Model save
Browse files- README.md +6 -12
- all_results.json +5 -5
- config.json +1 -1
- model-00001-of-00003.safetensors +1 -1
- model-00002-of-00003.safetensors +1 -1
- model-00003-of-00003.safetensors +1 -1
- train_results.json +5 -5
- trainer_state.json +581 -1781
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -2,15 +2,9 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: TTTXXX01/All_like-zephyr-7b-sft-full
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- dpo
|
| 8 |
- generated_from_trainer
|
| 9 |
-
- trl
|
| 10 |
-
- dpo
|
| 11 |
-
- generated_from_trainer
|
| 12 |
-
datasets:
|
| 13 |
-
- HuggingFaceH4/ultrafeedback_binarized
|
| 14 |
model-index:
|
| 15 |
- name: IL_BrierAll-7b-sft-full
|
| 16 |
results: []
|
|
@@ -21,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 21 |
|
| 22 |
# IL_BrierAll-7b-sft-full
|
| 23 |
|
| 24 |
-
This model is a fine-tuned version of [TTTXXX01/All_like-zephyr-7b-sft-full](https://huggingface.co/TTTXXX01/All_like-zephyr-7b-sft-full) on
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -41,13 +35,13 @@ More information needed
|
|
| 41 |
|
| 42 |
The following hyperparameters were used during training:
|
| 43 |
- learning_rate: 5e-07
|
| 44 |
-
- train_batch_size:
|
| 45 |
- eval_batch_size: 4
|
| 46 |
- seed: 42
|
| 47 |
- distributed_type: multi-GPU
|
| 48 |
- num_devices: 4
|
| 49 |
-
- gradient_accumulation_steps:
|
| 50 |
-
- total_train_batch_size:
|
| 51 |
- total_eval_batch_size: 16
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: cosine
|
|
@@ -61,6 +55,6 @@ The following hyperparameters were used during training:
|
|
| 61 |
### Framework versions
|
| 62 |
|
| 63 |
- Transformers 4.41.2
|
| 64 |
-
- Pytorch 2.
|
| 65 |
-
- Datasets 2.19.
|
| 66 |
- Tokenizers 0.19.1
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: TTTXXX01/All_like-zephyr-7b-sft-full
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- dpo
|
| 7 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: IL_BrierAll-7b-sft-full
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# IL_BrierAll-7b-sft-full
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [TTTXXX01/All_like-zephyr-7b-sft-full](https://huggingface.co/TTTXXX01/All_like-zephyr-7b-sft-full) on an unknown dataset.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
- learning_rate: 5e-07
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
- eval_batch_size: 4
|
| 40 |
- seed: 42
|
| 41 |
- distributed_type: multi-GPU
|
| 42 |
- num_devices: 4
|
| 43 |
+
- gradient_accumulation_steps: 8
|
| 44 |
+
- total_train_batch_size: 128
|
| 45 |
- total_eval_batch_size: 16
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: cosine
|
|
|
|
| 55 |
### Framework versions
|
| 56 |
|
| 57 |
- Transformers 4.41.2
|
| 58 |
+
- Pytorch 2.3.0+cu121
|
| 59 |
+
- Datasets 2.19.1
|
| 60 |
- Tokenizers 0.19.1
|
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"total_flos": 0.0,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 61135,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.998691442030882,
|
| 3 |
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.4856958399278813,
|
| 5 |
+
"train_runtime": 14922.9586,
|
| 6 |
"train_samples": 61135,
|
| 7 |
+
"train_samples_per_second": 4.097,
|
| 8 |
+
"train_steps_per_second": 0.032
|
| 9 |
}
|
config.json
CHANGED
|
@@ -21,6 +21,6 @@
|
|
| 21 |
"tie_word_embeddings": false,
|
| 22 |
"torch_dtype": "bfloat16",
|
| 23 |
"transformers_version": "4.41.2",
|
| 24 |
-
"use_cache":
|
| 25 |
"vocab_size": 32000
|
| 26 |
}
|
|
|
|
| 21 |
"tie_word_embeddings": false,
|
| 22 |
"torch_dtype": "bfloat16",
|
| 23 |
"transformers_version": "4.41.2",
|
| 24 |
+
"use_cache": false,
|
| 25 |
"vocab_size": 32000
|
| 26 |
}
|
model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4943162336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:925cf8eabacd6f9f2b9aa170e7cb2405450045b5697f3246bb1e2dfec02bbd80
|
| 3 |
size 4943162336
|
model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999819336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0162c86a91191d1c508c75efc8e079c5eb1bbb1bcf370e395b457968ab14f71b
|
| 3 |
size 4999819336
|
model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4540516344
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe3c79bf6fe2dfb8817d57795e9a13f191380f75cddc5bf0c1599b3ba68697bb
|
| 3 |
size 4540516344
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"total_flos": 0.0,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 61135,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.998691442030882,
|
| 3 |
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.4856958399278813,
|
| 5 |
+
"train_runtime": 14922.9586,
|
| 6 |
"train_samples": 61135,
|
| 7 |
+
"train_samples_per_second": 4.097,
|
| 8 |
+
"train_steps_per_second": 0.032
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,21 +1,21 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"logits/chosen":
|
| 16 |
-
"logits/rejected":
|
| 17 |
-
"logps/chosen": -
|
| 18 |
-
"logps/rejected": -
|
| 19 |
"loss": 0.5,
|
| 20 |
"rewards/accuracies": 0.0,
|
| 21 |
"rewards/chosen": 0.0,
|
|
@@ -24,1922 +24,722 @@
|
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"logits/chosen":
|
| 31 |
-
"logits/rejected":
|
| 32 |
-
"logps/chosen": -
|
| 33 |
-
"logps/rejected": -
|
| 34 |
"loss": 0.5,
|
| 35 |
-
"rewards/accuracies": 0.
|
| 36 |
-
"rewards/chosen":
|
| 37 |
-
"rewards/margins": 0.
|
| 38 |
-
"rewards/rejected": -0.
|
| 39 |
"step": 10
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"grad_norm":
|
| 44 |
-
"learning_rate":
|
| 45 |
-
"logits/chosen":
|
| 46 |
-
"logits/rejected":
|
| 47 |
-
"logps/chosen": -
|
| 48 |
-
"logps/rejected": -196.
|
| 49 |
-
"loss": 0.
|
| 50 |
-
"rewards/accuracies": 0.
|
| 51 |
-
"rewards/chosen":
|
| 52 |
-
"rewards/margins":
|
| 53 |
-
"rewards/rejected":
|
| 54 |
"step": 20
|
| 55 |
},
|
| 56 |
{
|
| 57 |
-
"epoch": 0.
|
| 58 |
-
"grad_norm":
|
| 59 |
-
"learning_rate":
|
| 60 |
-
"logits/chosen":
|
| 61 |
-
"logits/rejected":
|
| 62 |
-
"logps/chosen": -
|
| 63 |
-
"logps/rejected": -
|
| 64 |
-
"loss": 0.
|
| 65 |
-
"rewards/accuracies": 0.
|
| 66 |
-
"rewards/chosen": 0.
|
| 67 |
-
"rewards/margins": 0.
|
| 68 |
-
"rewards/rejected": -0.
|
| 69 |
"step": 30
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"grad_norm":
|
| 74 |
-
"learning_rate":
|
| 75 |
-
"logits/chosen":
|
| 76 |
-
"logits/rejected":
|
| 77 |
-
"logps/chosen": -
|
| 78 |
-
"logps/rejected": -
|
| 79 |
-
"loss": 0.
|
| 80 |
-
"rewards/accuracies": 0.
|
| 81 |
-
"rewards/chosen": 0.
|
| 82 |
-
"rewards/margins": 0.
|
| 83 |
-
"rewards/rejected": -0.
|
| 84 |
"step": 40
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 0.
|
| 88 |
-
"grad_norm": 2.
|
| 89 |
-
"learning_rate":
|
| 90 |
-
"logits/chosen":
|
| 91 |
-
"logits/rejected":
|
| 92 |
-
"logps/chosen": -
|
| 93 |
-
"logps/rejected": -
|
| 94 |
-
"loss": 0.
|
| 95 |
-
"rewards/accuracies": 0.
|
| 96 |
-
"rewards/chosen": 0.
|
| 97 |
-
"rewards/margins": 0.
|
| 98 |
-
"rewards/rejected": -0.
|
| 99 |
"step": 50
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
-
"grad_norm":
|
| 104 |
-
"learning_rate":
|
| 105 |
-
"logits/chosen":
|
| 106 |
-
"logits/rejected":
|
| 107 |
-
"logps/chosen": -
|
| 108 |
-
"logps/rejected": -
|
| 109 |
-
"loss": 0.
|
| 110 |
-
"rewards/accuracies": 0.
|
| 111 |
-
"rewards/chosen": 0.
|
| 112 |
-
"rewards/margins": 0.
|
| 113 |
-
"rewards/rejected": -0.
|
| 114 |
"step": 60
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 2.
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"logits/chosen":
|
| 121 |
-
"logits/rejected":
|
| 122 |
-
"logps/chosen": -
|
| 123 |
-
"logps/rejected": -
|
| 124 |
-
"loss": 0.
|
| 125 |
-
"rewards/accuracies": 0.
|
| 126 |
-
"rewards/chosen": 0.
|
| 127 |
-
"rewards/margins": 0.
|
| 128 |
-
"rewards/rejected": -0.
|
| 129 |
"step": 70
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm": 2.
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"logits/chosen":
|
| 136 |
-
"logits/rejected":
|
| 137 |
-
"logps/chosen": -
|
| 138 |
-
"logps/rejected": -
|
| 139 |
-
"loss": 0.
|
| 140 |
-
"rewards/accuracies": 0.
|
| 141 |
-
"rewards/chosen": -0.
|
| 142 |
-
"rewards/margins":
|
| 143 |
-
"rewards/rejected": -
|
| 144 |
"step": 80
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 0.
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"logits/chosen":
|
| 151 |
-
"logits/rejected":
|
| 152 |
-
"logps/chosen": -
|
| 153 |
-
"logps/rejected": -
|
| 154 |
-
"loss": 0.
|
| 155 |
-
"rewards/accuracies": 0.
|
| 156 |
-
"rewards/chosen": -
|
| 157 |
-
"rewards/margins":
|
| 158 |
-
"rewards/rejected": -
|
| 159 |
"step": 90
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 0.
|
| 163 |
-
"grad_norm":
|
| 164 |
-
"learning_rate":
|
| 165 |
-
"logits/chosen":
|
| 166 |
-
"logits/rejected":
|
| 167 |
-
"logps/chosen": -
|
| 168 |
-
"logps/rejected": -
|
| 169 |
-
"loss": 0.
|
| 170 |
-
"rewards/accuracies": 0.
|
| 171 |
-
"rewards/chosen": -
|
| 172 |
-
"rewards/margins":
|
| 173 |
-
"rewards/rejected": -
|
| 174 |
"step": 100
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 0.
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate": 4.
|
| 180 |
-
"logits/chosen":
|
| 181 |
-
"logits/rejected":
|
| 182 |
-
"logps/chosen": -
|
| 183 |
-
"logps/rejected": -
|
| 184 |
-
"loss": 0.
|
| 185 |
-
"rewards/accuracies": 0.
|
| 186 |
-
"rewards/chosen": -
|
| 187 |
-
"rewards/margins":
|
| 188 |
-
"rewards/rejected": -
|
| 189 |
"step": 110
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 0.
|
| 193 |
-
"grad_norm":
|
| 194 |
-
"learning_rate": 4.
|
| 195 |
-
"logits/chosen":
|
| 196 |
-
"logits/rejected":
|
| 197 |
-
"logps/chosen": -
|
| 198 |
-
"logps/rejected": -
|
| 199 |
-
"loss": 0.
|
| 200 |
-
"rewards/accuracies": 0.
|
| 201 |
-
"rewards/chosen": -
|
| 202 |
-
"rewards/margins":
|
| 203 |
-
"rewards/rejected": -
|
| 204 |
"step": 120
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch": 0.
|
| 208 |
-
"grad_norm":
|
| 209 |
-
"learning_rate": 4.
|
| 210 |
-
"logits/chosen":
|
| 211 |
-
"logits/rejected":
|
| 212 |
-
"logps/chosen": -
|
| 213 |
-
"logps/rejected": -
|
| 214 |
-
"loss": 0.
|
| 215 |
-
"rewards/accuracies": 0.
|
| 216 |
-
"rewards/chosen": -
|
| 217 |
-
"rewards/margins":
|
| 218 |
-
"rewards/rejected": -
|
| 219 |
"step": 130
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate": 4.
|
| 225 |
-
"logits/chosen":
|
| 226 |
-
"logits/rejected":
|
| 227 |
-
"logps/chosen": -
|
| 228 |
-
"logps/rejected": -
|
| 229 |
-
"loss": 0.
|
| 230 |
-
"rewards/accuracies": 0.
|
| 231 |
-
"rewards/chosen": -
|
| 232 |
-
"rewards/margins":
|
| 233 |
-
"rewards/rejected": -
|
| 234 |
"step": 140
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate": 4.
|
| 240 |
-
"logits/chosen":
|
| 241 |
-
"logits/rejected":
|
| 242 |
-
"logps/chosen": -
|
| 243 |
-
"logps/rejected": -
|
| 244 |
-
"loss": 0.
|
| 245 |
-
"rewards/accuracies": 0.
|
| 246 |
-
"rewards/chosen": -
|
| 247 |
-
"rewards/margins":
|
| 248 |
-
"rewards/rejected": -
|
| 249 |
"step": 150
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 0.
|
| 253 |
-
"grad_norm": 3.
|
| 254 |
-
"learning_rate": 4.
|
| 255 |
-
"logits/chosen":
|
| 256 |
-
"logits/rejected":
|
| 257 |
-
"logps/chosen": -
|
| 258 |
-
"logps/rejected": -
|
| 259 |
-
"loss": 0.
|
| 260 |
-
"rewards/accuracies": 0.
|
| 261 |
-
"rewards/chosen": -
|
| 262 |
-
"rewards/margins":
|
| 263 |
-
"rewards/rejected": -
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch": 0.
|
| 268 |
-
"grad_norm": 3.
|
| 269 |
-
"learning_rate": 4.
|
| 270 |
-
"logits/chosen":
|
| 271 |
-
"logits/rejected":
|
| 272 |
-
"logps/chosen": -
|
| 273 |
-
"logps/rejected": -
|
| 274 |
-
"loss": 0.
|
| 275 |
-
"rewards/accuracies": 0.
|
| 276 |
-
"rewards/chosen": -
|
| 277 |
-
"rewards/margins":
|
| 278 |
-
"rewards/rejected": -
|
| 279 |
"step": 170
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch": 0.
|
| 283 |
-
"grad_norm":
|
| 284 |
-
"learning_rate":
|
| 285 |
-
"logits/chosen":
|
| 286 |
-
"logits/rejected":
|
| 287 |
-
"logps/chosen": -
|
| 288 |
-
"logps/rejected": -
|
| 289 |
-
"loss": 0.
|
| 290 |
-
"rewards/accuracies": 0.
|
| 291 |
-
"rewards/chosen": -
|
| 292 |
-
"rewards/margins":
|
| 293 |
-
"rewards/rejected": -
|
| 294 |
"step": 180
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch": 0.
|
| 298 |
-
"grad_norm":
|
| 299 |
-
"learning_rate":
|
| 300 |
-
"logits/chosen":
|
| 301 |
-
"logits/rejected":
|
| 302 |
-
"logps/chosen": -
|
| 303 |
-
"logps/rejected": -
|
| 304 |
-
"loss": 0.
|
| 305 |
-
"rewards/accuracies": 0.
|
| 306 |
-
"rewards/chosen": -
|
| 307 |
-
"rewards/margins":
|
| 308 |
-
"rewards/rejected": -
|
| 309 |
"step": 190
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 0.
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"logits/chosen":
|
| 316 |
-
"logits/rejected":
|
| 317 |
-
"logps/chosen": -
|
| 318 |
-
"logps/rejected": -
|
| 319 |
-
"loss": 0.
|
| 320 |
-
"rewards/accuracies": 0.
|
| 321 |
-
"rewards/chosen": -
|
| 322 |
-
"rewards/margins":
|
| 323 |
-
"rewards/rejected": -
|
| 324 |
"step": 200
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"logits/chosen":
|
| 331 |
-
"logits/rejected":
|
| 332 |
-
"logps/chosen": -
|
| 333 |
-
"logps/rejected": -
|
| 334 |
-
"loss": 0.
|
| 335 |
-
"rewards/accuracies": 0.
|
| 336 |
-
"rewards/chosen": -
|
| 337 |
-
"rewards/margins":
|
| 338 |
-
"rewards/rejected": -
|
| 339 |
"step": 210
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 0.
