 3.4538362383165434e-05,
1003
+ "loss": 1.1939,
1004
+ "step": 830
1005
+ },
1006
+ {
1007
+ "epoch": 1.13,
1008
+ "learning_rate": 3.437394701829321e-05,
1009
+ "loss": 1.2424,
1010
+ "step": 835
1011
+ },
1012
+ {
1013
+ "epoch": 1.14,
1014
+ "learning_rate": 3.420905852075156e-05,
1015
+ "loss": 1.1147,
1016
+ "step": 840
1017
+ },
1018
+ {
1019
+ "epoch": 1.15,
1020
+ "learning_rate": 3.4043705212982987e-05,
1021
+ "loss": 1.1557,
1022
+ "step": 845
1023
+ },
1024
+ {
1025
+ "epoch": 1.15,
1026
+ "learning_rate": 3.38778954408904e-05,
1027
+ "loss": 1.2858,
1028
+ "step": 850
1029
+ },
1030
+ {
1031
+ "epoch": 1.16,
1032
+ "learning_rate": 3.371163757341597e-05,
1033
+ "loss": 1.1724,
1034
+ "step": 855
1035
+ },
1036
+ {
1037
+ "epoch": 1.17,
1038
+ "learning_rate": 3.354494000211856e-05,
1039
+ "loss": 1.2191,
1040
+ "step": 860
1041
+ },
1042
+ {
1043
+ "epoch": 1.17,
1044
+ "learning_rate": 3.337781114075035e-05,
1045
+ "loss": 1.1749,
1046
+ "step": 865
1047
+ },
1048
+ {
1049
+ "epoch": 1.18,
1050
+ "learning_rate": 3.321025942483205e-05,
1051
+ "loss": 1.1776,
1052
+ "step": 870
1053
+ },
1054
+ {
1055
+ "epoch": 1.19,
1056
+ "learning_rate": 3.3042293311227194e-05,
1057
+ "loss": 1.2484,
1058
+ "step": 875
1059
+ },
1060
+ {
1061
+ "epoch": 1.19,
1062
+ "learning_rate": 3.287392127771526e-05,
1063
+ "loss": 1.1541,
1064
+ "step": 880
1065
+ },
1066
+ {
1067
+ "epoch": 1.2,
1068
+ "learning_rate": 3.2705151822563784e-05,
1069
+ "loss": 1.2192,
1070
+ "step": 885
1071
+ },
1072
+ {
1073
+ "epoch": 1.21,
1074
+ "learning_rate": 3.2535993464099444e-05,
1075
+ "loss": 1.1494,
1076
+ "step": 890
1077
+ },
1078
+ {
1079
+ "epoch": 1.21,
1080
+ "learning_rate": 3.236645474027807e-05,
1081
+ "loss": 1.2086,
1082
+ "step": 895
1083
+ },
1084
+ {
1085
+ "epoch": 1.22,
1086
+ "learning_rate": 3.219654420825373e-05,
1087
+ "loss": 1.1576,
1088
+ "step": 900
1089
+ },
1090
+ {
1091
+ "epoch": 1.23,
1092
+ "learning_rate": 3.202627044394686e-05,
1093
+ "loss": 1.1459,
1094
+ "step": 905
1095
+ },
1096
+ {
1097
+ "epoch": 1.23,
1098
+ "learning_rate": 3.185564204161134e-05,
1099
+ "loss": 1.1342,
1100
+ "step": 910
1101
+ },
1102
+ {
1103
+ "epoch": 1.24,
1104
+ "learning_rate": 3.168466761340076e-05,
1105
+ "loss": 1.1235,
1106
+ "step": 915
1107
+ },
1108
+ {
1109
+ "epoch": 1.25,
1110
+ "learning_rate": 3.151335578893374e-05,
1111
+ "loss": 1.1784,
1112
+ "step": 920
1113
+ },
1114
+ {
1115
+ "epoch": 1.25,
1116
+ "learning_rate": 3.1341715214858316e-05,
1117
+ "loss": 1.178,
1118
+ "step": 925
1119
+ },
1120
+ {
1121
+ "epoch": 1.26,
1122
+ "learning_rate": 3.1169754554415605e-05,
1123
+ "loss": 1.1898,
1124
+ "step": 930
1125
+ },
1126
+ {
1127
+ "epoch": 1.27,
1128
+ "learning_rate": 3.099748248700245e-05,
1129
+ "loss": 1.1082,
1130
+ "step": 935
1131
+ },
1132
+ {
1133
+ "epoch": 1.27,
1134
+ "learning_rate": 3.082490770773341e-05,
1135
+ "loss": 1.2071,
1136
+ "step": 940
1137
+ },
1138
+ {
1139
+ "epoch": 1.28,
1140
+ "learning_rate": 3.065203892700186e-05,
1141
+ "loss": 1.1081,
1142
+ "step": 945
1143
+ },
1144
+ {
1145
+ "epoch": 1.29,
1146
+ "learning_rate": 3.0478884870040348e-05,
1147
+ "loss": 1.1357,
1148
+ "step": 950
1149
+ },
1150
+ {
1151
+ "epoch": 1.29,
1152
+ "learning_rate": 3.030545427648022e-05,
1153
+ "loss": 1.1597,
1154
+ "step": 955
1155
+ },
1156
+ {
1157
+ "epoch": 1.3,
1158
+ "learning_rate": 3.0131755899910502e-05,
1159
+ "loss": 1.1475,
1160
+ "step": 960
1161
+ },
1162
+ {
1163
+ "epoch": 1.31,
1164
+ "learning_rate": 2.9957798507436064e-05,
1165
+ "loss": 1.1334,
1166
+ "step": 965
1167
+ },
1168
+ {
1169
+ "epoch": 1.32,
1170
+ "learning_rate": 2.9783590879235124e-05,
1171
+ "loss": 1.169,
1172
+ "step": 970
1173
+ },
1174
+ {
1175
+ "epoch": 1.32,
1176
+ "learning_rate": 2.9609141808116103e-05,
1177
+ "loss": 1.1756,
1178
+ "step": 975
1179
+ },
1180
+ {
1181
+ "epoch": 1.33,
1182
+ "learning_rate": 2.9434460099073773e-05,
1183
+ "loss": 1.1572,
1184
+ "step": 980
1185
+ },
1186
+ {
1187
+ "epoch": 1.34,
1188
+ "learning_rate": 2.9259554568844916e-05,
1189
+ "loss": 1.1201,
1190
+ "step": 985
1191
+ },
1192
+ {
1193
+ "epoch": 1.34,
1194
+ "learning_rate": 2.9084434045463255e-05,
1195
+ "loss": 1.0908,
1196
+ "step": 990
1197
+ },
1198
+ {
1199
+ "epoch": 1.35,
1200
+ "learning_rate": 2.8909107367813893e-05,
1201
+ "loss": 1.1573,
1202
+ "step": 995
1203
+ },
1204
+ {
1205
+ "epoch": 1.36,
1206
+ "learning_rate": 2.8733583385187195e-05,
1207
+ "loss": 1.1656,
1208
+ "step": 1000
1209
+ },
1210
+ {
1211
+ "epoch": 1.36,
1212
+ "learning_rate": 2.8557870956832132e-05,
1213
+ "loss": 1.1984,
1214
+ "step": 1005
1215
+ },
1216
+ {
1217
+ "epoch": 1.37,
1218
+ "learning_rate": 2.8381978951509115e-05,
1219
+ "loss": 1.1605,
1220
+ "step": 1010
1221
+ },
1222
+ {
1223
+ "epoch": 1.38,
1224
+ "learning_rate": 2.8205916247042384e-05,
1225
+ "loss": 1.1765,
1226
+ "step": 1015
1227
+ },
1228
+ {
1229
+ "epoch": 1.38,
1230
+ "learning_rate": 2.8029691729871898e-05,
1231
+ "loss": 1.1306,
1232
+ "step": 1020
1233
+ },
1234
+ {
1235
+ "epoch": 1.39,
1236
