hZzy commited on
Commit
c4f2547
·
verified ·
1 Parent(s): 42b6699

Model save

Browse files
README.md ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: hZzy/qwen2.5-0.5b-sft-news-IFT
4
+ tags:
5
+ - trl
6
+ - expo
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: qwen2.5-0.5b-expo-L2EXPO-ES-0.1-W0
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhiyuzha-university-of-florida/huggingface/runs/ogsqjyxu)
17
+ # qwen2.5-0.5b-expo-L2EXPO-ES-0.1-W0
18
+
19
+ This model is a fine-tuned version of [hZzy/qwen2.5-0.5b-sft-news-IFT](https://huggingface.co/hZzy/qwen2.5-0.5b-sft-news-IFT) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 280.5317
22
+ - Logps: -82.9107
23
+ - Logits: -0.4190
24
+ - Objective: 274.7932
25
+ - Dpo Loss: 0.7169
26
+ - Regularize: 0.6260
27
+ - Ranking Simple: 0.5326
28
+ - Ranking Idealized: 0.6030
29
+ - Ranking Idealized Expo: 0.5223
30
+ - Wo Beta: 14.3418
31
+
32
+ ## Model description
33
+
34
+ More information needed
35
+
36
+ ## Intended uses & limitations
37
+
38
+ More information needed
39
+
40
+ ## Training and evaluation data
41
+
42
+ More information needed
43
+
44
+ ## Training procedure
45
+
46
+ ### Training hyperparameters
47
+
48
+ The following hyperparameters were used during training:
49
+ - learning_rate: 5e-06
50
+ - train_batch_size: 4
51
+ - eval_batch_size: 4
52
+ - seed: 42
53
+ - distributed_type: multi-GPU
54
+ - num_devices: 3
55
+ - gradient_accumulation_steps: 12
56
+ - total_train_batch_size: 144
57
+ - total_eval_batch_size: 12
58
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
+ - lr_scheduler_type: cosine
60
+ - lr_scheduler_warmup_ratio: 0.1
61
+ - num_epochs: 5
62
+
63
+ ### Training results
64
+
65
+ | Training Loss | Epoch | Step | Dpo Loss | Logits | Logps | Validation Loss | Objective | Ranking Idealized | Ranking Idealized Expo | Ranking Simple | Regularize | Wo Beta |
66
+ |:-------------:|:------:|:----:|:--------:|:-------:|:--------:|:---------------:|:---------:|:-----------------:|:----------------------:|:--------------:|:----------:|:-------:|
67
+ | 176.8183 | 0.1417 | 50 | 0.6874 | -1.4745 | -93.3291 | 185.6754 | 183.7365 | 0.6030 | 0.5223 | 0.5212 | 0.4178 | 16.4509 |
68
+ | 168.1755 | 0.2834 | 100 | 0.6819 | -1.4487 | -93.7829 | 195.2546 | 190.2426 | 0.6030 | 0.5223 | 0.5342 | 0.4320 | 16.2920 |
69
+ | 182.7148 | 0.4251 | 150 | 0.6969 | -1.2730 | -89.7329 | 218.5299 | 213.4884 | 0.6030 | 0.5223 | 0.5336 | 0.4859 | 15.8045 |
70
+ | 203.0993 | 0.5668 | 200 | 0.7051 | -1.0447 | -79.7062 | 251.9243 | 242.6406 | 0.6030 | 0.5223 | 0.5326 | 0.5518 | 14.6949 |
71
+ | 207.5481 | 0.7085 | 250 | 0.7055 | -1.0362 | -80.0940 | 251.9905 | 244.3510 | 0.6030 | 0.5223 | 0.5305 | 0.5542 | 14.8158 |
72
+ | 193.4843 | 0.8503 | 300 | 0.7150 | -0.7137 | -80.4296 | 266.7107 | 258.3957 | 0.6030 | 0.5223 | 0.5290 | 0.5881 | 14.5431 |
73
+ | 182.6922 | 0.9920 | 350 | 0.7073 | -0.6448 | -76.3638 | 262.3346 | 254.6360 | 0.6030 | 0.5223 | 0.5357 | 0.5802 | 14.6176 |
74
+ | 166.9683 | 1.1337 | 400 | 0.7152 | -0.6392 | -78.3482 | 272.3288 | 264.9111 | 0.6030 | 0.5223 | 0.5274 | 0.6056 | 14.6513 |
75
+ | 155.9364 | 1.2754 | 450 | 0.7186 | -0.4207 | -80.5230 | 275.0490 | 268.8637 | 0.6030 | 0.5223 | 0.5321 | 0.6129 | 14.7777 |
76
+ | 143.4724 | 1.4171 | 500 | 0.7209 | -0.5141 | -80.5587 | 275.9663 | 270.0383 | 0.6030 | 0.5223 | 0.5269 | 0.6150 | 14.4364 |
77
+ | 141.3444 | 1.5588 | 550 | 0.7139 | -0.6338 | -81.1271 | 275.0851 | 269.2189 | 0.6030 | 0.5223 | 0.5378 | 0.6159 | 14.6425 |
78
+ | 136.172 | 1.7029 | 600 | 273.6681 | -79.4221| -0.5857 | 264.6510 | 0.7111 | 0.6012 | 0.5373 | 0.6030 | 0.5223 | 14.5631 |
79
+ | 130.7133 | 1.8446 | 650 | 276.3609 | -80.2130| -0.4215 | 269.6939 | 0.7193 | 0.6141 | 0.5342 | 0.6030 | 0.5223 | 14.5456 |
80
+ | 122.624 | 1.9863 | 700 | 278.4690 | -80.9968| -0.5263 | 271.4757 | 0.7178 | 0.6190 | 0.5378 | 0.6030 | 0.5223 | 14.4664 |
81
+ | 108.7022 | 2.1280 | 750 | 282.5668 | -84.0088| -0.4657 | 276.0201 | 0.7207 | 0.6302 | 0.5347 | 0.6030 | 0.5223 | 14.4517 |
82
+ | 104.1923 | 2.2697 | 800 | 278.0555 | -81.6313| -0.4640 | 272.7622 | 0.7166 | 0.6210 | 0.5383 | 0.6030 | 0.5223 | 14.4307 |
83