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"logits/chosen":
|
| 346 |
-
"logits/rejected":
|
| 347 |
-
"logps/chosen": -
|
| 348 |
-
"logps/rejected": -
|
| 349 |
-
"loss": 0.
|
| 350 |
-
"rewards/accuracies": 0.
|
| 351 |
-
"rewards/chosen": -
|
| 352 |
-
"rewards/margins":
|
| 353 |
-
"rewards/rejected": -
|
| 354 |
"step": 220
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 0.
|
| 358 |
-
"grad_norm":
|
| 359 |
-
"learning_rate":
|
| 360 |
-
"logits/chosen":
|
| 361 |
-
"logits/rejected":
|
| 362 |
-
"logps/chosen": -
|
| 363 |
-
"logps/rejected": -
|
| 364 |
-
"loss": 0.
|
| 365 |
-
"rewards/accuracies": 0.
|
| 366 |
-
"rewards/chosen": -
|
| 367 |
-
"rewards/margins":
|
| 368 |
-
"rewards/rejected": -
|
| 369 |
"step": 230
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch": 0.
|
| 373 |
-
"grad_norm":
|
| 374 |
-
"learning_rate":
|
| 375 |
-
"logits/chosen":
|
| 376 |
-
"logits/rejected":
|
| 377 |
-
"logps/chosen": -
|
| 378 |
-
"logps/rejected": -
|
| 379 |
-
"loss": 0.
|
| 380 |
-
"rewards/accuracies": 0.
|
| 381 |
-
"rewards/chosen": -
|
| 382 |
-
"rewards/margins":
|
| 383 |
-
"rewards/rejected": -
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch": 0.
|
| 388 |
-
"grad_norm":
|
| 389 |
-
"learning_rate":
|
| 390 |
-
"logits/chosen":
|
| 391 |
-
"logits/rejected":
|
| 392 |
-
"logps/chosen": -
|
| 393 |
-
"logps/rejected": -
|
| 394 |
-
"loss": 0.
|
| 395 |
-
"rewards/accuracies": 0.
|
| 396 |
-
"rewards/chosen": -
|
| 397 |
-
"rewards/margins":
|
| 398 |
-
"rewards/rejected": -
|
| 399 |
"step": 250
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch": 0.
|
| 403 |
-
"grad_norm":
|
| 404 |
-
"learning_rate":
|
| 405 |
-
"logits/chosen":
|
| 406 |
-
"logits/rejected":
|
| 407 |
-
"logps/chosen": -
|
| 408 |
-
"logps/rejected": -
|
| 409 |
-
"loss": 0.
|
| 410 |
-
"rewards/accuracies": 0.
|
| 411 |
-
"rewards/chosen": -
|
| 412 |
-
"rewards/margins":
|
| 413 |
-
"rewards/rejected": -
|
| 414 |
"step": 260
|
| 415 |
},
|
| 416 |
{
|
| 417 |
-
"epoch": 0.
|
| 418 |
-
"grad_norm":
|
| 419 |
-
"learning_rate":
|
| 420 |
-
"logits/chosen":
|
| 421 |
-
"logits/rejected":
|
| 422 |
-
"logps/chosen": -
|
| 423 |
-
"logps/rejected": -
|
| 424 |
-
"loss": 0.
|
| 425 |
-
"rewards/accuracies": 0.
|
| 426 |
-
"rewards/chosen": -
|
| 427 |
-
"rewards/margins":
|
| 428 |
-
"rewards/rejected": -
|
| 429 |
"step": 270
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"logits/chosen":
|
| 436 |
-
"logits/rejected":
|
| 437 |
-
"logps/chosen": -
|
| 438 |
-
"logps/rejected": -
|
| 439 |
-
"loss": 0.
|
| 440 |
-
"rewards/accuracies": 0.
|
| 441 |
-
"rewards/chosen": -
|
| 442 |
-
"rewards/margins":
|
| 443 |
-
"rewards/rejected": -
|
| 444 |
"step": 280
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch": 0.
|
| 448 |
-
"grad_norm":
|
| 449 |
-
"learning_rate":
|
| 450 |
-
"logits/chosen":
|
| 451 |
-
"logits/rejected":
|
| 452 |
-
"logps/chosen": -
|
| 453 |
-
"logps/rejected": -
|
| 454 |
-
"loss": 0.
|
| 455 |
-
"rewards/accuracies": 0.
|
| 456 |
-
"rewards/chosen": -
|
| 457 |
-
"rewards/margins":
|
| 458 |
-
"rewards/rejected": -
|
| 459 |
"step": 290
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 0.
|
| 463 |
-
"grad_norm":
|
| 464 |
-
"learning_rate":
|
| 465 |
-
"logits/chosen":
|
| 466 |
-
"logits/rejected":
|
| 467 |
-
"logps/chosen": -
|
| 468 |
-
"logps/rejected": -
|
| 469 |
-
"loss": 0.
|
| 470 |
-
"rewards/accuracies": 0.
|
| 471 |
-
"rewards/chosen": -
|
| 472 |
-
"rewards/margins":
|
| 473 |
-
"rewards/rejected": -
|
| 474 |
"step": 300
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch": 0.
|
| 478 |
-
"grad_norm": 3.
|
| 479 |
-
"learning_rate":
|
| 480 |
-
"logits/chosen":
|
| 481 |
-
"logits/rejected":
|
| 482 |
-
"logps/chosen": -
|
| 483 |
-
"logps/rejected": -
|
| 484 |
-
"loss": 0.
|
| 485 |
-
"rewards/accuracies": 0.
|
| 486 |
-
"rewards/chosen": -
|
| 487 |
-
"rewards/margins":
|
| 488 |
-
"rewards/rejected": -
|
| 489 |
"step": 310
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch": 0.
|
| 493 |
-
"grad_norm": 5.
|
| 494 |
-
"learning_rate":
|
| 495 |
-
"logits/chosen":
|
| 496 |
-
"logits/rejected":
|
| 497 |
-
"logps/chosen": -
|
| 498 |
-
"logps/rejected": -
|
| 499 |
-
"loss": 0.
|
| 500 |
-
"rewards/accuracies": 0.
|
| 501 |
-
"rewards/chosen": -
|
| 502 |
-
"rewards/margins":
|
| 503 |
-
"rewards/rejected": -
|
| 504 |
"step": 320
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch": 0.
|
| 508 |
-
"grad_norm":
|
| 509 |
-
"learning_rate":
|
| 510 |
-
"logits/chosen":
|
| 511 |
-
"logits/rejected":
|
| 512 |
-
"logps/chosen": -
|
| 513 |
-
"logps/rejected": -
|
| 514 |
-
"loss": 0.
|
| 515 |
-
"rewards/accuracies": 0.
|
| 516 |
-
"rewards/chosen": -
|
| 517 |
-
"rewards/margins":
|
| 518 |
-
"rewards/rejected": -
|
| 519 |
"step": 330
|
| 520 |
},
|
| 521 |
{
|
| 522 |
-
"epoch": 0.
|
| 523 |
-
"grad_norm":
|
| 524 |
-
"learning_rate":
|
| 525 |
-
"logits/chosen":
|
| 526 |
-
"logits/rejected":
|
| 527 |
-
"logps/chosen": -
|
| 528 |
-
"logps/rejected": -
|
| 529 |
-
"loss": 0.
|
| 530 |
-
"rewards/accuracies": 0.
|
| 531 |
-
"rewards/chosen": -
|
| 532 |
-
"rewards/margins":
|
| 533 |
-
"rewards/rejected": -
|
| 534 |
"step": 340
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm":
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"logits/chosen":
|
| 541 |
-
"logits/rejected":
|
| 542 |
-
"logps/chosen": -
|
| 543 |
-
"logps/rejected": -
|
| 544 |
-
"loss": 0.
|
| 545 |
-
"rewards/accuracies": 0.
|
| 546 |
-
"rewards/chosen": -
|
| 547 |
-
"rewards/margins":
|
| 548 |
-
"rewards/rejected": -
|
| 549 |
"step": 350
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch": 0.
|
| 553 |
-
"grad_norm": 4.
|
| 554 |
-
"learning_rate":
|
| 555 |
-
"logits/chosen":
|
| 556 |
-
"logits/rejected":
|
| 557 |
-
"logps/chosen": -
|
| 558 |
-
"logps/rejected": -245.
|
| 559 |
-
"loss": 0.
|
| 560 |
-
"rewards/accuracies": 0.
|
| 561 |
-
"rewards/chosen": -
|
| 562 |
-
"rewards/margins":
|
| 563 |
-
"rewards/rejected": -
|
| 564 |
"step": 360
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"epoch": 0.
|
| 568 |
-
"grad_norm": 4.
|
| 569 |
-
"learning_rate":
|
| 570 |
-
"logits/chosen":
|
| 571 |
-
"logits/rejected":
|
| 572 |
-
"logps/chosen": -
|
| 573 |
-
"logps/rejected": -
|
| 574 |
-
"loss": 0.
|
| 575 |
-
"rewards/accuracies": 0.
|
| 576 |
-
"rewards/chosen": -
|
| 577 |
-
"rewards/margins":
|
| 578 |
-
"rewards/rejected": -
|
| 579 |
"step": 370
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch": 0.
|
| 583 |
-
"grad_norm":
|
| 584 |
-
"learning_rate":
|
| 585 |
-
"logits/chosen":
|
| 586 |
-
"logits/rejected":
|
| 587 |
-
"logps/chosen": -
|
| 588 |
-
"logps/rejected": -
|
| 589 |
-
"loss": 0.
|
| 590 |
-
"rewards/accuracies": 0.
|
| 591 |
-
"rewards/chosen": -
|
| 592 |
-
"rewards/margins":
|
| 593 |
-
"rewards/rejected": -
|
| 594 |
"step": 380
|
| 595 |
},
|
| 596 |
{
|
| 597 |
-
"epoch": 0.
|
| 598 |
-
"grad_norm":
|
| 599 |
-
"learning_rate": 4.
|
| 600 |
-
"logits/chosen":
|
| 601 |
-
"logits/rejected":
|
| 602 |
-
"logps/chosen": -
|
| 603 |
-
"logps/rejected": -
|
| 604 |
-
"loss": 0.
|
| 605 |
-
"rewards/accuracies": 0.
|
| 606 |
-
"rewards/chosen": -
|
| 607 |
-
"rewards/margins":
|
| 608 |
-
"rewards/rejected": -
|
| 609 |
"step": 390
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch": 0.
|
| 613 |
-
"grad_norm":
|
| 614 |
-
"learning_rate":
|
| 615 |
-
"logits/chosen":
|
| 616 |
-
"logits/rejected":
|
| 617 |
-
"logps/chosen": -
|
| 618 |
-
"logps/rejected": -
|
| 619 |
-
"loss": 0.
|
| 620 |
-
"rewards/accuracies": 0.
|
| 621 |
-
"rewards/chosen": -
|
| 622 |
-
"rewards/margins":
|
| 623 |
-
"rewards/rejected": -
|
| 624 |
"step": 400
|
| 625 |
},
|
| 626 |
{
|
| 627 |
-
"epoch": 0.
|
| 628 |
-
"grad_norm":
|
| 629 |
-
"learning_rate":
|
| 630 |
-
"logits/chosen":
|
| 631 |
-
"logits/rejected":
|
| 632 |
-
"logps/chosen": -
|
| 633 |
-
"logps/rejected": -
|
| 634 |
-
"loss": 0.
|
| 635 |
-
"rewards/accuracies": 0.
|
| 636 |
-
"rewards/chosen": -
|
| 637 |
-
"rewards/margins":
|
| 638 |
-
"rewards/rejected": -
|
| 639 |
"step": 410
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"grad_norm":
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"logits/chosen":
|
| 646 |
-
"logits/rejected":
|
| 647 |
-
"logps/chosen": -
|
| 648 |
-
"logps/rejected": -
|
| 649 |
-
"loss": 0.
|
| 650 |
-
"rewards/accuracies": 0.
|
| 651 |
-
"rewards/chosen": -
|
| 652 |
-
"rewards/margins":
|
| 653 |
-
"rewards/rejected": -
|
| 654 |
"step": 420
|
| 655 |
},
|
| 656 |
{
|
| 657 |
-
"epoch": 0.
|
| 658 |
-
"grad_norm":
|
| 659 |
-
"learning_rate":
|
| 660 |
-
"logits/chosen":
|
| 661 |
-
"logits/rejected":
|
| 662 |
-
"logps/chosen": -
|
| 663 |
-
"logps/rejected": -
|
| 664 |
-
"loss": 0.
|
| 665 |
-
"rewards/accuracies": 0.
|
| 666 |
-
"rewards/chosen": -
|
| 667 |
-
"rewards/margins":
|
| 668 |
-
"rewards/rejected": -
|
| 669 |
"step": 430
|
| 670 |
},
|
| 671 |
{
|
| 672 |
-
"epoch": 0.
|
| 673 |
-
"grad_norm":
|
| 674 |
-
"learning_rate":
|
| 675 |
-
"logits/chosen":
|
| 676 |
-
"logits/rejected":
|
| 677 |
-
"logps/chosen": -
|
| 678 |
-
"logps/rejected": -
|
| 679 |
-
"loss": 0.
|
| 680 |
-
"rewards/accuracies": 0.
|
| 681 |
-
"rewards/chosen": -
|
| 682 |
-
"rewards/margins":
|
| 683 |
-
"rewards/rejected": -
|
| 684 |
"step": 440
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch": 0.
|
| 688 |
-
"grad_norm":
|
| 689 |
-
"learning_rate": 4.
|
| 690 |
-
"logits/chosen":
|
| 691 |
-
"logits/rejected":
|
| 692 |
-
"logps/chosen": -
|
| 693 |
-
"logps/rejected": -
|
| 694 |
-
"loss": 0.
|
| 695 |
-
"rewards/accuracies": 0.
|
| 696 |
-
"rewards/chosen": -
|
| 697 |
-
"rewards/margins":
|
| 698 |
-
"rewards/rejected": -
|
| 699 |
"step": 450
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 0.