+ "learning_rate": 2.7853314294604803e-05,
1237
+ "loss": 1.1954,
1238
+ "step": 1025
1239
+ },
1240
+ {
1241
+ "epoch": 1.4,
1242
+ "learning_rate": 2.767679284356653e-05,
1243
+ "loss": 1.1751,
1244
+ "step": 1030
1245
+ },
1246
+ {
1247
+ "epoch": 1.4,
1248
+ "learning_rate": 2.7500136286351407e-05,
1249
+ "loss": 1.0568,
1250
+ "step": 1035
1251
+ },
1252
+ {
1253
+ "epoch": 1.41,
1254
+ "learning_rate": 2.732335353937302e-05,
1255
+ "loss": 1.1568,
1256
+ "step": 1040
1257
+ },
1258
+ {
1259
+ "epoch": 1.42,
1260
+ "learning_rate": 2.714645352541415e-05,
1261
+ "loss": 1.1541,
1262
+ "step": 1045
1263
+ },
1264
+ {
1265
+ "epoch": 1.42,
1266
+ "learning_rate": 2.69694451731764e-05,
1267
+ "loss": 1.1995,
1268
+ "step": 1050
1269
+ },
1270
+ {
1271
+ "epoch": 1.43,
1272
+ "learning_rate": 2.6792337416829554e-05,
1273
+ "loss": 1.1831,
1274
+ "step": 1055
1275
+ },
1276
+ {
1277
+ "epoch": 1.44,
1278
+ "learning_rate": 2.6615139195560635e-05,
1279
+ "loss": 1.1272,
1280
+ "step": 1060
1281
+ },
1282
+ {
1283
+ "epoch": 1.44,
1284
+ "learning_rate": 2.643785945312271e-05,
1285
+ "loss": 1.1833,
1286
+ "step": 1065
1287
+ },
1288
+ {
1289
+ "epoch": 1.45,
1290
+ "learning_rate": 2.626050713738349e-05,
1291
+ "loss": 1.1971,
1292
+ "step": 1070
1293
+ },
1294
+ {
1295
+ "epoch": 1.46,
1296
+ "learning_rate": 2.6083091199873677e-05,
1297
+ "loss": 1.1645,
1298
+ "step": 1075
1299
+ },
1300
+ {
1301
+ "epoch": 1.46,
1302
+ "learning_rate": 2.5905620595335172e-05,
1303
+ "loss": 1.1932,
1304
+ "step": 1080
1305
+ },
1306
+ {
1307
+ "epoch": 1.47,
1308
+ "learning_rate": 2.5728104281269087e-05,
1309
+ "loss": 1.1351,
1310
+ "step": 1085
1311
+ },
1312
+ {
1313
+ "epoch": 1.48,
1314
+ "learning_rate": 2.5550551217483642e-05,
1315
+ "loss": 1.0708,
1316
+ "step": 1090
1317
+ },
1318
+ {
1319
+ "epoch": 1.48,
1320
+ "learning_rate": 2.5372970365641928e-05,
1321
+ "loss": 1.1435,
1322
+ "step": 1095
1323
+ },
1324
+ {
1325
+ "epoch": 1.49,
1326
+ "learning_rate": 2.51953706888096e-05,
1327
+ "loss": 1.1483,
1328
+ "step": 1100
1329
+ }
1330
+ ],
1331
+ "max_steps": 2211,
1332
+ "num_train_epochs": 3,
1333
+ "total_flos": 5.303635290358579e+17,
1334
+ "trial_name": null,
1335
+ "trial_params": null
1336
+ }
checkpoint-1100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0090099d32ad29c1bd6d98560579b7c6ef7a6d711729f1926d4fd6f8220f669c
3
+ size 3361
checkpoint-1200/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-1200/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
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": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-1200/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c26573c2cdb9d165820e229615cf0542b148f3a767593f8853f2005c1f3efb9e
3
+ size 16821197
checkpoint-1200/finetuning_args.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dpo_beta": 0.1,
3
+ "finetuning_type": "lora",
4
+ "lora_alpha": 32.0,
5
+ "lora_dropout": 0.1,
6
+ "lora_rank": 8,
7
+ "lora_target": [
8
+ "q_proj",
9
+ "v_proj"
10
+ ],
11
+ "name_module_trainable": "mlp",
12
+ "num_hidden_layers": 32,
13
+ "num_layer_trainable": 3,
14
+ "ppo_score_norm": false,
15
+ "resume_lora_training": true
16
+ }
checkpoint-1200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a556285ede7b9aca8b4258224f67f71ba7ad5c6ac924d8f0495d9294fbdcfb16
3
+ size 33629893
checkpoint-1200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da4482f7a209e2a9fac2a5c900344827a341854ba81b902dae36d41b315e7c06
3
+ size 14575
checkpoint-1200/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:741cefeca9ef427f92406d2d10b81996655e2a9d50eb7aaa9614e6fdd1c9f529
3
+ size 557
checkpoint-1200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c871ee5de85294900a35e52c062fb81555f18a54f76532dcab4afb86eb783292
3
+ size 627
checkpoint-1200/trainer_state.json ADDED
@@ -0,0 +1,1456 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.6271186440677967,
5
+ "global_step": 1200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.999936908557769e-05,
13
+ "loss": 1.8685,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.999747637415502e-05,
19
+ "loss": 1.698,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.999432196126308e-05,
25
+ "loss": 1.513,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.998990600611505e-05,
31
+ "loss": 1.5436,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 4.998422873159811e-05,
37
+ "loss": 1.5148,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 4.9977290424262214e-05,
43
+ "loss": 1.4708,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.996909143430562e-05,
49
+ "loss": 1.4639,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.05,
54
+ "learning_rate": 4.995963217555719e-05,
55
+ "loss": 1.4528,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.06,
60
+ "learning_rate": 4.994891312545556e-05,
61
+ "loss": 1.5201,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.07,
66
+ "learning_rate": 4.9936934825025e-05,
67
+ "loss": 1.4794,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.07,
72
+ "learning_rate": 4.992369787884809e-05,