+ | 99.0867 | 2.4114 | 850 | 283.0078 | -81.0689| -0.5212 | 277.3703 | 0.7209 | 0.6310 | 0.5331 | 0.6030 | 0.5223 | 14.2695 |
84
+ | 91.7475 | 2.5531 | 900 | 279.6676 | -81.6144| -0.5149 | 275.1769 | 0.7200 | 0.6279 | 0.5373 | 0.6030 | 0.5223 | 14.3570 |
85
+ | 87.8681 | 2.6949 | 950 | 281.5718 | -81.8544| -0.4428 | 275.7560 | 0.7191 | 0.6277 | 0.5362 | 0.6030 | 0.5223 | 14.3509 |
86
+ | 81.742 | 2.8366 | 1000 | 279.1324 | -81.4412| -0.4951 | 274.5647 | 0.7197 | 0.6257 | 0.5336 | 0.6030 | 0.5223 | 14.3551 |
87
+ | 76.4372 | 2.9783 | 1050 | 279.1884 | -82.3960| -0.4502 | 273.9026 | 0.7184 | 0.6249 | 0.5336 | 0.6030 | 0.5223 | 14.3203 |
88
+ | 67.4698 | 3.1200 | 1100 | 280.5317 | -82.9107| -0.4190 | 274.7932 | 0.7169 | 0.6260 | 0.5326 | 0.6030 | 0.5223 | 14.3418 |
89
+
90
+
91
+ ### Framework versions
92
+
93
+ - Transformers 4.42.0
94
+ - Pytorch 2.3.0+cu121
95
+ - Datasets 3.2.0
96
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.119981105337742,
3
+ "total_flos": 0.0,
4
+ "train_loss": 50.30704811789773,
5
+ "train_runtime": 15116.5529,
6
+ "train_samples": 50802,
7
+ "train_samples_per_second": 16.803,
8
+ "train_steps_per_second": 0.116
9
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151644,
3
+ "eos_token_id": 151645,
4
+ "max_new_tokens": 2048,
5
+ "pad_token_id": 151645,
6
+ "transformers_version": "4.42.0"
7
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:74adf7c97a2de35551456a244dc6dbcf959345e4111a7757138f0dacd914e2dc
3
  size 1975192208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0efb989df654aa7604aec66f19416e087c42ddaa936854e9c942ac912595543c
3
  size 1975192208
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.119981105337742,
3
+ "total_flos": 0.0,
4
+ "train_loss": 50.30704811789773,
5
+ "train_runtime": 15116.5529,
6
+ "train_samples": 50802,
7
+ "train_samples_per_second": 16.803,
8
+ "train_steps_per_second": 0.116
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,793 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 14.269521713256836,
3
+ "best_model_checkpoint": "./qwen2.5-0.5b/qwen2.5-0.5b-expo-L2EXPO-ES-0.1-W0/checkpoint-850",
4
+ "epoch": 3.119981105337742,
5
+ "eval_steps": 50,
6
+ "global_step": 1100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "dpo_loss": 0.6931471824645996,
13
+ "epoch": 0.002834199338686821,
14
+ "grad_norm": 17829.928247930075,
15
+ "learning_rate": 2.840909090909091e-08,
16
+ "logits": -1.359458565711975,
17
+ "logps": -84.69721221923828,
18
+ "loss": 169.5214,
19
+ "objective": 153.4677734375,
20
+ "ranking_idealized": 0.6458333134651184,
21
+ "ranking_idealized_expo": 0.5833333134651184,
22
+ "ranking_simple": 0.5833333134651184,
23
+ "regularize": 0.3618059456348419,
24
+ "step": 1,
25
+ "wo_beta": 14.830931663513184
26
+ },
27
+ {
28
+ "dpo_loss": 0.6812583208084106,
29
+ "epoch": 0.14170996693434104,
30
+ "grad_norm": 16211.46107075598,
31
+ "learning_rate": 1.4204545454545458e-06,
32
+ "logits": -1.4696459770202637,
33
+ "logps": -84.0085220336914,
34
+ "loss": 176.8183,
35
+ "objective": 176.67568969726562,
36
+ "ranking_idealized": 0.608418345451355,
37
+ "ranking_idealized_expo": 0.5229591727256775,
38
+ "ranking_simple": 0.5229591727256775,
39
+ "regularize": 0.40743452310562134,
40
+ "step": 50,
41
+ "wo_beta": 15.723442077636719
42
+ },
43
+ {
44
+ "epoch": 0.14170996693434104,
45
+ "eval_dpo_loss": 0.6874236464500427,
46
+ "eval_logits": -1.4744971990585327,
47
+ "eval_logps": -93.32908630371094,
48
+ "eval_loss": 185.6754150390625,
49
+ "eval_objective": 183.7364959716797,
50
+ "eval_ranking_idealized": 0.6030020713806152,
51
+ "eval_ranking_idealized_expo": 0.5222567319869995,
52
+ "eval_ranking_simple": 0.5212215185165405,
53
+ "eval_regularize": 0.4177533686161041,
54
+ "eval_runtime": 313.3628,
55
+ "eval_samples_per_second": 18.477,
56
+ "eval_steps_per_second": 1.541,
57
+ "eval_wo_beta": 16.450912475585938,
58
+ "step": 50
59
+ },
60
+ {
61
+ "dpo_loss": 0.6344443559646606,
62
+ "epoch": 0.2834199338686821,
63
+ "grad_norm": 13549.385454159787,
64
+ "learning_rate": 2.8409090909090916e-06,
65
+ "logits": -1.4799116849899292,
66
+ "logps": -83.19428253173828,
67
+ "loss": 168.1755,
68
+ "objective": 170.85800170898438,
69
+ "ranking_idealized": 0.6016666889190674,
70