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate":
|
| 705 |
-
"logits/chosen":
|
| 706 |
-
"logits/rejected":
|
| 707 |
-
"logps/chosen": -
|
| 708 |
-
"logps/rejected": -
|
| 709 |
-
"loss": 0.4846,
|
| 710 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 711 |
-
"rewards/chosen": -11.391783714294434,
|
| 712 |
-
"rewards/margins": 24.004859924316406,
|
| 713 |
-
"rewards/rejected": -35.396644592285156,
|
| 714 |
-
"step": 460
|
| 715 |
-
},
|
| 716 |
-
{
|
| 717 |
-
"epoch": 0.36891679748822603,
|
| 718 |
-
"grad_norm": 6.4728785883394115,
|
| 719 |
-
"learning_rate": 3.979425312480629e-07,
|
| 720 |
-
"logits/chosen": 6111.62890625,
|
| 721 |
-
"logits/rejected": 5374.3291015625,
|
| 722 |
-
"logps/chosen": -224.78897094726562,
|
| 723 |
-
"logps/rejected": -242.69113159179688,
|
| 724 |
-
"loss": 0.4824,
|
| 725 |
-
"rewards/accuracies": 0.7583333253860474,
|
| 726 |
-
"rewards/chosen": -12.585268020629883,
|
| 727 |
-
"rewards/margins": 27.602148056030273,
|
| 728 |
-
"rewards/rejected": -40.18741226196289,
|
| 729 |
-
"step": 470
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"epoch": 0.37676609105180536,
|
| 733 |
-
"grad_norm": 5.3655573665549205,
|
| 734 |
-
"learning_rate": 3.923630753280357e-07,
|
| 735 |
-
"logits/chosen": 6582.39990234375,
|
| 736 |
-
"logits/rejected": 5717.3642578125,
|
| 737 |
-
"logps/chosen": -219.0083465576172,
|
| 738 |
-
"logps/rejected": -220.59555053710938,
|
| 739 |
-
"loss": 0.4817,
|
| 740 |
-
"rewards/accuracies": 0.7916667461395264,
|
| 741 |
-
"rewards/chosen": -11.603225708007812,
|
| 742 |
-
"rewards/margins": 34.595619201660156,
|
| 743 |
-
"rewards/rejected": -46.1988410949707,
|
| 744 |
-
"step": 480
|
| 745 |
-
},
|
| 746 |
-
{
|
| 747 |
-
"epoch": 0.38461538461538464,
|
| 748 |
-
"grad_norm": 5.979735399161116,
|
| 749 |
-
"learning_rate": 3.866766398810424e-07,
|
| 750 |
-
"logits/chosen": 6178.900390625,
|
| 751 |
-
"logits/rejected": 5935.67626953125,
|
| 752 |
-
"logps/chosen": -181.1698760986328,
|
| 753 |
-
"logps/rejected": -238.9916534423828,
|
| 754 |
-
"loss": 0.4775,
|
| 755 |
-
"rewards/accuracies": 0.8083332777023315,
|
| 756 |
-
"rewards/chosen": -4.562686920166016,
|
| 757 |
-
"rewards/margins": 32.35731887817383,
|
| 758 |
-
"rewards/rejected": -36.920005798339844,
|
| 759 |
-
"step": 490
|
| 760 |
-
},
|
| 761 |
-
{
|
| 762 |
-
"epoch": 0.3924646781789639,
|
| 763 |
-
"grad_norm": 10.290243596769058,
|
| 764 |
-
"learning_rate": 3.8088749801071496e-07,
|
| 765 |
-
"logits/chosen": 6733.89990234375,
|
| 766 |
-
"logits/rejected": 5217.791015625,
|
| 767 |
-
"logps/chosen": -243.348388671875,
|
| 768 |
-
"logps/rejected": -263.6581115722656,
|
| 769 |
-
"loss": 0.4843,
|
| 770 |
-
"rewards/accuracies": 0.7000000476837158,
|
| 771 |
-
"rewards/chosen": -21.35295867919922,
|
| 772 |
-
"rewards/margins": 39.74835968017578,
|
| 773 |
-
"rewards/rejected": -61.101318359375,
|
| 774 |
-
"step": 500
|
| 775 |
-
},
|
| 776 |
-
{
|
| 777 |
-
"epoch": 0.4003139717425432,
|
| 778 |
-
"grad_norm": 5.406771505983375,
|
| 779 |
-
"learning_rate": 3.75e-07,
|
| 780 |
-
"logits/chosen": 5354.39697265625,
|
| 781 |
-
"logits/rejected": 4763.21044921875,
|
| 782 |
-
"logps/chosen": -198.2074737548828,
|
| 783 |
-
"logps/rejected": -207.1690673828125,
|
| 784 |
-
"loss": 0.4824,
|
| 785 |
-
"rewards/accuracies": 0.7833333611488342,
|
| 786 |
-
"rewards/chosen": -8.783690452575684,
|
| 787 |
-
"rewards/margins": 32.053260803222656,
|
| 788 |
-
"rewards/rejected": -40.836944580078125,
|
| 789 |
-
"step": 510
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 0.40816326530612246,
|
| 793 |
-
"grad_norm": 3.9154423666136626,
|
| 794 |
-
"learning_rate": 3.6901857004211443e-07,
|
| 795 |
-
"logits/chosen": 5683.8447265625,
|
| 796 |
-
"logits/rejected": 5304.09033203125,
|
| 797 |
-
"logps/chosen": -210.19241333007812,
|
| 798 |
-
"logps/rejected": -231.1099090576172,
|
| 799 |
-
"loss": 0.4884,
|
| 800 |
-
"rewards/accuracies": 0.675000011920929,
|
| 801 |
-
"rewards/chosen": -11.772139549255371,
|
| 802 |
-
"rewards/margins": 23.49697494506836,
|
| 803 |
-
"rewards/rejected": -35.26911926269531,
|
| 804 |
-
"step": 520
|
| 805 |
-
},
|
| 806 |
-
{
|
| 807 |
-
"epoch": 0.41601255886970173,
|
| 808 |
-
"grad_norm": 5.038579766803527,
|
| 809 |
-
"learning_rate": 3.6294770291596076e-07,
|
| 810 |
-
"logits/chosen": 6436.02587890625,
|
| 811 |
-
"logits/rejected": 5317.28369140625,
|
| 812 |
-
"logps/chosen": -220.25057983398438,
|
| 813 |
-
"logps/rejected": -229.13784790039062,
|
| 814 |
-
"loss": 0.4798,
|
| 815 |
-
"rewards/accuracies": 0.6833333373069763,
|
| 816 |
-
"rewards/chosen": -7.315767765045166,
|
| 817 |
-
"rewards/margins": 23.054981231689453,
|
| 818 |
-
"rewards/rejected": -30.37074851989746,
|
| 819 |
-
"step": 530
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 0.423861852433281,
|
| 823 |
-
"grad_norm": 9.177958827560293,
|
| 824 |
-
"learning_rate": 3.5679196060850034e-07,
|
| 825 |
-
"logits/chosen": 6133.5283203125,
|
| 826 |
-
"logits/rejected": 5512.38818359375,
|
| 827 |
-
"logps/chosen": -224.00009155273438,
|
| 828 |
-
"logps/rejected": -239.9444122314453,
|
| 829 |
-
"loss": 0.484,
|
| 830 |
-
"rewards/accuracies": 0.7583333849906921,
|
| 831 |
-
"rewards/chosen": -13.195943832397461,
|
| 832 |
-
"rewards/margins": 38.481021881103516,
|
| 833 |
-
"rewards/rejected": -51.676963806152344,
|
| 834 |
-
"step": 540
|
| 835 |
-
},
|
| 836 |
-
{
|
| 837 |
-
"epoch": 0.4317111459968603,
|
| 838 |
-
"grad_norm": 5.4423546958299385,
|
| 839 |
-
"learning_rate": 3.505559688866229e-07,
|
| 840 |
-
"logits/chosen": 5929.5986328125,
|
| 841 |
-
"logits/rejected": 5545.6748046875,
|
| 842 |
-
"logps/chosen": -226.8612823486328,
|
| 843 |
-
"logps/rejected": -273.16229248046875,
|
| 844 |
-
"loss": 0.4784,
|
| 845 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 846 |
-
"rewards/chosen": -12.048417091369629,
|
| 847 |
-
"rewards/margins": 36.062232971191406,
|
| 848 |
-
"rewards/rejected": -48.110652923583984,
|
| 849 |
-
"step": 550
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 0.43956043956043955,
|
| 853 |
-
"grad_norm": 4.766412998974481,
|
| 854 |
-
"learning_rate": 3.4424441382108826e-07,
|
| 855 |
-
"logits/chosen": 5972.25830078125,
|
| 856 |
-
"logits/rejected": 5610.7490234375,
|
| 857 |
-
"logps/chosen": -217.2675018310547,
|
| 858 |
-
"logps/rejected": -234.3104705810547,
|
| 859 |
-
"loss": 0.4871,
|
| 860 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 861 |
-
"rewards/chosen": -13.991231918334961,
|
| 862 |
-
"rewards/margins": 27.081674575805664,
|
| 863 |
-
"rewards/rejected": -41.07290267944336,
|
| 864 |
-
"step": 560
|
| 865 |
-
},
|
| 866 |
-
{
|
| 867 |
-
"epoch": 0.4474097331240188,
|
| 868 |
-
"grad_norm": 6.338785824551311,
|
| 869 |
-
"learning_rate": 3.378620382651523e-07,
|
| 870 |
-
"logits/chosen": 6286.69287109375,
|
| 871 |
-
"logits/rejected": 5813.5283203125,
|
| 872 |
-
"logps/chosen": -255.1618194580078,
|
| 873 |
-
"logps/rejected": -267.93878173828125,
|
| 874 |
-
"loss": 0.4808,
|
| 875 |
-
"rewards/accuracies": 0.7583333849906921,
|
| 876 |
-
"rewards/chosen": -12.230340957641602,
|
| 877 |
-
"rewards/margins": 31.982650756835938,
|
| 878 |
-
"rewards/rejected": -44.212989807128906,
|
| 879 |
-
"step": 570
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 0.4552590266875981,
|
| 883 |
-
"grad_norm": 7.016140317833379,
|
| 884 |
-
"learning_rate": 3.314136382905234e-07,
|
| 885 |
-
"logits/chosen": 6247.73046875,
|
| 886 |
-
"logits/rejected": 5671.7021484375,
|
| 887 |
-
"logps/chosen": -222.32693481445312,
|
| 888 |
-
"logps/rejected": -259.5238952636719,
|
| 889 |
-
"loss": 0.4828,
|
| 890 |
-
"rewards/accuracies": 0.7416666746139526,
|
| 891 |
-
"rewards/chosen": -9.925975799560547,
|
| 892 |
-
"rewards/margins": 36.31413269042969,
|
| 893 |
-
"rewards/rejected": -46.24010467529297,
|
| 894 |
-
"step": 580
|
| 895 |
-
},
|
| 896 |
-
{
|
| 897 |
-
"epoch": 0.4631083202511774,
|
| 898 |
-
"grad_norm": 10.530000697305884,
|
| 899 |
-
"learning_rate": 3.249040595833274e-07,
|
| 900 |
-
"logits/chosen": 6807.6484375,
|
| 901 |
-
"logits/rejected": 5773.490234375,
|
| 902 |
-
"logps/chosen": -242.41659545898438,
|
| 903 |
-
"logps/rejected": -225.3022003173828,
|
| 904 |
-
"loss": 0.4788,
|
| 905 |
-
"rewards/accuracies": 0.783333420753479,
|
| 906 |
-
"rewards/chosen": -10.745205879211426,
|
| 907 |
-
"rewards/margins": 35.73249435424805,
|
| 908 |
-
"rewards/rejected": -46.47770309448242,
|
| 909 |
-
"step": 590
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 0.47095761381475665,
|
| 913 |
-
"grad_norm": 8.512339938567651,
|
| 914 |
-
"learning_rate": 3.1833819380279023e-07,
|
| 915 |
-
"logits/chosen": 6445.0234375,
|
| 916 |
-
"logits/rejected": 5518.9345703125,
|
| 917 |
-
"logps/chosen": -186.99856567382812,
|
| 918 |
-
"logps/rejected": -234.09817504882812,
|
| 919 |
-
"loss": 0.4785,
|
| 920 |
-
"rewards/accuracies": 0.8166667222976685,
|
| 921 |
-
"rewards/chosen": -9.450994491577148,
|
| 922 |
-
"rewards/margins": 34.949501037597656,
|
| 923 |
-
"rewards/rejected": -44.40049743652344,
|
| 924 |
-
"step": 600
|
| 925 |
-
},
|
| 926 |
-
{
|
| 927 |
-
"epoch": 0.478806907378336,
|
| 928 |
-
"grad_norm": 4.8733837949422885,
|
| 929 |
-
"learning_rate": 3.11720974905373e-07,
|
| 930 |
-
"logits/chosen": 6186.08447265625,
|
| 931 |
-
"logits/rejected": 5429.20361328125,
|
| 932 |
-
"logps/chosen": -218.30288696289062,
|
| 933 |
-
"logps/rejected": -232.8191680908203,
|
| 934 |
-
"loss": 0.4793,
|
| 935 |
-
"rewards/accuracies": 0.7500000596046448,
|
| 936 |
-
"rewards/chosen": -12.169063568115234,
|
| 937 |
-
"rewards/margins": 35.24351119995117,
|
| 938 |
-
"rewards/rejected": -47.412574768066406,
|
| 939 |
-
"step": 610
|
| 940 |
-
},
|
| 941 |
-
{
|
| 942 |
-
"epoch": 0.48665620094191525,
|
| 943 |
-
"grad_norm": 5.136229399526287,
|
| 944 |
-
"learning_rate": 3.0505737543712275e-07,
|
| 945 |
-
"logits/chosen": 5280.85546875,
|
| 946 |
-
"logits/rejected": 4370.572265625,
|
| 947 |
-
"logps/chosen": -196.75811767578125,
|
| 948 |
-
"logps/rejected": -205.48971557617188,
|
| 949 |
-
"loss": 0.4806,
|
| 950 |
-
"rewards/accuracies": 0.783333420753479,
|
| 951 |
-
"rewards/chosen": -12.993181228637695,
|
| 952 |
-
"rewards/margins": 31.400897979736328,
|
| 953 |
-
"rewards/rejected": -44.394081115722656,
|
| 954 |
-
"step": 620
|
| 955 |
-
},
|
| 956 |
-
{
|
| 957 |
-
"epoch": 0.4945054945054945,
|
| 958 |
-
"grad_norm": 6.5054877538720985,
|
| 959 |
-
"learning_rate": 2.9835240279702513e-07,
|
| 960 |
-
"logits/chosen": 6880.61083984375,
|
| 961 |
-
"logits/rejected": 5913.2353515625,
|
| 962 |
-
"logps/chosen": -251.8698272705078,
|
| 963 |
-
"logps/rejected": -250.5375518798828,
|
| 964 |
-
"loss": 0.4787,
|
| 965 |
-
"rewards/accuracies": 0.7916666269302368,
|
| 966 |
-
"rewards/chosen": -9.037474632263184,
|
| 967 |
-
"rewards/margins": 47.23527526855469,
|
| 968 |
-
"rewards/rejected": -56.27274703979492,
|
| 969 |
-
"step": 630
|
| 970 |
-
},
|
| 971 |
-
{
|
| 972 |
-
"epoch": 0.5023547880690737,
|
| 973 |
-
"grad_norm": 8.979595260533317,
|
| 974 |
-
"learning_rate": 2.9161109547416667e-07,
|
| 975 |
-
"logits/chosen": 6547.708984375,
|
| 976 |
-
"logits/rejected": 5625.1787109375,
|
| 977 |
-
"logps/chosen": -224.85488891601562,
|
| 978 |
-
"logps/rejected": -254.49667358398438,
|
| 979 |
-
"loss": 0.4768,
|
| 980 |
-
"rewards/accuracies": 0.7333332896232605,
|
| 981 |
-
"rewards/chosen": -14.077960014343262,
|
| 982 |
-
"rewards/margins": 25.92667007446289,
|
| 983 |
-
"rewards/rejected": -40.00463104248047,
|
| 984 |
-
"step": 640
|
| 985 |
-
},
|
| 986 |
-
{
|
| 987 |
-
"epoch": 0.5102040816326531,
|
| 988 |
-
"grad_norm": 6.717397549018566,
|
| 989 |
-
"learning_rate": 2.848385192615339e-07,
|
| 990 |
-
"logits/chosen": 5637.56005859375,
|
| 991 |
-
"logits/rejected": 4628.1474609375,
|
| 992 |
-
"logps/chosen": -208.9348907470703,
|
| 993 |
-
"logps/rejected": -219.1776885986328,
|
| 994 |
-
"loss": 0.4814,
|
| 995 |
-
"rewards/accuracies": 0.8166667222976685,
|
| 996 |
-
"rewards/chosen": -12.61233139038086,
|
| 997 |
-
"rewards/margins": 37.836753845214844,
|
| 998 |
-
"rewards/rejected": -50.44908905029297,
|
| 999 |
-
"step": 650
|
| 1000 |
-
},
|
| 1001 |
-
{
|
| 1002 |
-
"epoch": 0.5180533751962323,
|
| 1003 |
-
"grad_norm": 6.838400431176198,
|
| 1004 |
-
"learning_rate": 2.780397634492949e-07,
|
| 1005 |
-
"logits/chosen": 6302.87744140625,
|
| 1006 |
-
"logits/rejected": 5081.7353515625,
|
| 1007 |
-
"logps/chosen": -229.68081665039062,
|
| 1008 |
-
"logps/rejected": -247.7713165283203,
|
| 1009 |
-
"loss": 0.4833,
|
| 1010 |
-
"rewards/accuracies": 0.7916666269302368,
|
| 1011 |
-
"rewards/chosen": -12.300036430358887,
|
| 1012 |
-
"rewards/margins": 47.47599411010742,
|
| 1013 |
-
"rewards/rejected": -59.776031494140625,
|
| 1014 |
-
"step": 660
|
| 1015 |
-
},
|
| 1016 |
-
{
|
| 1017 |
-
"epoch": 0.5259026687598116,
|
| 1018 |
-
"grad_norm": 5.8476161724541535,
|
| 1019 |
-
"learning_rate": 2.71219937000424e-07,
|
| 1020 |
-
"logits/chosen": 6290.115234375,
|
| 1021 |
-
"logits/rejected": 5202.7998046875,
|
| 1022 |
-
"logps/chosen": -217.5615997314453,
|
| 1023 |
-
"logps/rejected": -227.6338348388672,
|
| 1024 |
-
"loss": 0.4827,
|
| 1025 |
-
"rewards/accuracies": 0.7916666269302368,
|
| 1026 |
-
"rewards/chosen": -10.82927131652832,
|
| 1027 |
-
"rewards/margins": 27.942428588867188,
|
| 1028 |
-
"rewards/rejected": -38.771697998046875,
|
| 1029 |
-
"step": 670
|
| 1030 |
-
},
|
| 1031 |
-
{
|
| 1032 |
-
"epoch": 0.533751962323391,
|
| 1033 |
-
"grad_norm": 5.146890250240064,
|
| 1034 |
-
"learning_rate": 2.6438416471154273e-07,
|
| 1035 |
-
"logits/chosen": 6111.2451171875,
|
| 1036 |
-
"logits/rejected": 5137.7705078125,
|
| 1037 |
-
"logps/chosen": -222.5435028076172,
|
| 1038 |
-
"logps/rejected": -225.12094116210938,
|
| 1039 |
-
"loss": 0.4815,
|
| 1040 |
-
"rewards/accuracies": 0.75,
|
| 1041 |
-
"rewards/chosen": -8.91177749633789,
|
| 1042 |
-
"rewards/margins": 36.991119384765625,
|
| 1043 |
-
"rewards/rejected": -45.902896881103516,
|
| 1044 |
-
"step": 680
|
| 1045 |
-
},
|
| 1046 |
-
{
|
| 1047 |
-
"epoch": 0.5416012558869702,
|
| 1048 |
-
"grad_norm": 8.8935250283176,
|
| 1049 |
-
"learning_rate": 2.5753758336186326e-07,
|
| 1050 |
-
"logits/chosen": 6043.0673828125,
|
| 1051 |
-
"logits/rejected": 5563.6806640625,
|
| 1052 |
-
"logps/chosen": -221.37338256835938,
|
| 1053 |
-
"logps/rejected": -271.20306396484375,
|
| 1054 |
-
"loss": 0.48,
|
| 1055 |
-
"rewards/accuracies": 0.7666667103767395,
|
| 1056 |
-
"rewards/chosen": -12.23568344116211,
|
| 1057 |
-
"rewards/margins": 42.76652145385742,
|
| 1058 |
-
"rewards/rejected": -55.00220489501953,
|
| 1059 |
-
"step": 690
|
| 1060 |
-
},
|
| 1061 |
-
{
|
| 1062 |
-
"epoch": 0.5494505494505495,
|
| 1063 |
-
"grad_norm": 5.847729785681332,
|
| 1064 |
-
"learning_rate": 2.5068533785312666e-07,
|
| 1065 |
-
"logits/chosen": 5745.1103515625,
|
| 1066 |
-
"logits/rejected": 5549.70849609375,
|
| 1067 |
-
"logps/chosen": -204.39468383789062,
|
| 1068 |
-
"logps/rejected": -238.9816436767578,
|
| 1069 |
-
"loss": 0.4882,
|
| 1070 |
-
"rewards/accuracies": 0.7333333492279053,
|
| 1071 |
-
"rewards/chosen": -15.167062759399414,
|
| 1072 |
-
"rewards/margins": 36.86690139770508,
|
| 1073 |
-
"rewards/rejected": -52.033958435058594,
|
| 1074 |
-
"step": 700
|
| 1075 |
-
},
|
| 1076 |
-
{
|
| 1077 |
-
"epoch": 0.5572998430141287,
|
| 1078 |
-
"grad_norm": 4.903588308451907,
|
| 1079 |
-
"learning_rate": 2.4383257734343794e-07,
|
| 1080 |
-
"logits/chosen": 5706.705078125,
|
| 1081 |
-
"logits/rejected": 5757.8916015625,
|
| 1082 |
-
"logps/chosen": -205.46023559570312,
|
| 1083 |
-
"logps/rejected": -243.0206756591797,
|
| 1084 |
-
"loss": 0.4795,
|
| 1085 |
-
"rewards/accuracies": 0.7750000953674316,
|
| 1086 |
-
"rewards/chosen": -11.52241039276123,
|
| 1087 |
-
"rewards/margins": 28.85659408569336,
|
| 1088 |
-
"rewards/rejected": -40.37900924682617,
|
| 1089 |
-
"step": 710
|
| 1090 |
-
},
|
| 1091 |
-
{
|
| 1092 |
-
"epoch": 0.565149136577708,
|
| 1093 |
-
"grad_norm": 4.5680367168979155,
|
| 1094 |
-
"learning_rate": 2.3698445137790258e-07,
|
| 1095 |
-
"logits/chosen": 6126.8349609375,
|
| 1096 |
-
"logits/rejected": 5314.02001953125,
|
| 1097 |
-
"logps/chosen": -227.4978790283203,
|
| 1098 |
-
"logps/rejected": -240.22860717773438,
|
| 1099 |
-
"loss": 0.4842,
|
| 1100 |
-
"rewards/accuracies": 0.7750000357627869,
|
| 1101 |
-
"rewards/chosen": -8.492025375366211,
|
| 1102 |
-
"rewards/margins": 31.947551727294922,
|
| 1103 |
-
"rewards/rejected": -40.4395751953125,
|
| 1104 |
-
"step": 720
|
| 1105 |
-
},
|
| 1106 |
-
{
|
| 1107 |
-
"epoch": 0.5729984301412873,
|
| 1108 |
-
"grad_norm": 6.161642634958935,
|
| 1109 |
-
"learning_rate": 2.3014610601897157e-07,
|
| 1110 |
-
"logits/chosen": 6653.5546875,
|
| 1111 |
-
"logits/rejected": 5136.146484375,
|
| 1112 |
-
"logps/chosen": -238.821533203125,
|
| 1113 |
-
"logps/rejected": -226.68203735351562,
|
| 1114 |
-
"loss": 0.4792,
|
| 1115 |
-
"rewards/accuracies": 0.7166666984558105,
|
| 1116 |
-
"rewards/chosen": -13.643511772155762,
|
| 1117 |
-
"rewards/margins": 37.98426818847656,
|
| 1118 |
-
"rewards/rejected": -51.627777099609375,
|
| 1119 |
-
"step": 730
|
| 1120 |
-
},
|
| 1121 |
-
{
|
| 1122 |
-
"epoch": 0.5808477237048666,
|
| 1123 |
-
"grad_norm": 5.261202948913745,
|
| 1124 |
-
"learning_rate": 2.2332267997940513e-07,
|
| 1125 |
-
"logits/chosen": 5534.0478515625,
|
| 1126 |
-
"logits/rejected": 4724.48095703125,
|
| 1127 |
-
"logps/chosen": -201.07333374023438,
|
| 1128 |
-
"logps/rejected": -208.75173950195312,
|
| 1129 |
-
"loss": 0.479,
|
| 1130 |
-
"rewards/accuracies": 0.7583333253860474,
|
| 1131 |
-
"rewards/chosen": -8.631407737731934,
|
| 1132 |
-
"rewards/margins": 37.47869110107422,
|
| 1133 |
-
"rewards/rejected": -46.1100959777832,
|
| 1134 |
-
"step": 740
|
| 1135 |
-
},
|
| 1136 |
-
{
|
| 1137 |
-
"epoch": 0.5886970172684458,
|
| 1138 |
-
"grad_norm": 6.0896040319265285,
|
| 1139 |
-
"learning_rate": 2.1651930076075723e-07,
|
| 1140 |
-
"logits/chosen": 6029.47021484375,
|
| 1141 |
-
"logits/rejected": 5496.3857421875,
|
| 1142 |
-
"logps/chosen": -191.79519653320312,
|
| 1143 |
-
"logps/rejected": -203.33026123046875,
|
| 1144 |
-
"loss": 0.4819,
|
| 1145 |
-
"rewards/accuracies": 0.6916667222976685,
|
| 1146 |
-
"rewards/chosen": -11.413000106811523,
|
| 1147 |
-
"rewards/margins": 24.80480194091797,
|
| 1148 |
-
"rewards/rejected": -36.217803955078125,
|
| 1149 |
-
"step": 750
|
| 1150 |
-
},
|
| 1151 |
-
{
|
| 1152 |
-
"epoch": 0.5965463108320251,
|
| 1153 |
-
"grad_norm": 4.761046588187656,
|
| 1154 |
-
"learning_rate": 2.0974108080028692e-07,
|
| 1155 |
-
"logits/chosen": 6314.89404296875,
|
| 1156 |
-
"logits/rejected": 5032.1162109375,
|
| 1157 |
-
"logps/chosen": -213.14517211914062,
|
| 1158 |
-
"logps/rejected": -213.0286865234375,
|
| 1159 |
"loss": 0.4832,
|
| 1160 |
-
"rewards/accuracies": 0.