73
+ "loss": 1.4549,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.08,
78
+ "learning_rate": 4.9909202955035264e-05,
79
+ "loss": 1.4842,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.09,
84
+ "learning_rate": 4.989345078519104e-05,
85
+ "loss": 1.4832,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.09,
90
+ "learning_rate": 4.9876442164377115e-05,
91
+ "loss": 1.4952,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.1,
96
+ "learning_rate": 4.98581779510722e-05,
97
+ "loss": 1.4078,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.11,
102
+ "learning_rate": 4.9838659067128766e-05,
103
+ "loss": 1.5192,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.12,
108
+ "learning_rate": 4.981788649772644e-05,
109
+ "loss": 1.4805,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.12,
114
+ "learning_rate": 4.97958612913223e-05,
115
+ "loss": 1.4352,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.13,
120
+ "learning_rate": 4.9772584559597976e-05,
121
+ "loss": 1.444,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.14,
126
+ "learning_rate": 4.9748057477403546e-05,
127
+ "loss": 1.4018,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.14,
132
+ "learning_rate": 4.9722281282698186e-05,
133
+ "loss": 1.4472,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.15,
138
+ "learning_rate": 4.9695257276487736e-05,
139
+ "loss": 1.3842,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.16,
144
+ "learning_rate": 4.966698682275903e-05,
145
+ "loss": 1.4116,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.16,
150
+ "learning_rate": 4.9637471348411005e-05,
151
+ "loss": 1.3827,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.17,
156
+ "learning_rate": 4.960671234318276e-05,
157
+ "loss": 1.3718,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.18,
162
+ "learning_rate": 4.957471135957828e-05,
163
+ "loss": 1.3204,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.18,
168
+ "learning_rate": 4.9541470012788135e-05,
169
+ "loss": 1.3307,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 0.19,
174
+ "learning_rate": 4.9506989980607945e-05,
175
+ "loss": 1.3216,
176
+ "step": 140
177
+ },
178
+ {
179
+ "epoch": 0.2,
180
+ "learning_rate": 4.9471273003353655e-05,
181
+ "loss": 1.3268,
182
+ "step": 145
183
+ },
184
+ {
185
+ "epoch": 0.2,
186
+ "learning_rate": 4.943432088377378e-05,
187
+ "loss": 1.4072,
188
+ "step": 150
189
+ },
190
+ {
191
+ "epoch": 0.21,
192
+ "learning_rate": 4.93961354869583e-05,
193
+ "loss": 1.2871,
194
+ "step": 155
195
+ },
196
+ {
197
+ "epoch": 0.22,
198
+ "learning_rate": 4.935671874024465e-05,
199
+ "loss": 1.3451,
200
+ "step": 160
201
+ },
202
+ {
203
+ "epoch": 0.22,
204
+ "learning_rate": 4.931607263312032e-05,
205
+ "loss": 1.3447,
206
+ "step": 165
207
+ },
208
+ {
209
+ "epoch": 0.23,
210
+ "learning_rate": 4.9274199217122565e-05,
211
+ "loss": 1.3573,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 0.24,
216
+ "learning_rate": 4.923110060573472e-05,
217
+ "loss": 1.2914,
218
+ "step": 175
219
+ },
220
+ {
221
+ "epoch": 0.24,
222
+ "learning_rate": 4.9186778974279625e-05,
223
+ "loss": 1.3249,
224
+ "step": 180
225
+ },
226
+ {
227
+ "epoch": 0.25,
228
+ "learning_rate": 4.914123655980981e-05,
229
+ "loss": 1.3253,
230
+ "step": 185
231
+ },
232
+ {
233
+ "epoch": 0.26,
234
+ "learning_rate": 4.909447566099457e-05,
235
+ "loss": 1.2495,
236
+ "step": 190
237
+ },
238
+ {
239
+ "epoch": 0.26,
240
+ "learning_rate": 4.9046498638003925e-05,
241
+ "loss": 1.2333,
242
+ "step": 195
243
+ },
244
+ {
245
+ "epoch": 0.27,
246
+ "learning_rate": 4.8997307912389554e-05,
247
+ "loss": 1.3015,
248
+ "step": 200
249
+ },
250
+ {
251
+ "epoch": 0.28,
252
+ "learning_rate": 4.89469059669625e-05,
253
+ "loss": 1.3331,
254
+ "step": 205
255
+ },
256
+ {
257
+ "epoch": 0.28,
258
+ "learning_rate": 4.889529534566791e-05,
259
+ "loss": 1.2935,
260
+ "step": 210
261
+ },
262
+ {
263
+ "epoch": 0.29,
264
+ "learning_rate": 4.884247865345662e-05,
265
+ "loss": 1.3166,
266
+ "step": 215
267
+ },
268
+ {
269
+ "epoch": 0.3,
270
+ "learning_rate": 4.8788458556153635e-05,
271
+ "loss": 1.3156,
272
+ "step": 220
273
+ },
274
+ {
275
+ "epoch": 0.31,
276
+ "learning_rate": 4.873323778032366e-05,
277
+ "loss": 1.2931,
278
+ "step": 225
279
+ },
280
+ {
281
+ "epoch": 0.31,
282
+ "learning_rate": 4.8676819113133386e-05,
283
+ "loss": 1.3569,
284
+ "step": 230
285
+ },
286
+ {
287
+ "epoch": 0.32,
288
+ "learning_rate": 4.861920540221089e-05,
289
+ "loss": 1.2342,
290
+ "step": 235
291
+ },
292
+ {
293
+ "epoch": 0.33,
294
+ "learning_rate": 4.856039955550186e-05,
295
+ "loss": 1.3015,
296
+ "step": 240
297
+ },
298
+ {
299
+ "epoch": 0.33,
300
+ "learning_rate": 4.8500404541122836e-05,
301
+ "loss": 1.3191,
302
+ "step": 245
303
+ },
304
+ {
305
+ "epoch": 0.34,
306
+ "learning_rate": 4.8439223387211406e-05,
307
+ "loss": 1.3178,
308
+ "step": 250
309
+ },
310
+ {
311
+ "epoch": 0.35,
312
+ "learning_rate": 4.837685918177337e-05,
313
+ "loss": 1.2669,
314
+ "step": 255
315
+ },
316
+ {
317
+ "epoch": 0.35,
318