+ "ranking_idealized_expo": 0.5141666531562805,
71
+ "ranking_simple": 0.5462499856948853,
72
+ "regularize": 0.3876563012599945,
73
+ "step": 100,
74
+ "wo_beta": 15.583632469177246
75
+ },
76
+ {
77
+ "epoch": 0.2834199338686821,
78
+ "eval_dpo_loss": 0.6818934679031372,
79
+ "eval_logits": -1.4487266540527344,
80
+ "eval_logps": -93.78287506103516,
81
+ "eval_loss": 195.25457763671875,
82
+ "eval_objective": 190.24264526367188,
83
+ "eval_ranking_idealized": 0.6030020713806152,
84
+ "eval_ranking_idealized_expo": 0.5222567319869995,
85
+ "eval_ranking_simple": 0.5341615080833435,
86
+ "eval_regularize": 0.4319508373737335,
87
+ "eval_runtime": 308.147,
88
+ "eval_samples_per_second": 18.79,
89
+ "eval_steps_per_second": 1.567,
90
+ "eval_wo_beta": 16.292007446289062,
91
+ "step": 100
92
+ },
93
+ {
94
+ "dpo_loss": 0.6211874485015869,
95
+ "epoch": 0.42512990080302315,
96
+ "grad_norm": 11468.6742630064,
97
+ "learning_rate": 4.2613636363636365e-06,
98
+ "logits": -1.3584699630737305,
99
+ "logps": -84.20758056640625,
100
+ "loss": 182.7148,
101
+ "objective": 183.446044921875,
102
+ "ranking_idealized": 0.6066666841506958,
103
+ "ranking_idealized_expo": 0.5287500023841858,
104
+ "ranking_simple": 0.5704166889190674,
105
+ "regularize": 0.408563494682312,
106
+ "step": 150,
107
+ "wo_beta": 15.468416213989258
108
+ },
109
+ {
110
+ "epoch": 0.42512990080302315,
111
+ "eval_dpo_loss": 0.696851909160614,
112
+ "eval_logits": -1.2730469703674316,
113
+ "eval_logps": -89.73287200927734,
114
+ "eval_loss": 218.5299072265625,
115
+ "eval_objective": 213.48837280273438,
116
+ "eval_ranking_idealized": 0.6030020713806152,
117
+ "eval_ranking_idealized_expo": 0.5222567319869995,
118
+ "eval_ranking_simple": 0.533643901348114,
119
+ "eval_regularize": 0.4859408438205719,
120
+ "eval_runtime": 310.345,
121
+ "eval_samples_per_second": 18.657,
122
+ "eval_steps_per_second": 1.556,
123
+ "eval_wo_beta": 15.804472923278809,
124
+ "step": 150
125
+ },
126
+ {
127
+ "dpo_loss": 0.6188575029373169,
128
+ "epoch": 0.5668398677373642,
129
+ "grad_norm": 9821.481225782692,
130
+ "learning_rate": 4.997168347957521e-06,
131
+ "logits": -1.112230896949768,
132
+ "logps": -78.28816223144531,
133
+ "loss": 203.0993,
134
+ "objective": 197.34710693359375,
135
+ "ranking_idealized": 0.5924999713897705,
136
+ "ranking_idealized_expo": 0.5166666507720947,
137
+ "ranking_simple": 0.5529166460037231,
138
+ "regularize": 0.43580684065818787,
139
+ "step": 200,
140
+ "wo_beta": 15.601910591125488
141
+ },
142
+ {
143
+ "epoch": 0.5668398677373642,
144
+ "eval_dpo_loss": 0.7050652503967285,
145
+ "eval_logits": -1.0446792840957642,
146
+ "eval_logps": -79.70624542236328,
147
+ "eval_loss": 251.92431640625,
148
+ "eval_objective": 242.6405792236328,
149
+ "eval_ranking_idealized": 0.6030020713806152,
150
+ "eval_ranking_idealized_expo": 0.5222567319869995,
151
+ "eval_ranking_simple": 0.532608687877655,
152
+ "eval_regularize": 0.5518472790718079,
153
+ "eval_runtime": 309.0635,
154
+ "eval_samples_per_second": 18.734,
155
+ "eval_steps_per_second": 1.563,
156
+ "eval_wo_beta": 14.694934844970703,
157
+ "step": 200
158
+ },
159
+ {
160
+ "dpo_loss": 0.6134654879570007,
161
+ "epoch": 0.7085498346717053,
162
+ "grad_norm": 9089.685589774383,
163
+ "learning_rate": 4.973122855144066e-06,
164
+ "logits": -0.831710934638977,
165
+ "logps": -75.95425415039062,
166
+ "loss": 207.5481,
167
+ "objective": 209.42929077148438,
168
+ "ranking_idealized": 0.5991666913032532,
169
+ "ranking_idealized_expo": 0.5170833468437195,
170
+ "ranking_simple": 0.5600000023841858,
171
+ "regularize": 0.46070510149002075,
172
+ "step": 250,
173
+ "wo_beta": 15.946144104003906
174
+ },
175
+ {
176
+ "epoch": 0.7085498346717053,
177
+ "eval_dpo_loss": 0.7055123448371887,
178
+ "eval_logits": -1.0362330675125122,
179
+ "eval_logps": -80.09400177001953,
180
+ "eval_loss": 251.99049377441406,
181
+ "eval_objective": 244.35101318359375,
182
+ "eval_ranking_idealized": 0.6030020713806152,
183
+ "eval_ranking_idealized_expo": 0.5222567319869995,
184
+ "eval_ranking_simple": 0.5305383205413818,
185
+ "eval_regularize": 0.5541524887084961,
186
+ "eval_runtime": 308.2828,
187
+ "eval_samples_per_second": 18.781,
188