|
| 1161 |
-
"rewards/chosen": -
|
| 1162 |
-
"rewards/margins":
|
| 1163 |
-
"rewards/rejected": -
|
| 1164 |
-
"step":
|
| 1165 |
-
},
|
| 1166 |
-
{
|
| 1167 |
-
"epoch": 0.6043956043956044,
|
| 1168 |
-
"grad_norm": 7.508020905443514,
|
| 1169 |
-
"learning_rate": 2.0299311362918773e-07,
|
| 1170 |
-
"logits/chosen": 6520.9560546875,
|
| 1171 |
-
"logits/rejected": 5636.296875,
|
| 1172 |
-
"logps/chosen": -242.4247589111328,
|
| 1173 |
-
"logps/rejected": -275.4010314941406,
|
| 1174 |
-
"loss": 0.4821,
|
| 1175 |
-
"rewards/accuracies": 0.7416666746139526,
|
| 1176 |
-
"rewards/chosen": -12.711216926574707,
|
| 1177 |
-
"rewards/margins": 38.81017303466797,
|
| 1178 |
-
"rewards/rejected": -51.521385192871094,
|
| 1179 |
-
"step": 770
|
| 1180 |
-
},
|
| 1181 |
-
{
|
| 1182 |
-
"epoch": 0.6122448979591837,
|
| 1183 |
-
"grad_norm": 5.044298723491718,
|
| 1184 |
-
"learning_rate": 1.962804700450265e-07,
|
| 1185 |
-
"logits/chosen": 6358.08837890625,
|
| 1186 |
-
"logits/rejected": 6072.24560546875,
|
| 1187 |
-
"logps/chosen": -226.471435546875,
|
| 1188 |
-
"logps/rejected": -281.1588134765625,
|
| 1189 |
-
"loss": 0.4829,
|
| 1190 |
-
"rewards/accuracies": 0.7500000596046448,
|
| 1191 |
-
"rewards/chosen": -10.3597993850708,
|
| 1192 |
-
"rewards/margins": 33.39599609375,
|
| 1193 |
-
"rewards/rejected": -43.75579071044922,
|
| 1194 |
-
"step": 780
|
| 1195 |
-
},
|
| 1196 |
-
{
|
| 1197 |
-
"epoch": 0.6200941915227629,
|
| 1198 |
-
"grad_norm": 7.21284462490354,
|
| 1199 |
-
"learning_rate": 1.8960819430126334e-07,
|
| 1200 |
-
"logits/chosen": 5915.1435546875,
|
| 1201 |
-
"logits/rejected": 5259.5546875,
|
| 1202 |
-
"logps/chosen": -213.5358428955078,
|
| 1203 |
-
"logps/rejected": -248.9722442626953,
|
| 1204 |
-
"loss": 0.4792,
|
| 1205 |
-
"rewards/accuracies": 0.800000011920929,
|
| 1206 |
-
"rewards/chosen": -15.374967575073242,
|
| 1207 |
-
"rewards/margins": 44.96841049194336,
|
| 1208 |
-
"rewards/rejected": -60.34337615966797,
|
| 1209 |
-
"step": 790
|
| 1210 |
-
},
|
| 1211 |
-
{
|
| 1212 |
-
"epoch": 0.6279434850863422,
|
| 1213 |
-
"grad_norm": 6.911749315995225,
|
| 1214 |
-
"learning_rate": 1.8298130031671972e-07,
|
| 1215 |
-
"logits/chosen": 5912.0732421875,
|
| 1216 |
-
"logits/rejected": 5204.81982421875,
|
| 1217 |
-
"logps/chosen": -231.18405151367188,
|
| 1218 |
-
"logps/rejected": -258.3150634765625,
|
| 1219 |
-
"loss": 0.4837,
|
| 1220 |
-
"rewards/accuracies": 0.7000000476837158,
|
| 1221 |
-
"rewards/chosen": -14.43272876739502,
|
| 1222 |
-
"rewards/margins": 31.22930335998535,
|
| 1223 |
-
"rewards/rejected": -45.66203689575195,
|
| 1224 |
-
"step": 800
|
| 1225 |
-
},
|
| 1226 |
-
{
|
| 1227 |
-
"epoch": 0.6357927786499215,
|
| 1228 |
-
"grad_norm": 5.218511863000425,
|
| 1229 |
-
"learning_rate": 1.7640476790784075e-07,
|
| 1230 |
-
"logits/chosen": 5465.1982421875,
|
| 1231 |
-
"logits/rejected": 4933.3720703125,
|
| 1232 |
-
"logps/chosen": -213.6913299560547,
|
| 1233 |
-
"logps/rejected": -266.97265625,
|
| 1234 |
-
"loss": 0.4817,
|
| 1235 |
-
"rewards/accuracies": 0.7749999761581421,
|
| 1236 |
-
"rewards/chosen": -11.240377426147461,
|
| 1237 |
-
"rewards/margins": 34.471500396728516,
|
| 1238 |
-
"rewards/rejected": -45.71187973022461,
|
| 1239 |
-
"step": 810
|
| 1240 |
-
},
|
| 1241 |
-
{
|
| 1242 |
-
"epoch": 0.6436420722135008,
|
| 1243 |
-
"grad_norm": 7.438249570403255,
|
| 1244 |
-
"learning_rate": 1.6988353904658492e-07,
|
| 1245 |
-
"logits/chosen": 5941.025390625,
|
| 1246 |
-
"logits/rejected": 4626.3564453125,
|
| 1247 |
-
"logps/chosen": -229.0679931640625,
|
| 1248 |
-
"logps/rejected": -208.60195922851562,
|
| 1249 |
-
"loss": 0.478,
|
| 1250 |
-
"rewards/accuracies": 0.75,
|
| 1251 |
-
"rewards/chosen": -7.8618903160095215,
|
| 1252 |
-
"rewards/margins": 32.07147216796875,
|
| 1253 |
-
"rewards/rejected": -39.9333610534668,
|
| 1254 |
-
"step": 820
|
| 1255 |
-
},
|
| 1256 |
-
{
|
| 1257 |
-
"epoch": 0.6514913657770801,
|
| 1258 |
-
"grad_norm": 8.778626873186747,
|
| 1259 |
-
"learning_rate": 1.634225141467513e-07,
|
| 1260 |
-
"logits/chosen": 5877.83447265625,
|
| 1261 |
-
"logits/rejected": 5282.060546875,
|
| 1262 |
-
"logps/chosen": -219.01065063476562,
|
| 1263 |
-
"logps/rejected": -245.612548828125,
|
| 1264 |
-
"loss": 0.4805,
|
| 1265 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 1266 |
-
"rewards/chosen": -14.819755554199219,
|
| 1267 |
-
"rewards/margins": 38.319461822509766,
|
| 1268 |
-
"rewards/rejected": -53.13921356201172,
|
| 1269 |
-
"step": 830
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
-
"epoch": 0.
|
| 1273 |
-
"grad_norm":
|
| 1274 |
-
"learning_rate":
|
| 1275 |
-
"logits/chosen":
|
| 1276 |
-
"logits/rejected":
|
| 1277 |
-
"logps/chosen": -
|
| 1278 |
-
"logps/rejected": -
|
| 1279 |
"loss": 0.481,
|
| 1280 |
-
"rewards/accuracies": 0.
|
| 1281 |
-
"rewards/chosen": -
|
| 1282 |
-
"rewards/margins":
|
| 1283 |
-
"rewards/rejected": -
|
| 1284 |
-
"step":
|
| 1285 |
-
},
|
| 1286 |
-
{
|
| 1287 |
-
"epoch": 0.6671899529042387,
|
| 1288 |
-
"grad_norm": 7.542089334458241,
|
| 1289 |
-
"learning_rate": 1.5070044803508691e-07,
|
| 1290 |
-
"logits/chosen": 5939.9482421875,
|
| 1291 |
-
"logits/rejected": 5369.12841796875,
|
| 1292 |
-
"logps/chosen": -227.2097930908203,
|
| 1293 |
-
"logps/rejected": -252.93820190429688,
|
| 1294 |
-
"loss": 0.4772,
|
| 1295 |
-
"rewards/accuracies": 0.7166667580604553,
|
| 1296 |
-
"rewards/chosen": -11.733682632446289,
|
| 1297 |
-
"rewards/margins": 40.6881217956543,
|
| 1298 |
-
"rewards/rejected": -52.42180252075195,
|
| 1299 |
-
"step": 850
|
| 1300 |
-
},
|
| 1301 |
-
{
|
| 1302 |
-
"epoch": 0.6750392464678179,
|
| 1303 |
-
"grad_norm": 10.41273121619664,
|
| 1304 |
-
"learning_rate": 1.444489668907914e-07,
|
| 1305 |
-
"logits/chosen": 6402.37109375,
|
| 1306 |
-
"logits/rejected": 5468.13134765625,
|
| 1307 |
-
"logps/chosen": -258.1214599609375,
|
| 1308 |
-
"logps/rejected": -253.7722625732422,
|
| 1309 |
-
"loss": 0.4809,
|
| 1310 |
-
"rewards/accuracies": 0.8000000715255737,
|
| 1311 |
-
"rewards/chosen": -10.320735931396484,
|
| 1312 |
-
"rewards/margins": 40.777584075927734,
|
| 1313 |
-
"rewards/rejected": -51.098323822021484,
|
| 1314 |
-
"step": 860
|
| 1315 |
-
},
|
| 1316 |
-
{
|
| 1317 |
-
"epoch": 0.6828885400313972,
|
| 1318 |
-
"grad_norm": 5.248825341405066,
|
| 1319 |
-
"learning_rate": 1.3827680265902232e-07,
|
| 1320 |
-
"logits/chosen": 6361.248046875,
|
| 1321 |
-
"logits/rejected": 5298.61572265625,
|
| 1322 |
-
"logps/chosen": -243.3575439453125,
|
| 1323 |
-
"logps/rejected": -250.5015106201172,
|
| 1324 |
-
"loss": 0.4825,
|
| 1325 |
-
"rewards/accuracies": 0.783333420753479,
|
| 1326 |
-
"rewards/chosen": -13.222381591796875,
|
| 1327 |
-
"rewards/margins": 36.560760498046875,
|
| 1328 |
-
"rewards/rejected": -49.783138275146484,
|
| 1329 |
-
"step": 870
|
| 1330 |
-
},
|
| 1331 |
-
{
|
| 1332 |
-
"epoch": 0.6907378335949764,
|
| 1333 |
-
"grad_norm": 8.056127375196619,
|
| 1334 |
-
"learning_rate": 1.3218859344701632e-07,
|
| 1335 |
-
"logits/chosen": 5598.6044921875,
|
| 1336 |
-
"logits/rejected": 5378.14208984375,
|
| 1337 |
-
"logps/chosen": -221.2474822998047,
|
| 1338 |
-
"logps/rejected": -275.5020446777344,
|
| 1339 |
-
"loss": 0.4827,
|
| 1340 |
-
"rewards/accuracies": 0.7249999642372131,
|
| 1341 |
-
"rewards/chosen": -10.712526321411133,
|
| 1342 |
-
"rewards/margins": 33.19308853149414,
|
| 1343 |
-
"rewards/rejected": -43.90561294555664,
|
| 1344 |
-
"step": 880
|
| 1345 |
-
},
|
| 1346 |
-
{
|
| 1347 |
-
"epoch": 0.6985871271585558,
|
| 1348 |
-
"grad_norm": 6.462283230726811,
|
| 1349 |
-
"learning_rate": 1.2618891427354172e-07,
|
| 1350 |
-
"logits/chosen": 6601.6826171875,
|
| 1351 |
-
"logits/rejected": 5406.0244140625,
|
| 1352 |
-
"logps/chosen": -266.26226806640625,
|
| 1353 |
-
"logps/rejected": -258.1165466308594,
|
| 1354 |
-
"loss": 0.4824,
|
| 1355 |
-
"rewards/accuracies": 0.8083332777023315,
|
| 1356 |
-
"rewards/chosen": -8.635444641113281,
|
| 1357 |
-
"rewards/margins": 37.71168518066406,
|
| 1358 |
-
"rewards/rejected": -46.347129821777344,
|
| 1359 |
-
"step": 890
|
| 1360 |
-
},
|
| 1361 |
-
{
|
| 1362 |
-
"epoch": 0.706436420722135,
|
| 1363 |
-
"grad_norm": 7.146931701552703,
|
| 1364 |
-
"learning_rate": 1.202822736309758e-07,
|
| 1365 |
-
"logits/chosen": 5596.5810546875,
|
| 1366 |
-
"logits/rejected": 5215.19140625,
|
| 1367 |
-
"logps/chosen": -214.41629028320312,
|
| 1368 |
-
"logps/rejected": -253.17379760742188,
|
| 1369 |
-
"loss": 0.4815,
|
| 1370 |
-
"rewards/accuracies": 0.73333340883255,
|
| 1371 |
-
"rewards/chosen": -10.091071128845215,
|
| 1372 |
-
"rewards/margins": 32.64034652709961,
|
| 1373 |
-
"rewards/rejected": -42.731414794921875,
|
| 1374 |
-
"step": 900
|
| 1375 |
-
},
|
| 1376 |
-
{
|
| 1377 |
-
"epoch": 0.7142857142857143,
|
| 1378 |
-
"grad_norm": 6.45263078416188,
|
| 1379 |
-
"learning_rate": 1.1447311009737299e-07,
|
| 1380 |
-
"logits/chosen": 5508.6884765625,
|
| 1381 |
-
"logits/rejected": 5250.3154296875,
|
| 1382 |
-
"logps/chosen": -222.135986328125,
|
| 1383 |
-
"logps/rejected": -262.9181213378906,
|
| 1384 |
-
"loss": 0.4803,
|
| 1385 |
-
"rewards/accuracies": 0.7749999761581421,
|
| 1386 |
-
"rewards/chosen": -13.866915702819824,
|
| 1387 |
-
"rewards/margins": 41.107566833496094,
|
| 1388 |
-
"rewards/rejected": -54.97447967529297,
|
| 1389 |
-
"step": 910
|
| 1390 |
-
},
|
| 1391 |
-
{
|
| 1392 |
-
"epoch": 0.7221350078492935,
|
| 1393 |
-
"grad_norm": 5.759526286673952,
|
| 1394 |
-
"learning_rate": 1.0876578900107053e-07,
|
| 1395 |
-
"logits/chosen": 6085.29931640625,
|
| 1396 |
-
"logits/rejected": 5071.2158203125,
|
| 1397 |
-
"logps/chosen": -246.2467498779297,
|
| 1398 |
-
"logps/rejected": -252.5704803466797,
|
| 1399 |
-
"loss": 0.4796,
|
| 1400 |
-
"rewards/accuracies": 0.783333420753479,
|
| 1401 |
-
"rewards/chosen": -13.800603866577148,
|
| 1402 |
-
"rewards/margins": 40.584354400634766,
|
| 1403 |
-
"rewards/rejected": -54.38496017456055,
|
| 1404 |
-
"step": 920
|
| 1405 |
-
},
|
| 1406 |
-
{
|
| 1407 |
-
"epoch": 0.7299843014128728,
|
| 1408 |
-
"grad_norm": 5.35980955418667,
|
| 1409 |
-
"learning_rate": 1.0316459914033793e-07,
|
| 1410 |
-
"logits/chosen": 5992.89306640625,
|
| 1411 |
-
"logits/rejected": 4551.4521484375,
|
| 1412 |
-
"logps/chosen": -253.8580780029297,
|
| 1413 |
-
"logps/rejected": -241.84286499023438,
|
| 1414 |
-
"loss": 0.4835,
|
| 1415 |
-
"rewards/accuracies": 0.7416666746139526,
|
| 1416 |
-
"rewards/chosen": -17.851383209228516,
|
| 1417 |
-
"rewards/margins": 38.075477600097656,
|
| 1418 |
-
"rewards/rejected": -55.9268684387207,
|
| 1419 |
-
"step": 930
|
| 1420 |
-
},
|
| 1421 |
-
{
|
| 1422 |
-
"epoch": 0.7378335949764521,
|
| 1423 |
-
"grad_norm": 5.546184868537026,
|
| 1424 |
-