+ "learning_rate": 4.8313315072526855e-05,
319
+ "loss": 1.2218,
320
+ "step": 260
321
+ },
322
+ {
323
+ "epoch": 0.36,
324
+ "learning_rate": 4.824859426674345e-05,
325
+ "loss": 1.304,
326
+ "step": 265
327
+ },
328
+ {
329
+ "epoch": 0.37,
330
+ "learning_rate": 4.818270003108635e-05,
331
+ "loss": 1.311,
332
+ "step": 270
333
+ },
334
+ {
335
+ "epoch": 0.37,
336
+ "learning_rate": 4.8115635691445436e-05,
337
+ "loss": 1.3036,
338
+ "step": 275
339
+ },
340
+ {
341
+ "epoch": 0.38,
342
+ "learning_rate": 4.8047404632769446e-05,
343
+ "loss": 1.258,
344
+ "step": 280
345
+ },
346
+ {
347
+ "epoch": 0.39,
348
+ "learning_rate": 4.797801029889508e-05,
349
+ "loss": 1.3186,
350
+ "step": 285
351
+ },
352
+ {
353
+ "epoch": 0.39,
354
+ "learning_rate": 4.7907456192373245e-05,
355
+ "loss": 1.395,
356
+ "step": 290
357
+ },
358
+ {
359
+ "epoch": 0.4,
360
+ "learning_rate": 4.783574587429219e-05,
361
+ "loss": 1.2793,
362
+ "step": 295
363
+ },
364
+ {
365
+ "epoch": 0.41,
366
+ "learning_rate": 4.776288296409785e-05,
367
+ "loss": 1.3625,
368
+ "step": 300
369
+ },
370
+ {
371
+ "epoch": 0.41,
372
+ "learning_rate": 4.768887113941107e-05,
373
+ "loss": 1.2453,
374
+ "step": 305
375
+ },
376
+ {
377
+ "epoch": 0.42,
378
+ "learning_rate": 4.761371413584209e-05,
379
+ "loss": 1.3312,
380
+ "step": 310
381
+ },
382
+ {
383
+ "epoch": 0.43,
384
+ "learning_rate": 4.7537415746801876e-05,
385
+ "loss": 1.2627,
386
+ "step": 315
387
+ },
388
+ {
389
+ "epoch": 0.43,
390
+ "learning_rate": 4.745997982331077e-05,
391
+ "loss": 1.2231,
392
+ "step": 320
393
+ },
394
+ {
395
+ "epoch": 0.44,
396
+ "learning_rate": 4.738141027380405e-05,
397
+ "loss": 1.2063,
398
+ "step": 325
399
+ },
400
+ {
401
+ "epoch": 0.45,
402
+ "learning_rate": 4.730171106393466e-05,
403
+ "loss": 1.3148,
404
+ "step": 330
405
+ },
406
+ {
407
+ "epoch": 0.45,
408
+ "learning_rate": 4.722088621637309e-05,
409
+ "loss": 1.2787,
410
+ "step": 335
411
+ },
412
+ {
413
+ "epoch": 0.46,
414
+ "learning_rate": 4.7138939810604286e-05,
415
+ "loss": 1.2111,
416
+ "step": 340
417
+ },
418
+ {
419
+ "epoch": 0.47,
420
+ "learning_rate": 4.7055875982721794e-05,
421
+ "loss": 1.3034,
422
+ "step": 345
423
+ },
424
+ {
425
+ "epoch": 0.47,
426
+ "learning_rate": 4.697169892521898e-05,
427
+ "loss": 1.288,
428
+ "step": 350
429
+ },
430
+ {
431
+ "epoch": 0.48,
432
+ "learning_rate": 4.688641288677739e-05,
433
+ "loss": 1.2935,
434
+ "step": 355
435
+ },
436
+ {
437
+ "epoch": 0.49,
438
+ "learning_rate": 4.680002217205238e-05,
439
+ "loss": 1.3183,
440
+ "step": 360
441
+ },
442
+ {
443
+ "epoch": 0.49,
444
+ "learning_rate": 4.6712531141455785e-05,
445
+ "loss": 1.1688,
446
+ "step": 365
447
+ },
448
+ {
449
+ "epoch": 0.5,
450
+ "learning_rate": 4.662394421093582e-05,
451
+ "loss": 1.2734,
452
+ "step": 370
453
+ },
454
+ {
455
+ "epoch": 0.51,
456
+ "learning_rate": 4.653426585175428e-05,
457
+ "loss": 1.2155,
458
+ "step": 375
459
+ },
460
+ {
461
+ "epoch": 0.52,
462
+ "learning_rate": 4.644350059026076e-05,
463
+ "loss": 1.2641,
464
+ "step": 380
465
+ },
466
+ {
467
+ "epoch": 0.52,
468
+ "learning_rate": 4.6351653007664276e-05,
469
+ "loss": 1.2819,
470
+ "step": 385
471
+ },
472
+ {
473
+ "epoch": 0.53,
474
+ "learning_rate": 4.625872773980197e-05,
475
+ "loss": 1.2435,
476
+ "step": 390
477
+ },
478
+ {
479
+ "epoch": 0.54,
480
+ "learning_rate": 4.61647294769052e-05,
481
+ "loss": 1.2541,
482
+ "step": 395
483
+ },
484
+ {
485
+ "epoch": 0.54,
486
+ "learning_rate": 4.6069662963362724e-05,
487
+ "loss": 1.2464,
488
+ "step": 400
489
+ },
490
+ {
491
+ "epoch": 0.55,
492
+ "learning_rate": 4.597353299748132e-05,
493
+ "loss": 1.2843,
494
+ "step": 405
495
+ },
496
+ {
497
+ "epoch": 0.56,
498
+ "learning_rate": 4.587634443124352e-05,
499
+ "loss": 1.3026,
500
+ "step": 410
501
+ },
502
+ {
503
+ "epoch": 0.56,
504
+ "learning_rate": 4.577810217006279e-05,
505
+ "loss": 1.2905,
506
+ "step": 415
507
+ },
508
+ {
509
+ "epoch": 0.57,
510
+ "learning_rate": 4.567881117253587e-05,
511
+ "loss": 1.2689,
512
+ "step": 420
513
+ },
514
+ {
515
+ "epoch": 0.58,
516
+ "learning_rate": 4.557847645019257e-05,
517
+ "loss": 1.2959,
518
+ "step": 425
519
+ },
520
+ {
521
+ "epoch": 0.58,
522
+ "learning_rate": 4.547710306724274e-05,
523
+ "loss": 1.2237,
524
+ "step": 430
525
+ },
526
+ {
527
+ "epoch": 0.59,
528
+ "learning_rate": 4.537469614032074e-05,
529
+ "loss": 1.2885,
530
+ "step": 435
531
+ },
532
+ {
533
+ "epoch": 0.6,
534
+ "learning_rate": 4.5271260838227124e-05,
535
+ "loss": 1.2622,
536
+ "step": 440
537
+ },
538
+ {
539
+ "epoch": 0.6,
540
+ "learning_rate": 4.516680238166783e-05,
541
+ "loss": 1.2567,
542
+ "step": 445
543
+ },
544
+ {
545
+ "epoch": 0.61,
546
+ "learning_rate": 4.506132604299057e-05,
547
+ "loss": 1.2618,
548
+ "step": 450
549
+ },
550
+ {
551
+ "epoch": 0.62,
552
+ "learning_rate": 4.4954837145918826e-05,
553
+ "loss": 1.249,
554
+ "step": 455
555
+ },
556
+ {
557
+ "epoch": 0.62,
558
+ "learning_rate": 4.484734106528307e-05,
559
+ "loss": 1.2891,
560
+ "step": 460
561
+ },
562
+ {
563
+ "epoch": 0.63,