+ "eval_steps_per_second": 1.567,
189
+ "eval_wo_beta": 14.815776824951172,
190
+ "step": 250
191
+ },
192
+ {
193
+ "dpo_loss": 0.6043393015861511,
194
+ "epoch": 0.8502598016060463,
195
+ "grad_norm": 8539.808891034761,
196
+ "learning_rate": 4.924776641419513e-06,
197
+ "logits": -0.7953253388404846,
198
+ "logps": -75.85888671875,
199
+ "loss": 193.4843,
200
+ "objective": 194.79408264160156,
201
+ "ranking_idealized": 0.5799999833106995,
202
+ "ranking_idealized_expo": 0.4970833361148834,
203
+ "ranking_simple": 0.5475000143051147,
204
+ "regularize": 0.43644893169403076,
205
+ "step": 300,
206
+ "wo_beta": 15.487529754638672
207
+ },
208
+ {
209
+ "epoch": 0.8502598016060463,
210
+ "eval_dpo_loss": 0.7149869799613953,
211
+ "eval_logits": -0.7137008905410767,
212
+ "eval_logps": -80.42964172363281,
213
+ "eval_loss": 266.710693359375,
214
+ "eval_objective": 258.3957214355469,
215
+ "eval_ranking_idealized": 0.6030020713806152,
216
+ "eval_ranking_idealized_expo": 0.5222567319869995,
217
+ "eval_ranking_simple": 0.5289855003356934,
218
+ "eval_regularize": 0.5880751013755798,
219
+ "eval_runtime": 308.0939,
220
+ "eval_samples_per_second": 18.793,
221
+ "eval_steps_per_second": 1.568,
222
+ "eval_wo_beta": 14.543147087097168,
223
+ "step": 300
224
+ },
225
+ {
226
+ "dpo_loss": 0.5903122425079346,
227
+ "epoch": 0.9919697685403873,
228
+ "grad_norm": 7990.460628234389,
229
+ "learning_rate": 4.8526047530778175e-06,
230
+ "logits": -0.5379437208175659,
231
+ "logps": -75.28119659423828,
232
+ "loss": 182.6922,
233
+ "objective": 183.501220703125,
234
+ "ranking_idealized": 0.60916668176651,
235
+ "ranking_idealized_expo": 0.5270833373069763,
236
+ "ranking_simple": 0.5887500047683716,
237
+ "regularize": 0.4042646884918213,
238
+ "step": 350,
239
+ "wo_beta": 15.643413543701172
240
+ },
241
+ {
242
+ "epoch": 0.9919697685403873,
243
+ "eval_dpo_loss": 0.7073115706443787,
244
+ "eval_logits": -0.6448346972465515,
245
+ "eval_logps": -76.36376190185547,
246
+ "eval_loss": 262.3345947265625,
247
+ "eval_objective": 254.63600158691406,
248
+ "eval_ranking_idealized": 0.6030020713806152,
249
+ "eval_ranking_idealized_expo": 0.5222567319869995,
250
+ "eval_ranking_simple": 0.5357142686843872,
251
+ "eval_regularize": 0.5802178978919983,
252
+ "eval_runtime": 307.6386,
253
+ "eval_samples_per_second": 18.821,
254
+ "eval_steps_per_second": 1.57,
255
+ "eval_wo_beta": 14.617557525634766,
256
+ "step": 350
257
+ },
258
+ {
259
+ "dpo_loss": 0.5762209892272949,
260
+ "epoch": 1.1336797354747283,
261
+ "grad_norm": 8197.08929637491,
262
+ "learning_rate": 4.757316345716554e-06,
263
+ "logits": -0.5033857226371765,
264
+ "logps": -73.0711441040039,
265
+ "loss": 166.9683,
266
+ "objective": 169.2793426513672,
267
+ "ranking_idealized": 0.6087499856948853,
268
+ "ranking_idealized_expo": 0.5337499976158142,
269
+ "ranking_simple": 0.5870833396911621,
270
+ "regularize": 0.3724302351474762,
271
+ "step": 400,
272
+ "wo_beta": 15.946381568908691
273
+ },
274
+ {
275
+ "epoch": 1.1336797354747283,
276
+ "eval_dpo_loss": 0.7151855230331421,
277
+ "eval_logits": -0.6391506791114807,
278
+ "eval_logps": -78.34819793701172,
279
+ "eval_loss": 272.3287658691406,
280
+ "eval_objective": 264.91107177734375,
281
+ "eval_ranking_idealized": 0.6030020713806152,
282
+ "eval_ranking_idealized_expo": 0.5222567319869995,
283
+ "eval_ranking_simple": 0.5274327397346497,
284
+ "eval_regularize": 0.6056078672409058,
285
+ "eval_runtime": 308.6832,
286
+ "eval_samples_per_second": 18.757,
287
+ "eval_steps_per_second": 1.565,
288
+ "eval_wo_beta": 14.651263236999512,
289
+ "step": 400
290
+ },
291
+ {
292
+ "dpo_loss": 0.5697375535964966,
293
+ "epoch": 1.2753897024090695,
294
+ "grad_norm": 7987.082885512546,
295
+ "learning_rate": 4.639847716126855e-06,
296
+ "logits": -0.49218180775642395,
297
+ "logps": -75.01708984375,
298
+ "loss": 155.9364,
299
+ "objective": 154.53318786621094,
300
+ "ranking_idealized": 0.5975000262260437,
301
+ "ranking_idealized_expo": 0.5199999809265137,
302
+ "ranking_simple": 0.5849999785423279,
303
+ "regularize": 0.34387490153312683,
304
+ "step": 450,
305
+ "wo_beta": 16.392471313476562
306
+ },
307
+ {
308
+ "epoch": 1.2753897024090695,
309