"learning_rate": 9.767374956053584e-08,
|
| 1425 |
-
"logits/chosen": 5804.6865234375,
|
| 1426 |
-
"logits/rejected": 5108.57958984375,
|
| 1427 |
-
"logps/chosen": -231.1918182373047,
|
| 1428 |
-
"logps/rejected": -260.5343017578125,
|
| 1429 |
-
"loss": 0.4798,
|
| 1430 |
-
"rewards/accuracies": 0.7333333492279053,
|
| 1431 |
-
"rewards/chosen": -12.856742858886719,
|
| 1432 |
-
"rewards/margins": 42.71269989013672,
|
| 1433 |
-
"rewards/rejected": -55.569435119628906,
|
| 1434 |
-
"step": 940
|
| 1435 |
-
},
|
| 1436 |
-
{
|
| 1437 |
-
"epoch": 0.7456828885400314,
|
| 1438 |
-
"grad_norm": 5.310959910799943,
|
| 1439 |
-
"learning_rate": 9.229736639120561e-08,
|
| 1440 |
-
"logits/chosen": 5984.3681640625,
|
| 1441 |
-
"logits/rejected": 5550.1865234375,
|
| 1442 |
-
"logps/chosen": -229.02523803710938,
|
| 1443 |
-
"logps/rejected": -247.6044464111328,
|
| 1444 |
-
"loss": 0.4835,
|
| 1445 |
-
"rewards/accuracies": 0.7833333611488342,
|
| 1446 |
-
"rewards/chosen": -11.238405227661133,
|
| 1447 |
-
"rewards/margins": 22.278400421142578,
|
| 1448 |
-
"rewards/rejected": -33.516807556152344,
|
| 1449 |
-
"step": 950
|
| 1450 |
-
},
|
| 1451 |
-
{
|
| 1452 |
-
"epoch": 0.7535321821036107,
|
| 1453 |
-
"grad_norm": 5.724504063036896,
|
| 1454 |
-
"learning_rate": 8.70394897454659e-08,
|
| 1455 |
-
"logits/chosen": 5837.259765625,
|
| 1456 |
-
"logits/rejected": 5220.193359375,
|
| 1457 |
-
"logps/chosen": -226.56478881835938,
|
| 1458 |
-
"logps/rejected": -249.77938842773438,
|
| 1459 |
-
"loss": 0.4772,
|
| 1460 |
-
"rewards/accuracies": 0.8083332777023315,
|
| 1461 |
-
"rewards/chosen": -8.066027641296387,
|
| 1462 |
-
"rewards/margins": 39.250091552734375,
|
| 1463 |
-
"rewards/rejected": -47.31611633300781,
|
| 1464 |
-
"step": 960
|
| 1465 |
-
},
|
| 1466 |
-
{
|
| 1467 |
-
"epoch": 0.7613814756671899,
|
| 1468 |
-
"grad_norm": 6.84872456451459,
|
| 1469 |
-
"learning_rate": 8.19040706840472e-08,
|
| 1470 |
-
"logits/chosen": 5938.76025390625,
|
| 1471 |
-
"logits/rejected": 5001.06396484375,
|
| 1472 |
-
"logps/chosen": -251.7744598388672,
|
| 1473 |
-
"logps/rejected": -265.7369689941406,
|
| 1474 |
-
"loss": 0.4772,
|
| 1475 |
-
"rewards/accuracies": 0.7749999761581421,
|
| 1476 |
-
"rewards/chosen": -11.534469604492188,
|
| 1477 |
-
"rewards/margins": 47.53069305419922,
|
| 1478 |
-
"rewards/rejected": -59.065155029296875,
|
| 1479 |
-
"step": 970
|
| 1480 |
-
},
|
| 1481 |
-
{
|
| 1482 |
-
"epoch": 0.7692307692307693,
|
| 1483 |
-
"grad_norm": 6.2584137075852455,
|
| 1484 |
-
"learning_rate": 7.689496824624525e-08,
|
| 1485 |
-
"logits/chosen": 5646.22802734375,
|
| 1486 |
-
"logits/rejected": 4573.884765625,
|
| 1487 |
-
"logps/chosen": -238.1552276611328,
|
| 1488 |
-
"logps/rejected": -266.3992614746094,
|
| 1489 |
-
"loss": 0.4767,
|
| 1490 |
-
"rewards/accuracies": 0.8333333134651184,
|
| 1491 |
-
"rewards/chosen": -9.31987190246582,
|
| 1492 |
-
"rewards/margins": 66.60228729248047,
|
| 1493 |
-
"rewards/rejected": -75.92216491699219,
|
| 1494 |
-
"step": 980
|
| 1495 |
-
},
|
| 1496 |
-
{
|
| 1497 |
-
"epoch": 0.7770800627943485,
|
| 1498 |
-
"grad_norm": 8.467713398392366,
|
| 1499 |
-
"learning_rate": 7.201594655002458e-08,
|
| 1500 |
-
"logits/chosen": 5968.3173828125,
|
| 1501 |
-
"logits/rejected": 5018.83740234375,
|
| 1502 |
-
"logps/chosen": -242.1980438232422,
|
| 1503 |
-
"logps/rejected": -261.9761657714844,
|
| 1504 |
-
"loss": 0.4798,
|
| 1505 |
-
"rewards/accuracies": 0.7083333730697632,
|
| 1506 |
-
"rewards/chosen": -16.304744720458984,
|
| 1507 |
-
"rewards/margins": 51.60383987426758,
|
| 1508 |
-
"rewards/rejected": -67.90858459472656,
|
| 1509 |
-
"step": 990
|
| 1510 |
-
},
|
| 1511 |
-
{
|
| 1512 |
-
"epoch": 0.7849293563579278,
|
| 1513 |
-
"grad_norm": 7.778614626465624,
|
| 1514 |
-
"learning_rate": 6.727067196345099e-08,
|
| 1515 |
-
"logits/chosen": 5659.57275390625,
|
| 1516 |
-
"logits/rejected": 4817.7607421875,
|
| 1517 |
-
"logps/chosen": -226.59005737304688,
|
| 1518 |
-
"logps/rejected": -226.68734741210938,
|
| 1519 |
-
"loss": 0.4817,
|
| 1520 |
-
"rewards/accuracies": 0.7166666388511658,
|
| 1521 |
-
"rewards/chosen": -12.7899169921875,
|
| 1522 |
-
"rewards/margins": 33.538944244384766,
|
| 1523 |
-
"rewards/rejected": -46.328861236572266,
|
| 1524 |
-
"step": 1000
|
| 1525 |
-
},
|
| 1526 |
-
{
|
| 1527 |
-
"epoch": 0.792778649921507,
|
| 1528 |
-
"grad_norm": 6.403550827254964,
|
| 1529 |
-
"learning_rate": 6.26627103495786e-08,
|
| 1530 |
-
"logits/chosen": 5840.86767578125,
|
| 1531 |
-
"logits/rejected": 4903.9775390625,
|
| 1532 |
-
"logps/chosen": -223.2283477783203,
|
| 1533 |
-
"logps/rejected": -242.93466186523438,
|
| 1534 |
-
"loss": 0.4798,
|
| 1535 |
-
"rewards/accuracies": 0.75,
|
| 1536 |
-
"rewards/chosen": -8.5188570022583,
|
| 1537 |
-
"rewards/margins": 43.02119064331055,
|
| 1538 |
-
"rewards/rejected": -51.54004669189453,
|
| 1539 |
-
"step": 1010
|
| 1540 |
-
},
|
| 1541 |
-
{
|
| 1542 |
-
"epoch": 0.8006279434850864,
|
| 1543 |
-
"grad_norm": 5.514137797353558,
|
| 1544 |
-
"learning_rate": 5.8195524386862374e-08,
|
| 1545 |
-
"logits/chosen": 5926.3291015625,
|
| 1546 |
-
"logits/rejected": 5305.26318359375,
|
| 1547 |
-
"logps/chosen": -256.4900207519531,
|
| 1548 |
-
"logps/rejected": -274.34906005859375,
|
| 1549 |
-
"loss": 0.4821,
|
| 1550 |
-
"rewards/accuracies": 0.7916666269302368,
|
| 1551 |
-
"rewards/chosen": -8.3116455078125,
|
| 1552 |
-
"rewards/margins": 39.88562774658203,
|
| 1553 |
-
"rewards/rejected": -48.19727325439453,
|
| 1554 |
-
"step": 1020
|
| 1555 |
-
},
|
| 1556 |
-
{
|
| 1557 |
-
"epoch": 0.8084772370486656,
|
| 1558 |
-
"grad_norm": 5.221372206353793,
|
| 1559 |
-
"learning_rate": 5.38724709671092e-08,
|
| 1560 |
-
"logits/chosen": 6330.0234375,
|
| 1561 |
-
"logits/rejected": 6003.18408203125,
|
| 1562 |
-
"logps/chosen": -241.54736328125,
|
| 1563 |
-
"logps/rejected": -281.74896240234375,
|
| 1564 |
-
"loss": 0.4787,
|
| 1565 |
-
"rewards/accuracies": 0.783333420753479,
|
| 1566 |
-
"rewards/chosen": -10.10380744934082,
|
| 1567 |
-
"rewards/margins": 38.526222229003906,
|
| 1568 |
-
"rewards/rejected": -48.630027770996094,
|
| 1569 |
-
"step": 1030
|
| 1570 |
-
},
|
| 1571 |
-
{
|
| 1572 |
-
"epoch": 0.8163265306122449,
|
| 1573 |
-
"grad_norm": 7.932681835732521,
|
| 1574 |
-
"learning_rate": 4.969679867292276e-08,
|
| 1575 |
-
"logits/chosen": 5629.4482421875,
|
| 1576 |
-
"logits/rejected": 5160.17578125,
|
| 1577 |
-
"logps/chosen": -234.3016357421875,
|
| 1578 |
-
"logps/rejected": -264.07110595703125,
|
| 1579 |
-
"loss": 0.4798,
|
| 1580 |
-
"rewards/accuracies": 0.7166667580604553,
|
| 1581 |
-
"rewards/chosen": -14.211596488952637,
|
| 1582 |
-
"rewards/margins": 40.67064666748047,
|
| 1583 |
-
"rewards/rejected": -54.88224411010742,
|
| 1584 |
-
"step": 1040
|
| 1585 |
-
},
|
| 1586 |
-
{
|
| 1587 |
-
"epoch": 0.8241758241758241,
|
| 1588 |
-
"grad_norm": 6.9567227767775615,
|
| 1589 |
-
"learning_rate": 4.5671645336537416e-08,
|
| 1590 |
-
"logits/chosen": 5678.796875,
|
| 1591 |
-
"logits/rejected": 5205.8046875,
|
| 1592 |
-
"logps/chosen": -250.764404296875,
|
| 1593 |
-
"logps/rejected": -270.3853759765625,
|
| 1594 |
-
"loss": 0.4788,
|
| 1595 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 1596 |
-
"rewards/chosen": -13.21537971496582,
|
| 1597 |
-
"rewards/margins": 41.3323860168457,
|
| 1598 |
-
"rewards/rejected": -54.54777145385742,
|
| 1599 |
-
"step": 1050
|
| 1600 |
-
},
|
| 1601 |
-
{
|
| 1602 |
-
"epoch": 0.8320251177394035,
|
| 1603 |
-
"grad_norm": 11.260543191909672,
|
| 1604 |
-
"learning_rate": 4.180003568187776e-08,
|
| 1605 |
-
"logits/chosen": 7013.36181640625,
|
| 1606 |
-
"logits/rejected": 5547.8046875,
|
| 1607 |
-
"logps/chosen": -277.34161376953125,
|
| 1608 |
-
"logps/rejected": -266.3480224609375,
|
| 1609 |
-
"loss": 0.4853,
|
| 1610 |
-
"rewards/accuracies": 0.6666666865348816,
|
| 1611 |
-
"rewards/chosen": -16.316293716430664,
|
| 1612 |
-
"rewards/margins": 29.96787452697754,
|
| 1613 |
-
"rewards/rejected": -46.28417205810547,
|
| 1614 |
-
"step": 1060
|
| 1615 |
-
},
|
| 1616 |
-
{
|
| 1617 |
-
"epoch": 0.8398744113029827,
|
| 1618 |
-
"grad_norm": 5.873157048382797,
|
| 1619 |
-
"learning_rate": 3.8084879051612144e-08,
|
| 1620 |
-
"logits/chosen": 5843.4775390625,
|
| 1621 |
-
"logits/rejected": 5391.5986328125,
|
| 1622 |
-
"logps/chosen": -234.66769409179688,
|
| 1623 |
-
"logps/rejected": -239.1238555908203,
|
| 1624 |
-
"loss": 0.4805,
|
| 1625 |
-
"rewards/accuracies": 0.73333340883255,
|
| 1626 |
-
"rewards/chosen": -9.391263961791992,
|
| 1627 |
-
"rewards/margins": 36.595001220703125,
|
| 1628 |
-
"rewards/rejected": -45.98626708984375,
|
| 1629 |
-
"step": 1070
|
| 1630 |
-
},
|
| 1631 |
-
{
|
| 1632 |
-
"epoch": 0.847723704866562,
|
| 1633 |
-
"grad_norm": 6.007255893505933,
|
| 1634 |
-
"learning_rate": 3.452896722091128e-08,
|
| 1635 |
-
"logits/chosen": 6401.373046875,
|
| 1636 |
-
"logits/rejected": 4987.56005859375,
|
| 1637 |
-
"logps/chosen": -275.00115966796875,
|
| 1638 |
-
"logps/rejected": -255.5666046142578,
|
| 1639 |
-
"loss": 0.4771,
|
| 1640 |
-
"rewards/accuracies": 0.8083333969116211,
|
| 1641 |
-
"rewards/chosen": -8.407303810119629,
|
| 1642 |
-
"rewards/margins": 46.17400360107422,
|
| 1643 |
-
"rewards/rejected": -54.5813102722168,
|
| 1644 |
-
"step": 1080
|
| 1645 |
-
},
|
| 1646 |
-
{
|
| 1647 |
-
"epoch": 0.8555729984301413,
|
| 1648 |
-
"grad_norm": 6.959675987878217,
|
| 1649 |
-
"learning_rate": 3.11349722995527e-08,
|
| 1650 |
-
"logits/chosen": 6485.73193359375,
|
| 1651 |
-
"logits/rejected": 4893.6123046875,
|
| 1652 |
-
"logps/chosen": -240.83041381835938,
|
| 1653 |
-
"logps/rejected": -262.59429931640625,
|
| 1654 |
-
"loss": 0.4823,
|
| 1655 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 1656 |
-
"rewards/chosen": -13.316200256347656,
|
| 1657 |
-
"rewards/margins": 36.36842727661133,
|
| 1658 |
-
"rewards/rejected": -49.684635162353516,
|
| 1659 |
-
"step": 1090
|
| 1660 |
-
},
|
| 1661 |
-
{
|
| 1662 |
-
"epoch": 0.8634222919937206,
|
| 1663 |
-
"grad_norm": 6.728053574170472,
|
| 1664 |
-
"learning_rate": 2.7905444723949762e-08,
|
| 1665 |
-
"logits/chosen": 6255.11572265625,
|
| 1666 |
-
"logits/rejected": 5196.1142578125,
|
| 1667 |
-
"logps/chosen": -252.09976196289062,
|
| 1668 |
-
"logps/rejected": -247.3374786376953,
|
| 1669 |
-
"loss": 0.4814,
|
| 1670 |
-
"rewards/accuracies": 0.7749999761581421,
|
| 1671 |
-
"rewards/chosen": -13.543619155883789,
|
| 1672 |
-
"rewards/margins": 44.29808807373047,
|
| 1673 |
-
"rewards/rejected": -57.841705322265625,
|
| 1674 |
-
"step": 1100
|
| 1675 |
-
},
|
| 1676 |
-
{
|
| 1677 |
-
"epoch": 0.8712715855572999,
|
| 1678 |
-
"grad_norm": 6.249434590126908,
|
| 1679 |
-
"learning_rate": 2.484281134061142e-08,
|
| 1680 |
-