564
+ "learning_rate": 4.473884322674952e-05,
565
+ "loss": 1.2423,
566
+ "step": 465
567
+ },
568
+ {
569
+ "epoch": 0.64,
570
+ "learning_rate": 4.462934910654624e-05,
571
+ "loss": 1.1933,
572
+ "step": 470
573
+ },
574
+ {
575
+ "epoch": 0.64,
576
+ "learning_rate": 4.451886423118682e-05,
577
+ "loss": 1.1707,
578
+ "step": 475
579
+ },
580
+ {
581
+ "epoch": 0.65,
582
+ "learning_rate": 4.4407394177191366e-05,
583
+ "loss": 1.2059,
584
+ "step": 480
585
+ },
586
+ {
587
+ "epoch": 0.66,
588
+ "learning_rate": 4.429494457080505e-05,
589
+ "loss": 1.2385,
590
+ "step": 485
591
+ },
592
+ {
593
+ "epoch": 0.66,
594
+ "learning_rate": 4.418152108771414e-05,
595
+ "loss": 1.2545,
596
+ "step": 490
597
+ },
598
+ {
599
+ "epoch": 0.67,
600
+ "learning_rate": 4.406712945275955e-05,
601
+ "loss": 1.2488,
602
+ "step": 495
603
+ },
604
+ {
605
+ "epoch": 0.68,
606
+ "learning_rate": 4.395177543964787e-05,
607
+ "loss": 1.2613,
608
+ "step": 500
609
+ },
610
+ {
611
+ "epoch": 0.68,
612
+ "learning_rate": 4.383546487065993e-05,
613
+ "loss": 1.1931,
614
+ "step": 505
615
+ },
616
+ {
617
+ "epoch": 0.69,
618
+ "learning_rate": 4.3718203616356954e-05,
619
+ "loss": 1.231,
620
+ "step": 510
621
+ },
622
+ {
623
+ "epoch": 0.7,
624
+ "learning_rate": 4.35999975952843e-05,
625
+ "loss": 1.1383,
626
+ "step": 515
627
+ },
628
+ {
629
+ "epoch": 0.71,
630
+ "learning_rate": 4.348085277367262e-05,
631
+ "loss": 1.1776,
632
+ "step": 520
633
+ },
634
+ {
635
+ "epoch": 0.71,
636
+ "learning_rate": 4.336077516513682e-05,
637
+ "loss": 1.2254,
638
+ "step": 525
639
+ },
640
+ {
641
+ "epoch": 0.72,
642
+ "learning_rate": 4.323977083037252e-05,
643
+ "loss": 1.2157,
644
+ "step": 530
645
+ },
646
+ {
647
+ "epoch": 0.73,
648
+ "learning_rate": 4.3117845876850085e-05,
649
+ "loss": 1.1688,
650
+ "step": 535
651
+ },
652
+ {
653
+ "epoch": 0.73,
654
+ "learning_rate": 4.299500645850647e-05,
655
+ "loss": 1.2387,
656
+ "step": 540
657
+ },
658
+ {
659
+ "epoch": 0.74,
660
+ "learning_rate": 4.287125877543453e-05,
661
+ "loss": 1.2281,
662
+ "step": 545
663
+ },
664
+ {
665
+ "epoch": 0.75,
666
+ "learning_rate": 4.274660907357009e-05,
667
+ "loss": 1.0799,
668
+ "step": 550
669
+ },
670
+ {
671
+ "epoch": 0.75,
672
+ "learning_rate": 4.262106364437671e-05,
673
+ "loss": 1.2725,
674
+ "step": 555
675
+ },
676
+ {
677
+ "epoch": 0.76,
678
+ "learning_rate": 4.2494628824528184e-05,
679
+ "loss": 1.2027,
680
+ "step": 560
681
+ },
682
+ {
683
+ "epoch": 0.77,
684
+ "learning_rate": 4.2367310995588585e-05,
685
+ "loss": 1.3018,
686
+ "step": 565
687
+ },
688
+ {
689
+ "epoch": 0.77,
690
+ "learning_rate": 4.223911658369027e-05,
691
+ "loss": 1.2029,
692
+ "step": 570
693
+ },
694
+ {
695
+ "epoch": 0.78,
696
+ "learning_rate": 4.211005205920953e-05,
697
+ "loss": 1.2697,
698
+ "step": 575
699
+ },
700
+ {
701
+ "epoch": 0.79,
702
+ "learning_rate": 4.198012393643994e-05,
703
+ "loss": 1.2548,
704
+ "step": 580
705
+ },
706
+ {
707
+ "epoch": 0.79,
708
+ "learning_rate": 4.1849338773263625e-05,
709
+ "loss": 1.2451,
710
+ "step": 585
711
+ },
712
+ {
713
+ "epoch": 0.8,
714
+ "learning_rate": 4.171770317082024e-05,
715
+ "loss": 1.2411,
716
+ "step": 590
717
+ },
718
+ {
719
+ "epoch": 0.81,
720
+ "learning_rate": 4.158522377317379e-05,
721
+ "loss": 1.2933,
722
+ "step": 595
723
+ },
724
+ {
725
+ "epoch": 0.81,
726
+ "learning_rate": 4.1451907266977285e-05,
727
+ "loss": 1.2684,
728
+ "step": 600
729
+ },
730
+ {
731
+ "epoch": 0.82,
732
+ "learning_rate": 4.131776038113524e-05,
733
+ "loss": 1.3233,
734
+ "step": 605
735
+ },
736
+ {
737
+ "epoch": 0.83,
738
+ "learning_rate": 4.118278988646407e-05,
739
+ "loss": 1.1725,
740
+ "step": 610
741
+ },
742
+ {
743
+ "epoch": 0.83,
744
+ "learning_rate": 4.104700259535029e-05,
745
+ "loss": 1.1811,
746
+ "step": 615
747
+ },
748
+ {
749
+ "epoch": 0.84,
750
+ "learning_rate": 4.091040536140673e-05,
751
+ "loss": 1.1264,
752
+ "step": 620
753
+ },
754
+ {
755
+ "epoch": 0.85,
756
+ "learning_rate": 4.07730050791266e-05,
757
+ "loss": 1.2772,
758
+ "step": 625
759
+ },
760
+ {
761
+ "epoch": 0.85,
762
+ "learning_rate": 4.063480868353546e-05,
763
+ "loss": 1.2536,
764
+ "step": 630
765
+ },
766
+ {
767
+ "epoch": 0.86,
768
+ "learning_rate": 4.049582314984124e-05,
769
+ "loss": 1.2617,
770
+ "step": 635
771
+ },
772
+ {
773
+ "epoch": 0.87,
774
+ "learning_rate": 4.0356055493082166e-05,
775
+ "loss": 1.2038,
776
+ "step": 640
777
+ },
778
+ {
779
+ "epoch": 0.87,
780
+ "learning_rate": 4.021551276777266e-05,
781
+ "loss": 1.2328,
782
+ "step": 645
783
+ },
784
+ {
785
+ "epoch": 0.88,
786
+ "learning_rate": 4.007420206754732e-05,
787
+ "loss": 1.244,
788
+ "step": 650
789
+ },
790
+ {
791
+ "epoch": 0.89,
792
+ "learning_rate": 3.993213052480285e-05,
793
+ "loss": 1.1877,
794
+ "step": 655
795
+ },
796
+ {
797
+ "epoch": 0.89,
798
+ "learning_rate": 3.978930531033807e-05,
799
+ "loss": 1.1767,
800
+ "step": 660
801
+ },
802
+ {
803
+ "epoch": 0.9,
804
+ "learning_rate": 3.964573363299198e-05,
805
+ "loss": 1.1808,
806
+ "step": 665
807
+ },
808
+ {