+ "eval_dpo_loss": 0.7185649871826172,
310
+ "eval_logits": -0.4207253158092499,
311
+ "eval_logps": -80.52300262451172,
312
+ "eval_loss": 275.0490417480469,
313
+ "eval_objective": 268.8637390136719,
314
+ "eval_ranking_idealized": 0.6030020713806152,
315
+ "eval_ranking_idealized_expo": 0.5222567319869995,
316
+ "eval_ranking_simple": 0.5320910811424255,
317
+ "eval_regularize": 0.6128751039505005,
318
+ "eval_runtime": 308.3205,
319
+ "eval_samples_per_second": 18.779,
320
+ "eval_steps_per_second": 1.567,
321
+ "eval_wo_beta": 14.77772331237793,
322
+ "step": 450
323
+ },
324
+ {
325
+ "dpo_loss": 0.5621860027313232,
326
+ "epoch": 1.4170996693434104,
327
+ "grad_norm": 8470.747695660142,
328
+ "learning_rate": 4.501353102310901e-06,
329
+ "logits": -0.3975617587566376,
330
+ "logps": -76.15510559082031,
331
+ "loss": 143.4724,
332
+ "objective": 144.99127197265625,
333
+ "ranking_idealized": 0.57833331823349,
334
+ "ranking_idealized_expo": 0.4983333349227905,
335
+ "ranking_simple": 0.5641666650772095,
336
+ "regularize": 0.3229096531867981,
337
+ "step": 500,
338
+ "wo_beta": 15.915486335754395
339
+ },
340
+ {
341
+ "epoch": 1.4170996693434104,
342
+ "eval_dpo_loss": 0.720949113368988,
343
+ "eval_logits": -0.5140562653541565,
344
+ "eval_logps": -80.55872344970703,
345
+ "eval_loss": 275.96630859375,
346
+ "eval_objective": 270.0382995605469,
347
+ "eval_ranking_idealized": 0.6030020713806152,
348
+ "eval_ranking_idealized_expo": 0.5222567319869995,
349
+ "eval_ranking_simple": 0.5269151329994202,
350
+ "eval_regularize": 0.6150393486022949,
351
+ "eval_runtime": 308.3699,
352
+ "eval_samples_per_second": 18.776,
353
+ "eval_steps_per_second": 1.566,
354
+ "eval_wo_beta": 14.436367988586426,
355
+ "step": 500
356
+ },
357
+ {
358
+ "dpo_loss": 0.5593892931938171,
359
+ "epoch": 1.5588096362777515,
360
+ "grad_norm": 8086.109814509309,
361
+ "learning_rate": 4.34319334202531e-06,
362
+ "logits": -0.5436981320381165,
363
+ "logps": -76.949462890625,
364
+ "loss": 141.3444,
365
+ "objective": 144.28431701660156,
366
+ "ranking_idealized": 0.5945833325386047,
367
+ "ranking_idealized_expo": 0.5116666555404663,
368
+ "ranking_simple": 0.5841666460037231,
369
+ "regularize": 0.3211959898471832,
370
+ "step": 550,
371
+ "wo_beta": 15.494115829467773
372
+ },
373
+ {
374
+ "epoch": 1.5588096362777515,
375
+ "eval_dpo_loss": 0.7139093279838562,
376
+ "eval_logits": -0.6337844729423523,
377
+ "eval_logps": -81.12705993652344,
378
+ "eval_loss": 275.0850524902344,
379
+ "eval_objective": 269.2188720703125,
380
+ "eval_ranking_idealized": 0.6030020713806152,
381
+ "eval_ranking_idealized_expo": 0.5222567319869995,
382
+ "eval_ranking_simple": 0.5377846956253052,
383
+ "eval_regularize": 0.6158860325813293,
384
+ "eval_runtime": 309.36,
385
+ "eval_samples_per_second": 18.716,
386
+ "eval_steps_per_second": 1.561,
387
+ "eval_wo_beta": 14.642486572265625,
388
+ "step": 550
389
+ },
390
+ {
391
+ "dpo_loss": 0.5621269345283508,
392
+ "epoch": 1.7028814359943316,
393
+ "grad_norm": 7358.036337966711,
394
+ "learning_rate": 4.16692250129073e-06,
395
+ "logits": -0.5125311613082886,
396
+ "logps": -76.2679443359375,
397
+ "loss": 136.172,
398
+ "objective": 138.94873046875,
399
+ "ranking_idealized": 0.6000000238418579,
400
+ "ranking_idealized_expo": 0.51583331823349,
401
+ "ranking_simple": 0.5874999761581421,
402
+ "regularize": 0.304604709148407,
403
+ "step": 600,
404
+ "wo_beta": 15.486053466796875
405
+ },
406
+ {
407
+ "epoch": 1.7028814359943316,
408
+ "eval_dpo_loss": 0.711104691028595,
409
+ "eval_logits": -0.5857469439506531,
410
+ "eval_logps": -79.42213439941406,
411
+ "eval_loss": 273.6680908203125,
412
+ "eval_objective": 264.6509704589844,
413
+ "eval_ranking_idealized": 0.6030020713806152,
414
+ "eval_ranking_idealized_expo": 0.5222567319869995,
415
+ "eval_ranking_simple": 0.5372670888900757,
416
+ "eval_regularize": 0.6011682748794556,
417
+ "eval_runtime": 311.317,
418
+ "eval_samples_per_second": 18.598,
419
+ "eval_steps_per_second": 1.551,
420
+ "eval_wo_beta": 14.563067436218262,
421
+ "step": 600
422
+ },
423
+ {
424
+ "dpo_loss": 0.5565243363380432,
425
+ "epoch": 1.8445914029286725,
426
+ "grad_norm": 7716.658918543234,