"logits/chosen": 6619.61474609375,
|
| 1681 |
-
"logits/rejected": 5369.69189453125,
|
| 1682 |
-
"logps/chosen": -277.9066162109375,
|
| 1683 |
-
"logps/rejected": -277.1863708496094,
|
| 1684 |
-
"loss": 0.4815,
|
| 1685 |
-
"rewards/accuracies": 0.7916666269302368,
|
| 1686 |
-
"rewards/chosen": -13.30717658996582,
|
| 1687 |
-
"rewards/margins": 41.61245346069336,
|
| 1688 |
-
"rewards/rejected": -54.91962814331055,
|
| 1689 |
-
"step": 1110
|
| 1690 |
-
},
|
| 1691 |
-
{
|
| 1692 |
-
"epoch": 0.8791208791208791,
|
| 1693 |
-
"grad_norm": 6.621447418753569,
|
| 1694 |
-
"learning_rate": 2.194937358247506e-08,
|
| 1695 |
-
"logits/chosen": 6474.90234375,
|
| 1696 |
-
"logits/rejected": 5295.9033203125,
|
| 1697 |
-
"logps/chosen": -260.36065673828125,
|
| 1698 |
-
"logps/rejected": -271.5030517578125,
|
| 1699 |
-
"loss": 0.479,
|
| 1700 |
-
"rewards/accuracies": 0.7749999761581421,
|
| 1701 |
-
"rewards/chosen": -15.229601860046387,
|
| 1702 |
-
"rewards/margins": 39.49894332885742,
|
| 1703 |
-
"rewards/rejected": -54.728553771972656,
|
| 1704 |
-
"step": 1120
|
| 1705 |
-
},
|
| 1706 |
-
{
|
| 1707 |
-
"epoch": 0.8869701726844584,
|
| 1708 |
-
"grad_norm": 8.33446410604382,
|
| 1709 |
-
"learning_rate": 1.9227305739481612e-08,
|
| 1710 |
-
"logits/chosen": 5892.1806640625,
|
| 1711 |
-
"logits/rejected": 4676.626953125,
|
| 1712 |
-
"logps/chosen": -243.70510864257812,
|
| 1713 |
-
"logps/rejected": -232.04556274414062,
|
| 1714 |
-
"loss": 0.4778,
|
| 1715 |
-
"rewards/accuracies": 0.7166666388511658,
|
| 1716 |
-
"rewards/chosen": -8.918004989624023,
|
| 1717 |
-
"rewards/margins": 42.899051666259766,
|
| 1718 |
-
"rewards/rejected": -51.81705856323242,
|
| 1719 |
-
"step": 1130
|
| 1720 |
-
},
|
| 1721 |
-
{
|
| 1722 |
-
"epoch": 0.8948194662480377,
|
| 1723 |
-
"grad_norm": 5.6890326351976155,
|
| 1724 |
-
"learning_rate": 1.6678653324693787e-08,
|
| 1725 |
-
"logits/chosen": 6475.4951171875,
|
| 1726 |
-
"logits/rejected": 5299.70751953125,
|
| 1727 |
-
"logps/chosen": -269.6833190917969,
|
| 1728 |
-
"logps/rejected": -268.32513427734375,
|
| 1729 |
-
"loss": 0.4802,
|
| 1730 |
-
"rewards/accuracies": 0.75,
|
| 1731 |
-
"rewards/chosen": -12.040689468383789,
|
| 1732 |
-
"rewards/margins": 35.418983459472656,
|
| 1733 |
-
"rewards/rejected": -47.459678649902344,
|
| 1734 |
-
"step": 1140
|
| 1735 |
-
},
|
| 1736 |
-
{
|
| 1737 |
-
"epoch": 0.902668759811617,
|
| 1738 |
-
"grad_norm": 6.048036034423531,
|
| 1739 |
-
"learning_rate": 1.4305331537183384e-08,
|
| 1740 |
-
"logits/chosen": 5732.1220703125,
|
| 1741 |
-
"logits/rejected": 5299.99560546875,
|
| 1742 |
-
"logps/chosen": -238.8845977783203,
|
| 1743 |
-
"logps/rejected": -264.5845031738281,
|
| 1744 |
-
"loss": 0.478,
|
| 1745 |
-
"rewards/accuracies": 0.7166667580604553,
|
| 1746 |
-
"rewards/chosen": -11.897860527038574,
|
| 1747 |
-
"rewards/margins": 33.151702880859375,
|
| 1748 |
-
"rewards/rejected": -45.04956817626953,
|
| 1749 |
-
"step": 1150
|
| 1750 |
-
},
|
| 1751 |
-
{
|
| 1752 |
-
"epoch": 0.9105180533751962,
|
| 1753 |
-
"grad_norm": 5.848377321734073,
|
| 1754 |
-
"learning_rate": 1.2109123822844653e-08,
|
| 1755 |
-
"logits/chosen": 5896.5595703125,
|
| 1756 |
-
"logits/rejected": 4711.38427734375,
|
| 1757 |
-
"logps/chosen": -243.69515991210938,
|
| 1758 |
-
"logps/rejected": -245.3145751953125,
|
| 1759 |
-
"loss": 0.4802,
|
| 1760 |
-
"rewards/accuracies": 0.7583333253860474,
|
| 1761 |
-
"rewards/chosen": -15.122274398803711,
|
| 1762 |
-
"rewards/margins": 34.2625846862793,
|
| 1763 |
-
"rewards/rejected": -49.384857177734375,
|
| 1764 |
-
"step": 1160
|
| 1765 |
-
},
|
| 1766 |
-
{
|
| 1767 |
-
"epoch": 0.9183673469387755,
|
| 1768 |
-
"grad_norm": 5.838867657991871,
|
| 1769 |
-
"learning_rate": 1.0091680534213387e-08,
|
| 1770 |
-
"logits/chosen": 6464.91552734375,
|
| 1771 |
-
"logits/rejected": 6239.65771484375,
|
| 1772 |
-
"logps/chosen": -256.83367919921875,
|
| 1773 |
-
"logps/rejected": -294.30145263671875,
|
| 1774 |
-
"loss": 0.4816,
|
| 1775 |
-
"rewards/accuracies": 0.7250000238418579,
|
| 1776 |
-
"rewards/chosen": -13.878461837768555,
|
| 1777 |
-
"rewards/margins": 33.63318634033203,
|
| 1778 |
-
"rewards/rejected": -47.511653900146484,
|
| 1779 |
-
"step": 1170
|
| 1780 |
-
},
|
| 1781 |
-
{
|
| 1782 |
-
"epoch": 0.9262166405023547,
|
| 1783 |
-
"grad_norm": 6.801398869039188,
|
| 1784 |
-
"learning_rate": 8.254517690300944e-09,
|
| 1785 |
-
"logits/chosen": 5692.70703125,
|
| 1786 |
-
"logits/rejected": 5195.38525390625,
|
| 1787 |
-
"logps/chosen": -251.6653289794922,
|
| 1788 |
-
"logps/rejected": -267.250244140625,
|
| 1789 |
-
"loss": 0.4814,
|
| 1790 |
-
"rewards/accuracies": 0.7416666746139526,
|
| 1791 |
-
"rewards/chosen": -11.756540298461914,
|
| 1792 |
-
"rewards/margins": 38.70287322998047,
|
| 1793 |
-
"rewards/rejected": -50.45941925048828,
|
| 1794 |
-
"step": 1180
|
| 1795 |
-
},
|
| 1796 |
-
{
|
| 1797 |
-
"epoch": 0.9340659340659341,
|
| 1798 |
-
"grad_norm": 5.963443433951819,
|
| 1799 |
-
"learning_rate": 6.599015837372907e-09,
|
| 1800 |
-
"logits/chosen": 6176.25244140625,
|
| 1801 |
-
"logits/rejected": 5420.193359375,
|
| 1802 |
-
"logps/chosen": -269.15130615234375,
|
| 1803 |
-
"logps/rejected": -276.1173095703125,
|
| 1804 |
-
"loss": 0.479,
|
| 1805 |
-
"rewards/accuracies": 0.699999988079071,
|
| 1806 |
-
"rewards/chosen": -20.287494659423828,
|
| 1807 |
-
"rewards/margins": 38.67988586425781,
|
| 1808 |
-
"rewards/rejected": -58.967376708984375,
|
| 1809 |
-
"step": 1190
|
| 1810 |
-
},
|
| 1811 |
-
{
|
| 1812 |
-
"epoch": 0.9419152276295133,
|
| 1813 |
-
"grad_norm": 6.813232634060442,
|
| 1814 |
-
"learning_rate": 5.126419011529992e-09,
|
| 1815 |
-
"logits/chosen": 6387.81396484375,
|
| 1816 |
-
"logits/rejected": 5468.4169921875,
|
| 1817 |
-
"logps/chosen": -266.6917724609375,
|
| 1818 |
-
"logps/rejected": -276.78839111328125,
|
| 1819 |
-
"loss": 0.4785,
|
| 1820 |
-
"rewards/accuracies": 0.8083333969116211,
|
| 1821 |
-
"rewards/chosen": -11.047189712524414,
|
| 1822 |
-
"rewards/margins": 47.05707931518555,
|
| 1823 |
-
"rewards/rejected": -58.104270935058594,
|
| 1824 |
-
"step": 1200
|
| 1825 |
-
},
|
| 1826 |
-
{
|
| 1827 |
-
"epoch": 0.9497645211930926,
|
| 1828 |
-
"grad_norm": 9.80121981882165,
|
| 1829 |
-
"learning_rate": 3.837833803870177e-09,
|
| 1830 |
-
"logits/chosen": 5973.63623046875,
|
| 1831 |
-
"logits/rejected": 5258.65087890625,
|
| 1832 |
-
"logps/chosen": -252.6921844482422,
|
| 1833 |
-
"logps/rejected": -273.18035888671875,
|
| 1834 |
-
"loss": 0.4811,
|
| 1835 |
-
"rewards/accuracies": 0.7750000357627869,
|
| 1836 |
-
"rewards/chosen": -12.484926223754883,
|
| 1837 |
-
"rewards/margins": 42.53102493286133,
|
| 1838 |
-
"rewards/rejected": -55.01594924926758,
|
| 1839 |
-
"step": 1210
|
| 1840 |
-
},
|
| 1841 |
-
{
|
| 1842 |
-
"epoch": 0.957613814756672,
|
| 1843 |
-
"grad_norm": 7.203558153981481,
|
| 1844 |
-
"learning_rate": 2.734228528934679e-09,
|
| 1845 |
-
"logits/chosen": 7448.9443359375,
|
| 1846 |
-
"logits/rejected": 5512.33203125,
|
| 1847 |
-
"logps/chosen": -312.1805725097656,
|
| 1848 |
-
"logps/rejected": -300.3201904296875,
|
| 1849 |
-
"loss": 0.4817,
|
| 1850 |
-
"rewards/accuracies": 0.7500000596046448,
|
| 1851 |
-
"rewards/chosen": -15.350790023803711,
|
| 1852 |
-
"rewards/margins": 39.79079055786133,
|
| 1853 |
-
"rewards/rejected": -55.14158248901367,
|
| 1854 |
-
"step": 1220
|
| 1855 |
-
},
|
| 1856 |
-
{
|
| 1857 |
-
"epoch": 0.9654631083202512,
|
| 1858 |
-
"grad_norm": 6.776742127412717,
|
| 1859 |
-
"learning_rate": 1.8164324970625645e-09,
|
| 1860 |
-
"logits/chosen": 6630.1591796875,
|
| 1861 |
-
"logits/rejected": 5257.56298828125,
|
| 1862 |
-
"logps/chosen": -270.0474548339844,
|
| 1863 |
-
"logps/rejected": -263.6984558105469,
|
| 1864 |
-
"loss": 0.4803,
|
| 1865 |
-
"rewards/accuracies": 0.7666666507720947,
|
| 1866 |
-
"rewards/chosen": -9.480215072631836,
|
| 1867 |
-
"rewards/margins": 41.2264289855957,
|
| 1868 |
-
"rewards/rejected": -50.706642150878906,
|
| 1869 |
-
"step": 1230
|
| 1870 |
-
},
|
| 1871 |
-
{
|
| 1872 |
-
"epoch": 0.9733124018838305,
|
| 1873 |
-
"grad_norm": 5.426445939101484,
|
| 1874 |
-
"learning_rate": 1.0851353912008642e-09,
|
| 1875 |
-
"logits/chosen": 5712.93408203125,
|
| 1876 |
-
"logits/rejected": 5265.7021484375,
|
| 1877 |
-
"logps/chosen": -247.56008911132812,
|
| 1878 |
-
"logps/rejected": -291.08746337890625,
|
| 1879 |
-
"loss": 0.4786,
|
| 1880 |
-
"rewards/accuracies": 0.7333333492279053,
|
| 1881 |
-
"rewards/chosen": -15.966026306152344,
|
| 1882 |
-
"rewards/margins": 39.92435073852539,
|
| 1883 |
-
"rewards/rejected": -55.890380859375,
|
| 1884 |
-
"step": 1240
|
| 1885 |
-
},
|
| 1886 |
-
{
|
| 1887 |
-
"epoch": 0.9811616954474097,
|
| 1888 |
-
"grad_norm": 5.411171726240316,
|
| 1889 |
-
"learning_rate": 5.408867486384471e-10,
|
| 1890 |
-
"logits/chosen": 5824.78125,
|
| 1891 |
-
"logits/rejected": 4941.5732421875,
|
| 1892 |
-
"logps/chosen": -240.45315551757812,
|
| 1893 |
-
"logps/rejected": -232.99081420898438,
|
| 1894 |
-
"loss": 0.4813,
|
| 1895 |
-
"rewards/accuracies": 0.7583333253860474,
|
| 1896 |
-
"rewards/chosen": -10.121699333190918,
|
| 1897 |
-
"rewards/margins": 33.4198112487793,
|
| 1898 |
-
"rewards/rejected": -43.54151153564453,
|
| 1899 |
-
"step": 1250
|
| 1900 |
-
},
|
| 1901 |
-
{
|
| 1902 |
-
"epoch": 0.989010989010989,
|
| 1903 |
-
"grad_norm": 6.255102070301854,
|
| 1904 |
-
"learning_rate": 1.840955480532924e-10,
|
| 1905 |
-
"logits/chosen": 5502.2841796875,
|
| 1906 |
-
"logits/rejected": 5238.97265625,
|
| 1907 |
-
"logps/chosen": -245.449951171875,
|
| 1908 |
-
"logps/rejected": -261.3154296875,
|
| 1909 |
-
"loss": 0.4794,
|
| 1910 |
-
"rewards/accuracies": 0.8083333969116211,
|
| 1911 |
-
"rewards/chosen": -13.299115180969238,
|
| 1912 |
-
"rewards/margins": 31.690093994140625,
|
| 1913 |
-
"rewards/rejected": -44.98920822143555,
|
| 1914 |
-
"step": 1260
|
| 1915 |
-
},
|
| 1916 |
-
{
|
| 1917 |
-
"epoch": 0.9968602825745683,
|
| 1918 |
-
"grad_norm": 6.339307198379936,
|
| 1919 |
-
"learning_rate": 1.502990218302247e-11,
|
| 1920 |
-
"logits/chosen": 5779.14501953125,
|
| 1921 |
-
"logits/rejected": 4718.3759765625,
|
| 1922 |
-
"logps/chosen": -237.53652954101562,
|
| 1923 |
-
"logps/rejected": -237.6223907470703,
|
| 1924 |
-
"loss": 0.4789,
|
| 1925 |
-
"rewards/accuracies": 0.7583333253860474,
|
| 1926 |
-
"rewards/chosen": -13.04780101776123,
|
| 1927 |
-
"rewards/margins": 38.266212463378906,
|
| 1928 |
-
"rewards/rejected": -51.31401443481445,
|
| 1929 |
-
"step": 1270
|
| 1930 |
},
|
| 1931 |
{
|
| 1932 |
-
"epoch":
|
| 1933 |
-
"step":
|
| 1934 |
"total_flos": 0.0,
|
| 1935 |
-
"train_loss": 0.
|
| 1936 |
-
"train_runtime":
|
| 1937 |
-
"train_samples_per_second":
|
| 1938 |
-
"train_steps_per_second": 0.