809
+ "epoch": 0.91,
810
+ "learning_rate": 3.9501422739279956e-05,
811
+ "loss": 1.1984,
812
+ "step": 670
813
+ },
814
+ {
815
+ "epoch": 0.92,
816
+ "learning_rate": 3.935637991302792e-05,
817
+ "loss": 1.234,
818
+ "step": 675
819
+ },
820
+ {
821
+ "epoch": 0.92,
822
+ "learning_rate": 3.921061247500474e-05,
823
+ "loss": 1.2115,
824
+ "step": 680
825
+ },
826
+ {
827
+ "epoch": 0.93,
828
+ "learning_rate": 3.906412778255274e-05,
829
+ "loss": 1.2645,
830
+ "step": 685
831
+ },
832
+ {
833
+ "epoch": 0.94,
834
+ "learning_rate": 3.8916933229216315e-05,
835
+ "loss": 1.1742,
836
+ "step": 690
837
+ },
838
+ {
839
+ "epoch": 0.94,
840
+ "learning_rate": 3.8769036244368815e-05,
841
+ "loss": 1.1966,
842
+ "step": 695
843
+ },
844
+ {
845
+ "epoch": 0.95,
846
+ "learning_rate": 3.8620444292837485e-05,
847
+ "loss": 1.1635,
848
+ "step": 700
849
+ },
850
+ {
851
+ "epoch": 0.96,
852
+ "learning_rate": 3.847116487452674e-05,
853
+ "loss": 1.1874,
854
+ "step": 705
855
+ },
856
+ {
857
+ "epoch": 0.96,
858
+ "learning_rate": 3.8321205524039635e-05,
859
+ "loss": 1.166,
860
+ "step": 710
861
+ },
862
+ {
863
+ "epoch": 0.97,
864
+ "learning_rate": 3.817057381029752e-05,
865
+ "loss": 1.1213,
866
+ "step": 715
867
+ },
868
+ {
869
+ "epoch": 0.98,
870
+ "learning_rate": 3.801927733615804e-05,
871
+ "loss": 1.0497,
872
+ "step": 720
873
+ },
874
+ {
875
+ "epoch": 0.98,
876
+ "learning_rate": 3.786732373803141e-05,
877
+ "loss": 1.1248,
878
+ "step": 725
879
+ },
880
+ {
881
+ "epoch": 0.99,
882
+ "learning_rate": 3.771472068549496e-05,
883
+ "loss": 1.2836,
884
+ "step": 730
885
+ },
886
+ {
887
+ "epoch": 1.0,
888
+ "learning_rate": 3.7561475880906026e-05,
889
+ "loss": 1.1569,
890
+ "step": 735
891
+ },
892
+ {
893
+ "epoch": 1.0,
894
+ "learning_rate": 3.7407597059013195e-05,
895
+ "loss": 1.1206,
896
+ "step": 740
897
+ },
898
+ {
899
+ "epoch": 1.01,
900
+ "learning_rate": 3.72530919865659e-05,
901
+ "loss": 1.175,
902
+ "step": 745
903
+ },
904
+ {
905
+ "epoch": 1.02,
906
+ "learning_rate": 3.709796846192243e-05,
907
+ "loss": 1.1392,
908
+ "step": 750
909
+ },
910
+ {
911
+ "epoch": 1.02,
912
+ "learning_rate": 3.694223431465631e-05,
913
+ "loss": 1.1942,
914
+ "step": 755
915
+ },
916
+ {
917
+ "epoch": 1.03,
918
+ "learning_rate": 3.678589740516109e-05,
919
+ "loss": 1.2052,
920
+ "step": 760
921
+ },
922
+ {
923
+ "epoch": 1.04,
924
+ "learning_rate": 3.662896562425364e-05,
925
+ "loss": 1.2162,
926
+ "step": 765
927
+ },
928
+ {
929
+ "epoch": 1.04,
930
+ "learning_rate": 3.6471446892775895e-05,
931
+ "loss": 1.2136,
932
+ "step": 770
933
+ },
934
+ {
935
+ "epoch": 1.05,
936
+ "learning_rate": 3.6313349161194995e-05,
937
+ "loss": 1.146,
938
+ "step": 775
939
+ },
940
+ {
941
+ "epoch": 1.06,
942
+ "learning_rate": 3.6154680409202064e-05,
943
+ "loss": 1.187,
944
+ "step": 780
945
+ },
946
+ {
947
+ "epoch": 1.06,
948
+ "learning_rate": 3.5995448645309424e-05,
949
+ "loss": 1.2493,
950
+ "step": 785
951
+ },
952
+ {
953
+ "epoch": 1.07,
954
+ "learning_rate": 3.583566190644638e-05,
955
+ "loss": 1.2171,
956
+ "step": 790
957
+ },
958
+ {
959
+ "epoch": 1.08,
960
+ "learning_rate": 3.567532825755358e-05,
961
+ "loss": 1.214,
962
+ "step": 795
963
+ },
964
+ {
965
+ "epoch": 1.08,
966
+ "learning_rate": 3.551445579117592e-05,
967
+ "loss": 1.1566,
968
+ "step": 800
969
+ },
970
+ {
971
+ "epoch": 1.09,
972
+ "learning_rate": 3.5353052627054155e-05,
973
+ "loss": 1.1658,
974
+ "step": 805
975
+ },
976
+ {
977
+ "epoch": 1.1,
978
+ "learning_rate": 3.519112691171501e-05,
979
+ "loss": 1.182,
980
+ "step": 810
981
+ },
982
+ {
983
+ "epoch": 1.11,
984
+ "learning_rate": 3.5028686818060016e-05,
985
+ "loss": 1.1485,
986
+ "step": 815
987
+ },
988
+ {
989
+ "epoch": 1.11,
990
+ "learning_rate": 3.4865740544952995e-05,
991
+ "loss": 1.1847,
992
+ "step": 820
993
+ },
994
+ {
995
+ "epoch": 1.12,
996
+ "learning_rate": 3.4702296316806244e-05,
997
+ "loss": 1.134,
998
+ "step": 825
999
+ },
1000
+ {
1001
+ "epoch": 1.13,
1002
+ "learning_rate": 3.4538362383165434e-05,
1003
+ "loss": 1.1939,
1004
+ "step": 830
1005
+ },
1006
+ {
1007
+ "epoch": 1.13,
1008
+ "learning_rate": 3.437394701829321e-05,
1009
+ "loss": 1.2424,
1010
+ "step": 835
1011
+ },
1012
+ {
1013
+ "epoch": 1.14,
1014
+ "learning_rate": 3.420905852075156e-05,
1015
+ "loss": 1.1147,
1016
+ "step": 840
1017
+ },
1018
+ {
1019
+ "epoch": 1.15,
1020
+ "learning_rate": 3.4043705212982987e-05,
1021
+ "loss": 1.1557,
1022
+ "step": 845
1023
+ },
1024
+ {
1025
+ "epoch": 1.15,
1026
+ "learning_rate": 3.38778954408904e-05,
1027
+ "loss": 1.2858,
1028
+ "step": 850
1029
+ },
1030
+ {
1031
+ "epoch": 1.16,
1032
+ "learning_rate": 3.371163757341597e-05,
1033
+ "loss": 1.1724,
1034
+ "step": 855
1035
+ },
1036
+ {
1037
+ "epoch": 1.17,
1038
+ "learning_rate": 3.354494000211856e-05,
1039
+ "loss": 1.2191,
1040
+ "step": 860
1041
+ },
1042
+ {
1043
+ "epoch": 1.17,
1044
+ "learning_rate": 3.337781114075035e-05,
1045
+ "loss": 1.1749,
1046
+ "step": 865
1047
+ },
1048
+ {
1049
+ "epoch": 1.18,
1050
+ "learning_rate": 3.321025942483205e-05,