427
+ "learning_rate": 3.974272604254906e-06,
428
+ "logits": -0.45386597514152527,
429
+ "logps": -76.72352600097656,
430
+ "loss": 130.7133,
431
+ "objective": 137.35403442382812,
432
+ "ranking_idealized": 0.6075000166893005,
433
+ "ranking_idealized_expo": 0.5308333039283752,
434
+ "ranking_simple": 0.5920833349227905,
435
+ "regularize": 0.2999296486377716,
436
+ "step": 650,
437
+ "wo_beta": 16.43378448486328
438
+ },
439
+ {
440
+ "epoch": 1.8445914029286725,
441
+ "eval_dpo_loss": 0.7193200588226318,
442
+ "eval_logits": -0.4215196371078491,
443
+ "eval_logps": -80.21295166015625,
444
+ "eval_loss": 276.3609313964844,
445
+ "eval_objective": 269.6939392089844,
446
+ "eval_ranking_idealized": 0.6030020713806152,
447
+ "eval_ranking_idealized_expo": 0.5222567319869995,
448
+ "eval_ranking_simple": 0.5341615080833435,
449
+ "eval_regularize": 0.614128828048706,
450
+ "eval_runtime": 309.0561,
451
+ "eval_samples_per_second": 18.734,
452
+ "eval_steps_per_second": 1.563,
453
+ "eval_wo_beta": 14.545560836791992,
454
+ "step": 650
455
+ },
456
+ {
457
+ "dpo_loss": 0.5614636540412903,
458
+ "epoch": 1.9863013698630136,
459
+ "grad_norm": 6878.133616740314,
460
+ "learning_rate": 3.767136614452458e-06,
461
+ "logits": -0.4337880313396454,
462
+ "logps": -76.46927642822266,
463
+ "loss": 122.624,
464
+ "objective": 121.14984130859375,
465
+ "ranking_idealized": 0.5933333039283752,
466
+ "ranking_idealized_expo": 0.5104166865348816,
467
+ "ranking_simple": 0.5766666531562805,
468
+ "regularize": 0.27313005924224854,
469
+ "step": 700,
470
+ "wo_beta": 15.872406005859375
471
+ },
472
+ {
473
+ "epoch": 1.9863013698630136,
474
+ "eval_dpo_loss": 0.717802107334137,
475
+ "eval_logits": -0.5263403058052063,
476
+ "eval_logps": -80.99676513671875,
477
+ "eval_loss": 278.4689636230469,
478
+ "eval_objective": 271.4756774902344,
479
+ "eval_ranking_idealized": 0.6030020713806152,
480
+ "eval_ranking_idealized_expo": 0.5222567319869995,
481
+ "eval_ranking_simple": 0.5377846956253052,
482
+ "eval_regularize": 0.6189925074577332,
483
+ "eval_runtime": 308.2823,
484
+ "eval_samples_per_second": 18.781,
485
+ "eval_steps_per_second": 1.567,
486
+ "eval_wo_beta": 14.466358184814453,
487
+ "step": 700
488
+ },
489
+ {
490
+ "dpo_loss": 0.5429660677909851,
491
+ "epoch": 2.1280113367973548,
492
+ "grad_norm": 7099.084874469886,
493
+ "learning_rate": 3.547549834686222e-06,
494
+ "logits": -0.456377774477005,
495
+ "logps": -78.5710678100586,
496
+ "loss": 108.7022,
497
+ "objective": 107.37686157226562,
498
+ "ranking_idealized": 0.6087499856948853,
499
+ "ranking_idealized_expo": 0.5170833468437195,
500
+ "ranking_simple": 0.5899999737739563,
501
+ "regularize": 0.2386901080608368,
502
+ "step": 750,
503
+ "wo_beta": 16.154056549072266
504
+ },
505
+ {
506
+ "epoch": 2.1280113367973548,
507
+ "eval_dpo_loss": 0.7206873297691345,
508
+ "eval_logits": -0.4656512439250946,
509
+ "eval_logps": -84.00879669189453,
510
+ "eval_loss": 282.56683349609375,
511
+ "eval_objective": 276.0200500488281,
512
+ "eval_ranking_idealized": 0.6030020713806152,
513
+ "eval_ranking_idealized_expo": 0.5222567319869995,
514
+ "eval_ranking_simple": 0.534679114818573,
515
+ "eval_regularize": 0.6302040815353394,
516
+ "eval_runtime": 308.3238,
517
+ "eval_samples_per_second": 18.779,
518
+ "eval_steps_per_second": 1.567,
519
+ "eval_wo_beta": 14.451681137084961,
520
+ "step": 750
521
+ },
522
+ {
523
+ "dpo_loss": 0.5408411026000977,
524
+ "epoch": 2.269721303731696,
525
+ "grad_norm": 7050.1218576698175,
526
+ "learning_rate": 3.3176699082935546e-06,
527
+ "logits": -0.44169890880584717,
528
+ "logps": -79.25837707519531,
529
+ "loss": 104.1923,
530
+ "objective": 101.39546966552734,
531
+ "ranking_idealized": 0.5924999713897705,
532
+ "ranking_idealized_expo": 0.5095833539962769,
533
+ "ranking_simple": 0.5879166722297668,
534
+ "regularize": 0.22695893049240112,
535
+ "step": 800,
536
+ "wo_beta": 15.370372772216797
537
+ },
538
+ {
539
+ "epoch": 2.269721303731696,
540
+ "eval_dpo_loss": 0.7166009545326233,
541
+ "eval_logits": -0.46399185061454773,
542
+ "eval_logps": -81.63129425048828,
543
+ "eval_loss": 278.05548095703125,
544
+ "eval_objective": 272.76220703125,
545