|
| 1939 |
}
|
| 1940 |
],
|
| 1941 |
"logging_steps": 10,
|
| 1942 |
-
"max_steps":
|
| 1943 |
"num_input_tokens_seen": 0,
|
| 1944 |
"num_train_epochs": 1,
|
| 1945 |
"save_steps": 1000,
|
|
@@ -1949,14 +749,14 @@
|
|
| 1949 |
"should_epoch_stop": false,
|
| 1950 |
"should_evaluate": false,
|
| 1951 |
"should_log": false,
|
| 1952 |
-
"should_save":
|
| 1953 |
"should_training_stop": false
|
| 1954 |
},
|
| 1955 |
"attributes": {}
|
| 1956 |
}
|
| 1957 |
},
|
| 1958 |
"total_flos": 0.0,
|
| 1959 |
-
"train_batch_size":
|
| 1960 |
"trial_name": null,
|
| 1961 |
"trial_params": null
|
| 1962 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.998691442030882,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 477,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.002093692750588851,
|
| 13 |
+
"grad_norm": 1.7795599560645496,
|
| 14 |
+
"learning_rate": 1.0416666666666666e-08,
|
| 15 |
+
"logits/chosen": 4963.591796875,
|
| 16 |
+
"logits/rejected": 4602.0634765625,
|
| 17 |
+
"logps/chosen": -222.9984130859375,
|
| 18 |
+
"logps/rejected": -158.74850463867188,
|
| 19 |
"loss": 0.5,
|
| 20 |
"rewards/accuracies": 0.0,
|
| 21 |
"rewards/chosen": 0.0,
|
|
|
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 0.02093692750588851,
|
| 28 |
+
"grad_norm": 1.664277641531783,
|
| 29 |
+
"learning_rate": 1.0416666666666667e-07,
|
| 30 |
+
"logits/chosen": 6011.06005859375,
|
| 31 |
+
"logits/rejected": 5064.162109375,
|
| 32 |
+
"logps/chosen": -240.0664825439453,
|
| 33 |
+
"logps/rejected": -208.95372009277344,
|
| 34 |
"loss": 0.5,
|
| 35 |
+
"rewards/accuracies": 0.5034722089767456,
|
| 36 |
+
"rewards/chosen": 0.00021603053028229624,
|
| 37 |
+
"rewards/margins": 0.0005849565495736897,
|
| 38 |
+
"rewards/rejected": -0.0003689260338433087,
|
| 39 |
"step": 10
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.04187385501177702,
|
| 43 |
+
"grad_norm": 1.6870127509120665,
|
| 44 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 45 |
+
"logits/chosen": 5908.4755859375,
|
| 46 |
+
"logits/rejected": 4882.568359375,
|
| 47 |
+
"logps/chosen": -231.4028778076172,
|
| 48 |
+
"logps/rejected": -196.9695281982422,
|
| 49 |
+
"loss": 0.4999,
|
| 50 |
+
"rewards/accuracies": 0.5531250238418579,
|
| 51 |
+
"rewards/chosen": 0.0007881122874096036,
|
| 52 |
+
"rewards/margins": 0.0005757558392360806,
|
| 53 |
+
"rewards/rejected": 0.00021235644817352295,
|
| 54 |
"step": 20
|
| 55 |
},
|
| 56 |
{
|
| 57 |
+
"epoch": 0.06281078251766553,
|
| 58 |
+
"grad_norm": 1.6176895769159747,
|
| 59 |
+
"learning_rate": 3.1249999999999997e-07,
|
| 60 |
+
"logits/chosen": 5731.59716796875,
|
| 61 |
+
"logits/rejected": 4970.974609375,
|
| 62 |
+
"logps/chosen": -205.80563354492188,
|
| 63 |
+
"logps/rejected": -182.66543579101562,
|
| 64 |
+
"loss": 0.4996,
|
| 65 |
+
"rewards/accuracies": 0.625,
|
| 66 |
+
"rewards/chosen": 0.0034130061976611614,
|
| 67 |
+
"rewards/margins": 0.003777171252295375,
|
| 68 |
+
"rewards/rejected": -0.0003641647635959089,
|
| 69 |
"step": 30
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.08374771002355404,
|
| 73 |
+
"grad_norm": 1.7199138258926487,
|
| 74 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 75 |
+
"logits/chosen": 5881.6220703125,
|
| 76 |
+
"logits/rejected": 5133.04638671875,
|
| 77 |
+
"logps/chosen": -199.26327514648438,
|
| 78 |
+
"logps/rejected": -189.4043426513672,
|
| 79 |
+
"loss": 0.4984,
|
| 80 |
+
"rewards/accuracies": 0.625,
|
| 81 |
+
"rewards/chosen": 0.0059957681223750114,
|
| 82 |
+
"rewards/margins": 0.01181001029908657,
|
| 83 |
+
"rewards/rejected": -0.005814241245388985,
|
| 84 |
"step": 40
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 0.10468463752944256,
|
| 88 |
+
"grad_norm": 2.335343455285067,
|
| 89 |
+
"learning_rate": 4.999731868769026e-07,
|
| 90 |
+
"logits/chosen": 6244.0546875,
|
| 91 |
+
"logits/rejected": 5419.8291015625,
|
| 92 |
+
"logps/chosen": -200.3682403564453,
|
| 93 |
+
"logps/rejected": -195.88189697265625,
|
| 94 |
+
"loss": 0.4965,
|
| 95 |
+
"rewards/accuracies": 0.640625,
|
| 96 |
+
"rewards/chosen": -0.022861136123538017,
|
| 97 |
+
"rewards/margins": 0.02738131210207939,
|
| 98 |
+
"rewards/rejected": -0.05024244636297226,
|
| 99 |
"step": 50
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 0.12562156503533106,
|
| 103 |
+
"grad_norm": 1.985885842678977,
|
| 104 |
+
"learning_rate": 4.990353313429303e-07,
|
| 105 |
+
"logits/chosen": 6085.1591796875,
|
| 106 |
+
"logits/rejected": 5170.4423828125,
|
| 107 |
+
"logps/chosen": -202.0069122314453,
|
| 108 |
+
"logps/rejected": -189.2211456298828,
|
| 109 |
+
"loss": 0.494,
|
| 110 |
+
"rewards/accuracies": 0.640625,
|
| 111 |
+
"rewards/chosen": -0.06540889292955399,
|
| 112 |
+
"rewards/margins": 0.07115975767374039,
|
| 113 |
+
"rewards/rejected": -0.13656863570213318,
|
| 114 |
"step": 60
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.14655849254121958,
|
| 118 |
+
"grad_norm": 2.02720438835768,
|
| 119 |
+
"learning_rate": 4.967625656594781e-07,
|
| 120 |
+
"logits/chosen": 5949.5009765625,
|
| 121 |
+
"logits/rejected": 5459.0986328125,
|
| 122 |
+
"logps/chosen": -203.56887817382812,
|
| 123 |
+
"logps/rejected": -203.3651885986328,
|
| 124 |
+
"loss": 0.493,
|
| 125 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 126 |
+
"rewards/chosen": -0.06014765053987503,
|
| 127 |
+
"rewards/margins": 0.07473154366016388,
|
| 128 |
+
"rewards/rejected": -0.1348791867494583,
|
| 129 |
"step": 70
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.16749542004710807,
|
| 133 |
+
"grad_norm": 2.2298529628194035,
|
| 134 |
+
"learning_rate": 4.93167072587771e-07,
|
| 135 |
+
"logits/chosen": 6108.1171875,
|
| 136 |
+
"logits/rejected": 5237.1611328125,
|
| 137 |
+
"logps/chosen": -200.6221160888672,
|
| 138 |
+
"logps/rejected": -197.19618225097656,
|
| 139 |
+
"loss": 0.4914,
|
| 140 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 141 |
+
"rewards/chosen": -0.08121322095394135,
|
| 142 |
+
"rewards/margins": 0.09713159501552582,
|
| 143 |
+
"rewards/rejected": -0.17834481596946716,
|
| 144 |
"step": 80
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.1884323475529966,
|
| 148 |
+
"grad_norm": 2.2234876584054275,
|
| 149 |
+
"learning_rate": 4.882681251368548e-07,
|
| 150 |
+
"logits/chosen": 6233.083984375,
|
| 151 |
+
"logits/rejected": 5252.95068359375,
|
| 152 |
+
"logps/chosen": -201.39984130859375,
|
| 153 |
+
"logps/rejected": -200.0003204345703,
|
| 154 |
+
"loss": 0.4901,
|
| 155 |
+
"rewards/accuracies": 0.6656249761581421,
|
| 156 |
+
"rewards/chosen": -0.08355093002319336,
|
| 157 |
+
"rewards/margins": 0.12262304872274399,
|
| 158 |
+
"rewards/rejected": -0.20617397129535675,
|
| 159 |
"step": 90
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 0.2093692750588851,
|
| 163 |
+
"grad_norm": 2.1862542481259792,
|
| 164 |
+
"learning_rate": 4.820919832540181e-07,
|
| 165 |
+
"logits/chosen": 6330.16259765625,
|
| 166 |
+
"logits/rejected": 5832.1982421875,
|
| 167 |
+
"logps/chosen": -203.98583984375,
|
| 168 |
+
"logps/rejected": -211.58010864257812,
|
| 169 |
+
"loss": 0.491,
|
| 170 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 171 |
+
"rewards/chosen": -0.07759369909763336,
|
| 172 |
+
"rewards/margins": 0.12538166344165802,
|
| 173 |
+
"rewards/rejected": -0.20297536253929138,
|
| 174 |
"step": 100
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 0.23030620256477363,
|
| 178 |
+
"grad_norm": 2.7054972610472565,
|
| 179 |
+
"learning_rate": 4.7467175306295647e-07,
|
| 180 |
+
"logits/chosen": 6421.41015625,
|
| 181 |
+
"logits/rejected": 5193.19287109375,
|
| 182 |
+
"logps/chosen": -198.57032775878906,
|
| 183 |
+
"logps/rejected": -209.23104858398438,
|
| 184 |
+
"loss": 0.4869,
|
| 185 |
+
"rewards/accuracies": 0.721875011920929,
|
| 186 |
+
"rewards/chosen": -0.0885535329580307,
|
| 187 |
+
"rewards/margins": 0.1617024838924408,
|
| 188 |
+
"rewards/rejected": -0.2502560019493103,
|
| 189 |
"step": 110
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 0.2512431300706621,
|
| 193 |
+
"grad_norm": 2.490243429862705,
|
| 194 |
+
"learning_rate": 4.6604720940421207e-07,
|
| 195 |
+
"logits/chosen": 6070.7314453125,
|
| 196 |
+
"logits/rejected": 5633.78369140625,
|
| 197 |
+
"logps/chosen": -196.9094696044922,
|
| 198 |
+
"logps/rejected": -213.2709197998047,
|
| 199 |
+
"loss": 0.4892,
|
| 200 |
+
"rewards/accuracies": 0.6781250238418579,
|
| 201 |
+
"rewards/chosen": -0.10950146615505219,
|
| 202 |
+
"rewards/margins": 0.1391303390264511,
|
| 203 |
+
"rewards/rejected": -0.2486318051815033,
|
| 204 |
"step": 120
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 0.2721800575765506,
|
| 208 |
+
"grad_norm": 2.7610246742258715,
|
| 209 |
+
"learning_rate": 4.5626458262912735e-07,
|
| 210 |
+
"logits/chosen": 6097.5146484375,
|
| 211 |
+
"logits/rejected": 5424.9013671875,
|
| 212 |
+
"logps/chosen": -197.1998748779297,
|
| 213 |
+
"logps/rejected": -216.65188598632812,
|
| 214 |
+
"loss": 0.4879,
|
| 215 |
+
"rewards/accuracies": 0.715624988079071,
|
| 216 |
+
"rewards/chosen": -0.08192776143550873,
|
| 217 |
+
"rewards/margins": 0.15857011079788208,
|
| 218 |
+
"rewards/rejected": -0.2404978722333908,
|
| 219 |
"step": 130
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.29311698508243916,
|
| 223 |
+
"grad_norm": 2.5211223485513323,
|
| 224 |
+
"learning_rate": 4.453763107901675e-07,
|
| 225 |
+
"logits/chosen": 5834.34814453125,
|
| 226 |
+
"logits/rejected": 5199.22119140625,
|
| 227 |
+
"logps/chosen": -204.2546844482422,
|
| 228 |
+
"logps/rejected": -208.1846923828125,
|
| 229 |
+
"loss": 0.4883,
|
| 230 |
+
"rewards/accuracies": 0.684374988079071,
|
| 231 |
+
"rewards/chosen": -0.10303564369678497,
|
| 232 |
+
"rewards/margins": 0.19196416437625885,
|
| 233 |
+
"rewards/rejected": -0.29499977827072144,
|
| 234 |
"step": 140
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.31405391258832765,
|
| 238 |
+
"grad_norm": 2.8015563362051052,
|
| 239 |
+
"learning_rate": 4.3344075855595097e-07,
|
| 240 |
+
"logits/chosen": 6411.4541015625,
|
| 241 |
+
"logits/rejected": 5283.06689453125,
|
| 242 |
+
"logps/chosen": -216.62496948242188,
|
| 243 |
+
"logps/rejected": -208.1221466064453,
|
| 244 |
+
"loss": 0.4855,
|
| 245 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 246 |
+
"rewards/chosen": -0.07201126962900162,
|
| 247 |
+
"rewards/margins": 0.21424655616283417,
|
| 248 |
+
"rewards/rejected": -0.286257803440094,
|
| 249 |
"step": 150
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 0.33499084009421615,
|
| 253 |
+
"grad_norm": 3.380927510289668,
|
| 254 |
+
"learning_rate": 4.2052190435769554e-07,
|
| 255 |
+
"logits/chosen": 6130.6318359375,
|
| 256 |
+
"logits/rejected": 5110.82080078125,
|
| 257 |
+
"logps/chosen": -209.1565704345703,
|
| 258 |
+
"logps/rejected": -219.35220336914062,
|
| 259 |
+
"loss": 0.4824,
|
| 260 |
+
"rewards/accuracies": 0.7281249761581421,
|
| 261 |
+
"rewards/chosen": -0.10825005918741226,
|
| 262 |
+
"rewards/margins": 0.2637324929237366,
|
| 263 |
+
"rewards/rejected": -0.37198254466056824,
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 0.3559277676001047,
|
| 268 |
+
"grad_norm": 3.2112846087091484,
|
| 269 |
+
"learning_rate": 4.0668899744407567e-07,
|
| 270 |
+
"logits/chosen": 6476.86767578125,
|
| 271 |
+
"logits/rejected": 5530.86767578125,
|
| 272 |
+
"logps/chosen": -220.4525146484375,
|
| 273 |
+
"logps/rejected": -208.67733764648438,
|
| 274 |
+
"loss": 0.4833,
|
| 275 |
+
"rewards/accuracies": 0.703125,
|
| 276 |
+
"rewards/chosen": -0.12977644801139832,
|
| 277 |
+
"rewards/margins": 0.23194870352745056,
|
| 278 |
+
"rewards/rejected": -0.3617251515388489,
|
| 279 |
"step": 170
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 0.3768646951059932,
|
| 283 |
+
"grad_norm": 2.5422470233501713,
|
| 284 |
+
"learning_rate": 3.920161866827889e-07,
|
| 285 |
+
"logits/chosen": 6021.7919921875,
|
| 286 |
+
"logits/rejected": 5178.1259765625,
|
| 287 |
+
"logps/chosen": -220.57601928710938,
|
| 288 |
+
"logps/rejected": -207.26339721679688,
|
| 289 |
+
"loss": 0.4847,
|
| 290 |
+
"rewards/accuracies": 0.71875,
|
| 291 |
+
"rewards/chosen": -0.09060440957546234,
|
| 292 |
+
"rewards/margins": 0.20774026215076447,
|
| 293 |
+
"rewards/rejected": -0.2983446717262268,
|
| 294 |
"step": 180
|
| 295 |
},
|
| 296 |
{
|
| 297 |
+
"epoch": 0.39780162261188173,
|
| 298 |
+
"grad_norm": 5.303607389601098,
|
| 299 |
+
"learning_rate": 3.765821230985757e-07,
|
| 300 |
+
"logits/chosen": 6147.5517578125,
|
| 301 |
+
"logits/rejected": 5167.7958984375,
|
| 302 |
+
"logps/chosen": -208.33944702148438,
|
| 303 |
+
"logps/rejected": -221.4107208251953,
|
| 304 |
+
"loss": 0.4836,
|
| 305 |
+
"rewards/accuracies": 0.7406250238418579,
|
| 306 |
+
"rewards/chosen": -0.1318391114473343,
|
| 307 |
+
"rewards/margins": 0.3124375343322754,
|
| 308 |
+
"rewards/rejected": -0.4442766308784485,
|
| 309 |
"step": 190
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 0.4187385501177702,
|
| 313 |
+
"grad_norm": 2.6432475137123794,
|
| 314 |
+
"learning_rate": 3.604695382782159e-07,
|
| 315 |
+
"logits/chosen": 5748.025390625,
|
| 316 |
+
"logits/rejected": 4992.97021484375,
|
| 317 |
+
"logps/chosen": -193.7339630126953,
|
| 318 |
+
"logps/rejected": -207.1187286376953,
|
| 319 |
+
"loss": 0.4868,
|
| 320 |
+
"rewards/accuracies": 0.703125,
|
| 321 |
+
"rewards/chosen": -0.08920423686504364,
|
| 322 |
+
"rewards/margins": 0.2186780869960785,
|
| 323 |
+
"rewards/rejected": -0.30788230895996094,
|
| 324 |
"step": 200
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.4396754776236587,
|
| 328 |
+
"grad_norm": 3.3944179691537864,
|
| 329 |
+
"learning_rate": 3.4376480090239047e-07,
|
| 330 |
+
"logits/chosen": 6434.7412109375,
|
| 331 |
+
"logits/rejected": 5470.30224609375,
|
| 332 |
+
"logps/chosen": -219.40087890625,
|