1051
+ "loss": 1.1776,
1052
+ "step": 870
1053
+ },
1054
+ {
1055
+ "epoch": 1.19,
1056
+ "learning_rate": 3.3042293311227194e-05,
1057
+ "loss": 1.2484,
1058
+ "step": 875
1059
+ },
1060
+ {
1061
+ "epoch": 1.19,
1062
+ "learning_rate": 3.287392127771526e-05,
1063
+ "loss": 1.1541,
1064
+ "step": 880
1065
+ },
1066
+ {
1067
+ "epoch": 1.2,
1068
+ "learning_rate": 3.2705151822563784e-05,
1069
+ "loss": 1.2192,
1070
+ "step": 885
1071
+ },
1072
+ {
1073
+ "epoch": 1.21,
1074
+ "learning_rate": 3.2535993464099444e-05,
1075
+ "loss": 1.1494,
1076
+ "step": 890
1077
+ },
1078
+ {
1079
+ "epoch": 1.21,
1080
+ "learning_rate": 3.236645474027807e-05,
1081
+ "loss": 1.2086,
1082
+ "step": 895
1083
+ },
1084
+ {
1085
+ "epoch": 1.22,
1086
+ "learning_rate": 3.219654420825373e-05,
1087
+ "loss": 1.1576,
1088
+ "step": 900
1089
+ },
1090
+ {
1091
+ "epoch": 1.23,
1092
+ "learning_rate": 3.202627044394686e-05,
1093
+ "loss": 1.1459,
1094
+ "step": 905
1095
+ },
1096
+ {
1097
+ "epoch": 1.23,
1098
+ "learning_rate": 3.185564204161134e-05,
1099
+ "loss": 1.1342,
1100
+ "step": 910
1101
+ },
1102
+ {
1103
+ "epoch": 1.24,
1104
+ "learning_rate": 3.168466761340076e-05,
1105
+ "loss": 1.1235,
1106
+ "step": 915
1107
+ },
1108
+ {
1109
+ "epoch": 1.25,
1110
+ "learning_rate": 3.151335578893374e-05,
1111
+ "loss": 1.1784,
1112
+ "step": 920
1113
+ },
1114
+ {
1115
+ "epoch": 1.25,
1116
+ "learning_rate": 3.1341715214858316e-05,
1117
+ "loss": 1.178,
1118
+ "step": 925
1119
+ },
1120
+ {
1121
+ "epoch": 1.26,
1122
+ "learning_rate": 3.1169754554415605e-05,
1123
+ "loss": 1.1898,
1124
+ "step": 930
1125
+ },
1126
+ {
1127
+ "epoch": 1.27,
1128
+ "learning_rate": 3.099748248700245e-05,
1129
+ "loss": 1.1082,
1130
+ "step": 935
1131
+ },
1132
+ {
1133
+ "epoch": 1.27,
1134
+ "learning_rate": 3.082490770773341e-05,
1135
+ "loss": 1.2071,
1136
+ "step": 940
1137
+ },
1138
+ {
1139
+ "epoch": 1.28,
1140
+ "learning_rate": 3.065203892700186e-05,
1141
+ "loss": 1.1081,
1142
+ "step": 945
1143
+ },
1144
+ {
1145
+ "epoch": 1.29,
1146
+ "learning_rate": 3.0478884870040348e-05,
1147
+ "loss": 1.1357,
1148
+ "step": 950
1149
+ },
1150
+ {
1151
+ "epoch": 1.29,
1152
+ "learning_rate": 3.030545427648022e-05,
1153
+ "loss": 1.1597,
1154
+ "step": 955
1155
+ },
1156
+ {
1157
+ "epoch": 1.3,
1158
+ "learning_rate": 3.0131755899910502e-05,
1159
+ "loss": 1.1475,
1160
+ "step": 960
1161
+ },
1162
+ {
1163
+ "epoch": 1.31,
1164
+ "learning_rate": 2.9957798507436064e-05,
1165
+ "loss": 1.1334,
1166
+ "step": 965
1167
+ },
1168
+ {
1169
+ "epoch": 1.32,
1170
+ "learning_rate": 2.9783590879235124e-05,
1171
+ "loss": 1.169,
1172
+ "step": 970
1173
+ },
1174
+ {
1175
+ "epoch": 1.32,
1176
+ "learning_rate": 2.9609141808116103e-05,
1177
+ "loss": 1.1756,
1178
+ "step": 975
1179
+ },
1180
+ {
1181
+ "epoch": 1.33,
1182
+ "learning_rate": 2.9434460099073773e-05,
1183
+ "loss": 1.1572,
1184
+ "step": 980
1185
+ },
1186
+ {
1187
+ "epoch": 1.34,
1188
+ "learning_rate": 2.9259554568844916e-05,
1189
+ "loss": 1.1201,
1190
+ "step": 985
1191
+ },
1192
+ {
1193
+ "epoch": 1.34,
1194
+ "learning_rate": 2.9084434045463255e-05,
1195
+ "loss": 1.0908,
1196
+ "step": 990
1197
+ },
1198
+ {
1199
+ "epoch": 1.35,
1200
+ "learning_rate": 2.8909107367813893e-05,
1201
+ "loss": 1.1573,
1202
+ "step": 995
1203
+ },
1204
+ {
1205
+ "epoch": 1.36,
1206
+ "learning_rate": 2.8733583385187195e-05,
1207
+ "loss": 1.1656,
1208
+ "step": 1000
1209
+ },
1210
+ {
1211
+ "epoch": 1.36,
1212
+ "learning_rate": 2.8557870956832132e-05,
1213
+ "loss": 1.1984,
1214
+ "step": 1005
1215
+ },
1216
+ {
1217
+ "epoch": 1.37,
1218
+ "learning_rate": 2.8381978951509115e-05,
1219
+ "loss": 1.1605,
1220
+ "step": 1010
1221
+ },
1222
+ {
1223
+ "epoch": 1.38,
1224
+ "learning_rate": 2.8205916247042384e-05,
1225
+ "loss": 1.1765,
1226
+ "step": 1015
1227
+ },
1228
+ {
1229
+ "epoch": 1.38,
1230
+ "learning_rate": 2.8029691729871898e-05,
1231
+ "loss": 1.1306,
1232
+ "step": 1020
1233
+ },
1234
+ {
1235
+ "epoch": 1.39,
1236
+ "learning_rate": 2.7853314294604803e-05,
1237
+ "loss": 1.1954,
1238
+ "step": 1025
1239
+ },
1240
+ {
1241
+ "epoch": 1.4,
1242
+ "learning_rate": 2.767679284356653e-05,
1243
+ "loss": 1.1751,
1244
+ "step": 1030
1245
+ },
1246
+ {
1247
+ "epoch": 1.4,
1248
+ "learning_rate": 2.7500136286351407e-05,
1249
+ "loss": 1.0568,
1250
+ "step": 1035
1251
+ },
1252
+ {
1253
+ "epoch": 1.41,
1254
+ "learning_rate": 2.732335353937302e-05,
1255
+ "loss": 1.1568,
1256
+ "step": 1040
1257
+ },
1258
+ {
1259
+ "epoch": 1.42,
1260
+ "learning_rate": 2.714645352541415e-05,
1261
+ "loss": 1.1541,
1262
+ "step": 1045
1263
+ },
1264
+ {
1265
+ "epoch": 1.42,
1266
+ "learning_rate": 2.69694451731764e-05,
1267
+ "loss": 1.1995,
1268
+ "step": 1050
1269
+ },
1270
+ {
1271
+ "epoch": 1.43,
1272
+ "learning_rate": 2.6792337416829554e-05,
1273
+ "loss": 1.1831,
1274
+ "step": 1055
1275
+ },
1276
+ {
1277
+ "epoch": 1.44,
1278
+ "learning_rate": 2.6615139195560635e-05,
1279
+ "loss": 1.1272,
1280
+ "step": 1060
1281
+ },
1282
+ {
1283
+ "epoch": 1.44,
1284