+ "eval_ranking_idealized": 0.6030020713806152,
546
+ "eval_ranking_idealized_expo": 0.5222567319869995,
547
+ "eval_ranking_simple": 0.5383023023605347,
548
+ "eval_regularize": 0.6210389137268066,
549
+ "eval_runtime": 316.0101,
550
+ "eval_samples_per_second": 18.322,
551
+ "eval_steps_per_second": 1.528,
552
+ "eval_wo_beta": 14.43066692352295,
553
+ "step": 800
554
+ },
555
+ {
556
+ "dpo_loss": 0.5408244132995605,
557
+ "epoch": 2.411431270666037,
558
+ "grad_norm": 6732.756367827086,
559
+ "learning_rate": 3.0797556183036582e-06,
560
+ "logits": -0.398872435092926,
561
+ "logps": -77.83311462402344,
562
+ "loss": 99.0867,
563
+ "objective": 97.56111145019531,
564
+ "ranking_idealized": 0.5975000262260437,
565
+ "ranking_idealized_expo": 0.5166666507720947,
566
+ "ranking_simple": 0.5879166722297668,
567
+ "regularize": 0.21405553817749023,
568
+ "step": 850,
569
+ "wo_beta": 15.753022193908691
570
+ },
571
+ {
572
+ "epoch": 2.411431270666037,
573
+ "eval_dpo_loss": 0.7208675742149353,
574
+ "eval_logits": -0.5211770534515381,
575
+ "eval_logps": -81.06893157958984,
576
+ "eval_loss": 283.0077819824219,
577
+ "eval_objective": 277.3703308105469,
578
+ "eval_ranking_idealized": 0.6030020713806152,
579
+ "eval_ranking_idealized_expo": 0.5222567319869995,
580
+ "eval_ranking_simple": 0.5331262946128845,
581
+ "eval_regularize": 0.6309527158737183,
582
+ "eval_runtime": 308.4936,
583
+ "eval_samples_per_second": 18.769,
584
+ "eval_steps_per_second": 1.566,
585
+ "eval_wo_beta": 14.269521713256836,
586
+ "step": 850
587
+ },
588
+ {
589
+ "dpo_loss": 0.5474214553833008,
590
+ "epoch": 2.5531412376003777,
591
+ "grad_norm": 6765.065529775399,
592
+ "learning_rate": 2.8361446928038298e-06,
593
+ "logits": -0.45377472043037415,
594
+ "logps": -77.1721420288086,
595
+ "loss": 91.7475,
596
+ "objective": 90.64097595214844,
597
+ "ranking_idealized": 0.5887500047683716,
598
+ "ranking_idealized_expo": 0.5191666483879089,
599
+ "ranking_simple": 0.5808333158493042,
600
+ "regularize": 0.20561054348945618,
601
+ "step": 900,
602
+ "wo_beta": 16.313594818115234
603
+ },
604
+ {
605
+ "epoch": 2.5531412376003777,
606
+ "eval_dpo_loss": 0.7200337052345276,
607
+ "eval_logits": -0.5149067044258118,
608
+ "eval_logps": -81.61436462402344,
609
+ "eval_loss": 279.6676025390625,
610
+ "eval_objective": 275.1768798828125,
611
+ "eval_ranking_idealized": 0.6030020713806152,
612
+ "eval_ranking_idealized_expo": 0.5222567319869995,
613
+ "eval_ranking_simple": 0.5372670888900757,
614
+ "eval_regularize": 0.6278803944587708,
615
+ "eval_runtime": 308.2613,
616
+ "eval_samples_per_second": 18.783,
617
+ "eval_steps_per_second": 1.567,
618
+ "eval_wo_beta": 14.35700798034668,
619
+ "step": 900
620
+ },
621
+ {
622
+ "dpo_loss": 0.536076545715332,
623
+ "epoch": 2.694851204534719,
624
+ "grad_norm": 6780.927379854703,
625
+ "learning_rate": 2.5892308345974517e-06,
626
+ "logits": -0.3948451280593872,
627
+ "logps": -78.59798431396484,
628
+ "loss": 87.8681,
629
+ "objective": 91.65276336669922,
630
+ "ranking_idealized": 0.5983333587646484,
631
+ "ranking_idealized_expo": 0.5087500214576721,
632
+ "ranking_simple": 0.5895833373069763,
633
+ "regularize": 0.1982572227716446,
634
+ "step": 950,
635
+ "wo_beta": 16.053390502929688
636
+ },
637
+ {
638
+ "epoch": 2.694851204534719,
639
+ "eval_dpo_loss": 0.7190758585929871,
640
+ "eval_logits": -0.44280436635017395,
641
+ "eval_logps": -81.85442352294922,
642
+ "eval_loss": 281.5718078613281,
643
+ "eval_objective": 275.7560119628906,
644
+ "eval_ranking_idealized": 0.6030020713806152,
645
+ "eval_ranking_idealized_expo": 0.5222567319869995,
646
+ "eval_ranking_simple": 0.5362318754196167,
647
+ "eval_regularize": 0.6276679039001465,
648
+ "eval_runtime": 308.5458,
649
+ "eval_samples_per_second": 18.765,
650
+ "eval_steps_per_second": 1.565,
651
+ "eval_wo_beta": 14.350942611694336,
652
+ "step": 950
653
+ },
654
+ {
655
+ "dpo_loss": 0.5394951105117798,
656
+ "epoch": 2.83656117146906,
657
+ "grad_norm": 7326.8730595182415,
658
+ "learning_rate": 2.341440200858589e-06,
659
+ "logits": -0.389419287443161,
660
+ "logps": -77.48452758789062,
661
+ "loss": 81.742,
662
+ "objective": 80.39604187011719,
663