+ "learning_rate": 2.643785945312271e-05,
1285
+ "loss": 1.1833,
1286
+ "step": 1065
1287
+ },
1288
+ {
1289
+ "epoch": 1.45,
1290
+ "learning_rate": 2.626050713738349e-05,
1291
+ "loss": 1.1971,
1292
+ "step": 1070
1293
+ },
1294
+ {
1295
+ "epoch": 1.46,
1296
+ "learning_rate": 2.6083091199873677e-05,
1297
+ "loss": 1.1645,
1298
+ "step": 1075
1299
+ },
1300
+ {
1301
+ "epoch": 1.46,
1302
+ "learning_rate": 2.5905620595335172e-05,
1303
+ "loss": 1.1932,
1304
+ "step": 1080
1305
+ },
1306
+ {
1307
+ "epoch": 1.47,
1308
+ "learning_rate": 2.5728104281269087e-05,
1309
+ "loss": 1.1351,
1310
+ "step": 1085
1311
+ },
1312
+ {
1313
+ "epoch": 1.48,
1314
+ "learning_rate": 2.5550551217483642e-05,
1315
+ "loss": 1.0708,
1316
+ "step": 1090
1317
+ },
1318
+ {
1319
+ "epoch": 1.48,
1320
+ "learning_rate": 2.5372970365641928e-05,
1321
+ "loss": 1.1435,
1322
+ "step": 1095
1323
+ },
1324
+ {
1325
+ "epoch": 1.49,
1326
+ "learning_rate": 2.51953706888096e-05,
1327
+ "loss": 1.1483,
1328
+ "step": 1100
1329
+ },
1330
+ {
1331
+ "epoch": 1.5,
1332
+ "learning_rate": 2.5017761151002443e-05,
1333
+ "loss": 1.1173,
1334
+ "step": 1105
1335
+ },
1336
+ {
1337
+ "epoch": 1.51,
1338
+ "learning_rate": 2.4840150716733977e-05,
1339
+ "loss": 1.2243,
1340
+ "step": 1110
1341
+ },
1342
+ {
1343
+ "epoch": 1.51,
1344
+ "learning_rate": 2.466254835056298e-05,
1345
+ "loss": 1.1516,
1346
+ "step": 1115
1347
+ },
1348
+ {
1349
+ "epoch": 1.52,
1350
+ "learning_rate": 2.448496301664097e-05,
1351
+ "loss": 1.0848,
1352
+ "step": 1120
1353
+ },
1354
+ {
1355
+ "epoch": 1.53,
1356
+ "learning_rate": 2.4307403678259832e-05,
1357
+ "loss": 1.0821,
1358
+ "step": 1125
1359
+ },
1360
+ {
1361
+ "epoch": 1.53,
1362
+ "learning_rate": 2.4129879297399353e-05,
1363
+ "loss": 1.1737,
1364
+ "step": 1130
1365
+ },
1366
+ {
1367
+ "epoch": 1.54,
1368
+ "learning_rate": 2.3952398834274902e-05,
1369
+ "loss": 1.1498,
1370
+ "step": 1135
1371
+ },
1372
+ {
1373
+ "epoch": 1.55,
1374
+ "learning_rate": 2.3774971246885197e-05,
1375
+ "loss": 1.1553,
1376
+ "step": 1140
1377
+ },
1378
+ {
1379
+ "epoch": 1.55,
1380
+ "learning_rate": 2.3597605490560134e-05,
1381
+ "loss": 1.1054,
1382
+ "step": 1145
1383
+ },
1384
+ {
1385
+ "epoch": 1.56,
1386
+ "learning_rate": 2.342031051750882e-05,
1387
+ "loss": 1.2051,
1388
+ "step": 1150
1389
+ },
1390
+ {
1391
+ "epoch": 1.57,
1392
+ "learning_rate": 2.3243095276367685e-05,
1393
+ "loss": 1.1477,
1394
+ "step": 1155
1395
+ },
1396
+ {
1397
+ "epoch": 1.57,
1398
+ "learning_rate": 2.3065968711748838e-05,
1399
+ "loss": 1.1917,
1400
+ "step": 1160
1401
+ },
1402
+ {
1403
+ "epoch": 1.58,
1404
+ "learning_rate": 2.2888939763788637e-05,
1405
+ "loss": 1.088,
1406
+ "step": 1165
1407
+ },
1408
+ {
1409
+ "epoch": 1.59,
1410
+ "learning_rate": 2.271201736769639e-05,
1411
+ "loss": 1.1747,
1412
+ "step": 1170
1413
+ },
1414
+ {
1415
+ "epoch": 1.59,
1416
+ "learning_rate": 2.2535210453303395e-05,
1417
+ "loss": 1.2312,
1418
+ "step": 1175
1419
+ },
1420
+ {
1421
+ "epoch": 1.6,
1422
+ "learning_rate": 2.2358527944612244e-05,
1423
+ "loss": 1.1272,
1424
+ "step": 1180
1425
+ },
1426
+ {
1427
+ "epoch": 1.61,
1428
+ "learning_rate": 2.2181978759346357e-05,
1429
+ "loss": 1.1451,
1430
+ "step": 1185
1431
+ },
1432
+ {
1433
+ "epoch": 1.61,
1434
+ "learning_rate": 2.2005571808499916e-05,
1435
+ "loss": 1.1075,
1436
+ "step": 1190
1437
+ },
1438
+ {
1439
+ "epoch": 1.62,
1440
+ "learning_rate": 2.1829315995888087e-05,
1441
+ "loss": 1.1211,
1442
+ "step": 1195
1443
+ },
1444
+ {
1445
+ "epoch": 1.63,
1446
+ "learning_rate": 2.1653220217697594e-05,
1447
+ "loss": 1.1199,
1448
+ "step": 1200
1449
+ }
1450
+ ],
1451
+ "max_steps": 2211,
1452
+ "num_train_epochs": 3,
1453
+ "total_flos": 5.783610728583168e+17,
1454
+ "trial_name": null,
1455
+ "trial_params": null
1456
+ }
checkpoint-1200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0090099d32ad29c1bd6d98560579b7c6ef7a6d711729f1926d4fd6f8220f669c
3
+ size 3361
checkpoint-1300/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-1300/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
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": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-1300/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab3427ece84e2d941b59f16297bc61cb49e3a96114b66c1642e31c981c58b121
3
+ size 16821197
checkpoint-1300/finetuning_args.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dpo_beta": 0.1,
3
+ "finetuning_type": "lora",
4
+ "lora_alpha": 32.0,
5
+ "lora_dropout": 0.1,
6
+ "lora_rank": 8,
7
+ "lora_target": [
8
+ "q_proj",
9
+ "v_proj"
10
+ ],
11
+ "name_module_trainable": "mlp",
12
+ "num_hidden_layers": 32,
13
+ "num_layer_trainable": 3,
14
+ "ppo_score_norm": false,
15
+ "resume_lora_training": true
16
+ }
checkpoint-1300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bda7effbe9c3826198daac76dbc925c941133356a44a1003bd142430ebb500b8
3
+ size 33629893
checkpoint-1300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:803d2d24da78ef65e06fcb990193643ac404026a49c32e712f80b32a7765aa72
3
+ size 14575