+ "ranking_idealized": 0.6016666889190674,
664
+ "ranking_idealized_expo": 0.5099999904632568,
665
+ "ranking_simple": 0.5883333086967468,
666
+ "regularize": 0.18344084918498993,
667
+ "step": 1000,
668
+ "wo_beta": 15.534008979797363
669
+ },
670
+ {
671
+ "epoch": 2.83656117146906,
672
+ "eval_dpo_loss": 0.7197140455245972,
673
+ "eval_logits": -0.49511978030204773,
674
+ "eval_logps": -81.44123077392578,
675
+ "eval_loss": 279.1323547363281,
676
+ "eval_objective": 274.5647277832031,
677
+ "eval_ranking_idealized": 0.6030020713806152,
678
+ "eval_ranking_idealized_expo": 0.5222567319869995,
679
+ "eval_ranking_simple": 0.533643901348114,
680
+ "eval_regularize": 0.6257451176643372,
681
+ "eval_runtime": 309.1776,
682
+ "eval_samples_per_second": 18.727,
683
+ "eval_steps_per_second": 1.562,
684
+ "eval_wo_beta": 14.355087280273438,
685
+ "step": 1000
686
+ },
687
+ {
688
+ "dpo_loss": 0.5420763492584229,
689
+ "epoch": 2.978271138403401,
690
+ "grad_norm": 6821.455863031989,
691
+ "learning_rate": 2.0952075638923656e-06,
692
+ "logits": -0.4298148453235626,
693
+ "logps": -78.06723022460938,
694
+ "loss": 76.4372,
695
+ "objective": 76.12059783935547,
696
+ "ranking_idealized": 0.6083333492279053,
697
+ "ranking_idealized_expo": 0.518750011920929,
698
+ "ranking_simple": 0.5958333611488342,
699
+ "regularize": 0.1691569834947586,
700
+ "step": 1050,
701
+ "wo_beta": 16.169239044189453
702
+ },
703
+ {
704
+ "epoch": 2.978271138403401,
705
+ "eval_dpo_loss": 0.7184009552001953,
706
+ "eval_logits": -0.4501994550228119,
707
+ "eval_logps": -82.39596557617188,
708
+ "eval_loss": 279.18841552734375,
709
+ "eval_objective": 273.9026184082031,
710
+ "eval_ranking_idealized": 0.6030020713806152,
711
+ "eval_ranking_idealized_expo": 0.5222567319869995,
712
+ "eval_ranking_simple": 0.533643901348114,
713
+ "eval_regularize": 0.6249070167541504,
714
+ "eval_runtime": 310.2594,
715
+ "eval_samples_per_second": 18.662,
716
+ "eval_steps_per_second": 1.557,
717
+ "eval_wo_beta": 14.32027816772461,
718
+ "step": 1050
719
+ },
720
+ {
721
+ "dpo_loss": 0.5327512621879578,
722
+ "epoch": 3.119981105337742,
723
+ "grad_norm": 6539.491492464407,
724
+ "learning_rate": 1.852952387243698e-06,
725
+ "logits": -0.39219775795936584,
726
+ "logps": -78.88080596923828,
727
+ "loss": 67.4698,
728
+ "objective": 66.67189025878906,
729
+ "ranking_idealized": 0.6045833230018616,
730
+ "ranking_idealized_expo": 0.528333306312561,
731
+ "ranking_simple": 0.6029166579246521,
732
+ "regularize": 0.14756560325622559,
733
+ "step": 1100,
734
+ "wo_beta": 16.35073471069336
735
+ },
736
+ {
737
+ "epoch": 3.119981105337742,
738
+ "eval_dpo_loss": 0.7169165015220642,
739
+ "eval_logits": -0.41896748542785645,
740
+ "eval_logps": -82.91073608398438,
741
+ "eval_loss": 280.53173828125,
742
+ "eval_objective": 274.793212890625,
743
+ "eval_ranking_idealized": 0.6030020713806152,
744
+ "eval_ranking_idealized_expo": 0.5222567319869995,
745
+ "eval_ranking_simple": 0.532608687877655,
746
+ "eval_regularize": 0.6259706616401672,
747
+ "eval_runtime": 310.4871,
748
+ "eval_samples_per_second": 18.648,
749
+ "eval_steps_per_second": 1.556,
750
+ "eval_wo_beta": 14.341768264770508,
751
+ "step": 1100
752
+ },
753
+ {
754
+ "epoch": 3.119981105337742,
755
+ "step": 1100,
756
+ "total_flos": 0.0,
757
+ "train_loss": 50.30704811789773,
758
+ "train_runtime": 15116.5529,
759
+ "train_samples_per_second": 16.803,
760
+ "train_steps_per_second": 0.116
761
+ }
762
+ ],
763
+ "logging_steps": 50,
764
+ "max_steps": 1760,
765
+ "num_input_tokens_seen": 0,
766
+ "num_train_epochs": 5,
767
+ "save_steps": 50,
768
+ "stateful_callbacks": {
769
+ "EarlyStoppingCallback": {
770
+ "args": {
771
+ "early_stopping_patience": 5,
772
+ "early_stopping_threshold": 0.0
773
+ },
774
+ "attributes": {
775
+ "early_stopping_patience_counter": 0
776
+ }
777
+ },
778
+ "TrainerControl": {
779
+ "args": {
780
+ "should_epoch_stop": false,
781
+ "should_evaluate": false,
782
+ "should_log": false,
783
+ "should_save": true,
784
+ "should_training_stop": true
785
+ },
786
+ "attributes": {}
787
+ }
788
+ },
789
+ "total_flos": 0.0,
790
+ "train_batch_size": 4,
791
+ "trial_name": null,
792
+ "trial_params": null
793
+ }