| 333 |
+
"logps/rejected": -221.01614379882812,
|
| 334 |
+
"loss": 0.4847,
|
| 335 |
+
"rewards/accuracies": 0.746874988079071,
|
| 336 |
+
"rewards/chosen": -0.07487013936042786,
|
| 337 |
+
"rewards/margins": 0.21102896332740784,
|
| 338 |
+
"rewards/rejected": -0.2858991026878357,
|
| 339 |
"step": 210
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 0.46061240512954726,
|
| 343 |
+
"grad_norm": 2.9009193632951207,
|
| 344 |
+
"learning_rate": 3.265574537815398e-07,
|
| 345 |
+
"logits/chosen": 5810.1513671875,
|
| 346 |
+
"logits/rejected": 5406.72021484375,
|
| 347 |
+
"logps/chosen": -212.43923950195312,
|
| 348 |
+
"logps/rejected": -227.1441192626953,
|
| 349 |
+
"loss": 0.4829,
|
| 350 |
+
"rewards/accuracies": 0.753125011920929,
|
| 351 |
+
"rewards/chosen": -0.105693019926548,
|
| 352 |
+
"rewards/margins": 0.26153022050857544,
|
| 353 |
+
"rewards/rejected": -0.36722320318222046,
|
| 354 |
"step": 220
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 0.48154933263543576,
|
| 358 |
+
"grad_norm": 2.6189951178492614,
|
| 359 |
+
"learning_rate": 3.0893973387735683e-07,
|
| 360 |
+
"logits/chosen": 6018.71484375,
|
| 361 |
+
"logits/rejected": 5094.3134765625,
|
| 362 |
+
"logps/chosen": -207.9360809326172,
|
| 363 |
+
"logps/rejected": -231.08480834960938,
|
| 364 |
+
"loss": 0.4805,
|
| 365 |
+
"rewards/accuracies": 0.78125,
|
| 366 |
+
"rewards/chosen": -0.13205979764461517,
|
| 367 |
+
"rewards/margins": 0.36629724502563477,
|
| 368 |
+
"rewards/rejected": -0.49835705757141113,
|
| 369 |
"step": 230
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 0.5024862601413242,
|
| 373 |
+
"grad_norm": 2.9120071931474514,
|
| 374 |
+
"learning_rate": 2.910060778827554e-07,
|
| 375 |
+
"logits/chosen": 6167.8330078125,
|
| 376 |
+
"logits/rejected": 5222.9423828125,
|
| 377 |
+
"logps/chosen": -206.72628784179688,
|
| 378 |
+
"logps/rejected": -222.31649780273438,
|
| 379 |
+
"loss": 0.4816,
|
| 380 |
+
"rewards/accuracies": 0.737500011920929,
|
| 381 |
+
"rewards/chosen": -0.10643552243709564,
|
| 382 |
+
"rewards/margins": 0.27724453806877136,
|
| 383 |
+
"rewards/rejected": -0.3836800754070282,
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 0.5234231876472127,
|
| 388 |
+
"grad_norm": 3.640526921626944,
|
| 389 |
+
"learning_rate": 2.7285261601056697e-07,
|
| 390 |
+
"logits/chosen": 6262.5078125,
|
| 391 |
+
"logits/rejected": 5280.09619140625,
|
| 392 |
+
"logps/chosen": -224.32699584960938,
|
| 393 |
+
"logps/rejected": -236.7519989013672,
|
| 394 |
+
"loss": 0.4832,
|
| 395 |
+
"rewards/accuracies": 0.7437499761581421,
|
| 396 |
+
"rewards/chosen": -0.11429344117641449,
|
| 397 |
+
"rewards/margins": 0.28984737396240234,
|
| 398 |
+
"rewards/rejected": -0.4041408598423004,
|
| 399 |
"step": 250
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 0.5443601151531012,
|
| 403 |
+
"grad_norm": 3.480502498738401,
|
| 404 |
+
"learning_rate": 2.5457665670441937e-07,
|
| 405 |
+
"logits/chosen": 6063.48681640625,
|
| 406 |
+
"logits/rejected": 5488.103515625,
|
| 407 |
+
"logps/chosen": -227.7340850830078,
|
| 408 |
+
"logps/rejected": -235.0940399169922,
|
| 409 |
+
"loss": 0.4823,
|
| 410 |
+
"rewards/accuracies": 0.746874988079071,
|
| 411 |
+
"rewards/chosen": -0.11375202983617783,
|
| 412 |
+
"rewards/margins": 0.27239111065864563,
|
| 413 |
+
"rewards/rejected": -0.38614311814308167,
|
| 414 |
"step": 260
|
| 415 |
},
|
| 416 |
{
|
| 417 |
+
"epoch": 0.5652970426589898,
|
| 418 |
+
"grad_norm": 3.100193815819693,
|
| 419 |
+
"learning_rate": 2.3627616503391812e-07,
|
| 420 |
+
"logits/chosen": 6130.5126953125,
|
| 421 |
+
"logits/rejected": 5626.0517578125,
|
| 422 |
+
"logps/chosen": -217.3065643310547,
|
| 423 |
+
"logps/rejected": -236.91329956054688,
|
| 424 |
+
"loss": 0.4834,
|
| 425 |
+
"rewards/accuracies": 0.753125011920929,
|
| 426 |
+
"rewards/chosen": -0.1151232123374939,
|
| 427 |
+
"rewards/margins": 0.2929432988166809,
|
| 428 |
+
"rewards/rejected": -0.40806645154953003,
|
| 429 |
"step": 270
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.5862339701648783,
|
| 433 |
+
"grad_norm": 3.6139410491493034,
|
| 434 |
+
"learning_rate": 2.1804923757009882e-07,
|
| 435 |
+
"logits/chosen": 6155.16357421875,
|
| 436 |
+
"logits/rejected": 5117.0361328125,
|
| 437 |
+
"logps/chosen": -224.9839630126953,
|
| 438 |
+
"logps/rejected": -229.80514526367188,
|
| 439 |
+
"loss": 0.4807,
|
| 440 |
+
"rewards/accuracies": 0.7718750238418579,
|
| 441 |
+
"rewards/chosen": -0.10252990573644638,
|
| 442 |
+
"rewards/margins": 0.33688706159591675,
|
| 443 |
+
"rewards/rejected": -0.4394169747829437,
|
| 444 |
"step": 280
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 0.6071708976707668,
|
| 448 |
+
"grad_norm": 3.0356982001656423,
|
| 449 |
+
"learning_rate": 1.9999357655598891e-07,
|
| 450 |
+
"logits/chosen": 6409.4736328125,
|
| 451 |
+
"logits/rejected": 6006.4580078125,
|
| 452 |
+
"logps/chosen": -239.07656860351562,
|
| 453 |
+
"logps/rejected": -265.92523193359375,
|
| 454 |
+
"loss": 0.4842,
|
| 455 |
+
"rewards/accuracies": 0.6968749761581421,
|
| 456 |
+
"rewards/chosen": -0.15593001246452332,
|
| 457 |
+
"rewards/margins": 0.2926904857158661,
|
| 458 |
+
"rewards/rejected": -0.4486205577850342,
|
| 459 |
"step": 290
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 0.6281078251766553,
|
| 463 |
+
"grad_norm": 4.271112938322883,
|
| 464 |
+
"learning_rate": 1.8220596619089573e-07,
|
| 465 |
+
"logits/chosen": 5861.82177734375,
|
| 466 |
+
"logits/rejected": 5508.8740234375,
|
| 467 |
+
"logps/chosen": -222.1393585205078,
|
| 468 |
+
"logps/rejected": -245.2710418701172,
|
| 469 |
+
"loss": 0.4845,
|
| 470 |
+
"rewards/accuracies": 0.75,
|
| 471 |
+
"rewards/chosen": -0.11530385166406631,
|
| 472 |
+
"rewards/margins": 0.2702717185020447,
|
| 473 |
+
"rewards/rejected": -0.3855755925178528,
|
| 474 |
"step": 300
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 0.6490447526825438,
|
| 478 |
+
"grad_norm": 3.0832085833192004,
|
| 479 |
+
"learning_rate": 1.647817538357072e-07,
|
| 480 |
+
"logits/chosen": 6551.92529296875,
|
| 481 |
+
"logits/rejected": 5745.2666015625,
|
| 482 |
+
"logps/chosen": -234.02554321289062,
|
| 483 |
+
"logps/rejected": -243.51119995117188,
|
| 484 |
+
"loss": 0.4815,
|
| 485 |
+
"rewards/accuracies": 0.7281249761581421,
|
| 486 |
+
"rewards/chosen": -0.10852447897195816,
|
| 487 |
+
"rewards/margins": 0.22780513763427734,
|
| 488 |
+
"rewards/rejected": -0.3363296091556549,
|
| 489 |
"step": 310
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 0.6699816801884323,
|
| 493 |
+
"grad_norm": 5.627304230331988,
|
| 494 |
+
"learning_rate": 1.478143389201113e-07,
|
| 495 |
+
"logits/chosen": 6057.1337890625,
|
| 496 |
+
"logits/rejected": 5255.4736328125,
|
| 497 |
+
"logps/chosen": -222.4619903564453,
|
| 498 |
+
"logps/rejected": -239.05459594726562,
|
| 499 |
+
"loss": 0.4817,
|
| 500 |
+
"rewards/accuracies": 0.746874988079071,
|
| 501 |
+
"rewards/chosen": -0.09626330435276031,
|
| 502 |
+
"rewards/margins": 0.28026360273361206,
|
| 503 |
+
"rewards/rejected": -0.37652695178985596,
|
| 504 |
"step": 320
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 0.6909186076943209,
|
| 508 |
+
"grad_norm": 4.070544491993295,
|
| 509 |
+
"learning_rate": 1.3139467229135998e-07,
|
| 510 |
+
"logits/chosen": 5855.13671875,
|
| 511 |
+
"logits/rejected": 5428.80224609375,
|
| 512 |
+
"logps/chosen": -231.0357208251953,
|
| 513 |
+
"logps/rejected": -246.7843780517578,
|
| 514 |
+
"loss": 0.484,
|
| 515 |
+
"rewards/accuracies": 0.768750011920929,
|
| 516 |
+
"rewards/chosen": -0.11221712827682495,
|
| 517 |
+
"rewards/margins": 0.30028775334358215,
|
| 518 |
+
"rewards/rejected": -0.4125048518180847,
|
| 519 |
"step": 330
|
| 520 |
},
|
| 521 |
{
|
| 522 |
+
"epoch": 0.7118555352002094,
|
| 523 |
+
"grad_norm": 4.9587556176352905,
|
| 524 |
+
"learning_rate": 1.1561076868822755e-07,
|
| 525 |
+
"logits/chosen": 5686.6806640625,
|
| 526 |
+
"logits/rejected": 4937.55615234375,
|
| 527 |
+
"logps/chosen": -216.03945922851562,
|
| 528 |
+
"logps/rejected": -238.09805297851562,
|
| 529 |
+
"loss": 0.4819,
|
| 530 |
+
"rewards/accuracies": 0.7593749761581421,
|
| 531 |
+
"rewards/chosen": -0.16062280535697937,
|
| 532 |
+
"rewards/margins": 0.34666866064071655,
|
| 533 |
+
"rewards/rejected": -0.5072914958000183,
|
| 534 |
"step": 340
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.7327924627060979,
|
| 538 |
+
"grad_norm": 3.8596857041552393,
|
| 539 |
+
"learning_rate": 1.0054723495346482e-07,
|
| 540 |
+
"logits/chosen": 6366.03955078125,
|
| 541 |
+
"logits/rejected": 5157.458984375,
|
| 542 |
+
"logps/chosen": -255.4525604248047,
|
| 543 |
+
"logps/rejected": -263.61639404296875,
|
| 544 |
+
"loss": 0.4836,
|
| 545 |
+
"rewards/accuracies": 0.7437499761581421,
|
| 546 |
+
"rewards/chosen": -0.14537517726421356,
|
| 547 |
+
"rewards/margins": 0.3442801833152771,
|
| 548 |
+
"rewards/rejected": -0.48965534567832947,
|
| 549 |
"step": 350
|
| 550 |
},
|
| 551 |
{
|
| 552 |
+
"epoch": 0.7537293902119864,
|
| 553 |
+
"grad_norm": 4.404236860707489,
|
| 554 |
+
"learning_rate": 8.628481651367875e-08,
|
| 555 |
+
"logits/chosen": 6031.80615234375,
|
| 556 |
+
"logits/rejected": 5262.02734375,
|
| 557 |
+
"logps/chosen": -231.36318969726562,
|
| 558 |
+
"logps/rejected": -245.8846893310547,
|
| 559 |
+
"loss": 0.4812,
|
| 560 |
+
"rewards/accuracies": 0.765625,
|
| 561 |
+
"rewards/chosen": -0.10767906904220581,
|
| 562 |
+
"rewards/margins": 0.31250181794166565,
|
| 563 |
+
"rewards/rejected": -0.42018088698387146,
|
| 564 |
"step": 360
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 0.7746663177178749,
|
| 568 |
+
"grad_norm": 4.286082049703798,
|
| 569 |
+
"learning_rate": 7.289996455765748e-08,
|
| 570 |
+
"logits/chosen": 5882.796875,
|
| 571 |
+
"logits/rejected": 5024.01171875,
|
| 572 |
+
"logps/chosen": -250.63809204101562,
|
| 573 |
+
"logps/rejected": -253.8474884033203,
|
| 574 |
+
"loss": 0.4799,
|
| 575 |
+
"rewards/accuracies": 0.737500011920929,
|
| 576 |
+
"rewards/chosen": -0.12153621017932892,
|
| 577 |
+
"rewards/margins": 0.3065151572227478,
|
| 578 |
+
"rewards/rejected": -0.4280513823032379,
|
| 579 |
"step": 370
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 0.7956032452237635,
|
| 583 |
+
"grad_norm": 4.2771383904693545,
|
| 584 |
+
"learning_rate": 6.046442623320145e-08,
|
| 585 |
+
"logits/chosen": 6166.33447265625,
|
| 586 |
+
"logits/rejected": 5370.65185546875,
|
| 587 |
+
"logps/chosen": -244.5850067138672,
|
| 588 |
+
"logps/rejected": -238.40518188476562,
|
| 589 |
+
"loss": 0.4816,
|
| 590 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 591 |
+
"rewards/chosen": -0.10372950881719589,
|
| 592 |
+
"rewards/margins": 0.2737034261226654,
|
| 593 |
+
"rewards/rejected": -0.3774329721927643,
|
| 594 |
"step": 380
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 0.816540172729652,
|
| 598 |
+
"grad_norm": 4.7598934741028565,
|
| 599 |
+
"learning_rate": 4.904486005914027e-08,
|
| 600 |
+
"logits/chosen": 6437.08251953125,
|
| 601 |
+
"logits/rejected": 5671.439453125,
|
| 602 |
+
"logps/chosen": -249.28125,
|
| 603 |
+
"logps/rejected": -271.4849548339844,
|
| 604 |
+
"loss": 0.4818,
|
| 605 |
+
"rewards/accuracies": 0.753125011920929,
|
| 606 |
+
"rewards/chosen": -0.1305597722530365,
|
| 607 |
+
"rewards/margins": 0.3592670261859894,
|
| 608 |
+
"rewards/rejected": -0.4898267686367035,
|
| 609 |
"step": 390
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 0.8374771002355405,
|
| 613 |
+
"grad_norm": 3.0633087764980194,
|
| 614 |
+
"learning_rate": 3.8702478614051345e-08,
|
| 615 |
+
"logits/chosen": 5975.5859375,
|
| 616 |
+
"logits/rejected": 5307.0068359375,
|
| 617 |
+
"logps/chosen": -249.6516876220703,
|
| 618 |
+
"logps/rejected": -247.6414031982422,
|
| 619 |
+
"loss": 0.4821,
|
| 620 |
+
"rewards/accuracies": 0.75,
|
| 621 |
+
"rewards/chosen": -0.12782196700572968,
|
| 622 |
+
"rewards/margins": 0.30399927496910095,
|
| 623 |
+
"rewards/rejected": -0.4318212568759918,
|
| 624 |
"step": 400
|
| 625 |
},
|
| 626 |
{
|
| 627 |
+
"epoch": 0.8584140277414289,
|
| 628 |
+
"grad_norm": 5.414232808843427,
|
| 629 |
+
"learning_rate": 2.9492720416985e-08,
|
| 630 |
+
"logits/chosen": 5987.8955078125,
|
| 631 |
+
"logits/rejected": 5288.86669921875,
|
| 632 |
+
"logps/chosen": -239.4297637939453,
|
| 633 |
+
"logps/rejected": -252.67214965820312,
|
| 634 |
+
"loss": 0.4809,
|
| 635 |
+
"rewards/accuracies": 0.731249988079071,
|
| 636 |
+
"rewards/chosen": -0.14551712572574615,
|
| 637 |
+
"rewards/margins": 0.2954447865486145,
|
| 638 |
+
"rewards/rejected": -0.44096192717552185,
|
| 639 |
"step": 410
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.8793509552473174,
|
| 643 |
+
"grad_norm": 3.7651155714131734,
|
| 644 |
+
"learning_rate": 2.1464952759020856e-08,
|
| 645 |
+
"logits/chosen": 6339.23046875,
|
| 646 |
+
"logits/rejected": 5529.13037109375,
|
| 647 |
+
"logps/chosen": -254.3148193359375,
|
| 648 |
+
"logps/rejected": -267.2603759765625,
|
| 649 |
+
"loss": 0.481,
|
| 650 |
+
"rewards/accuracies": 0.734375,
|
| 651 |
+
"rewards/chosen": -0.12421433627605438,
|
| 652 |
+
"rewards/margins": 0.3070390820503235,
|
| 653 |
+
"rewards/rejected": -0.43125343322753906,
|
| 654 |
"step": 420
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 0.9002878827532059,
|
| 658 |
+
"grad_norm": 3.0456608776131104,
|
| 659 |
+
"learning_rate": 1.4662207078575684e-08,
|
| 660 |
+
"logits/chosen": 6244.36669921875,
|
| 661 |
+
"logits/rejected": 5038.8232421875,
|
| 662 |
+
"logps/chosen": -249.1537628173828,
|
| 663 |
+
"logps/rejected": -258.28509521484375,
|
| 664 |
+
"loss": 0.4806,
|
| 665 |
+
"rewards/accuracies": 0.765625,
|
| 666 |
+
"rewards/chosen": -0.11752188205718994,
|
| 667 |
+
"rewards/margins": 0.3456159830093384,
|
| 668 |
+
"rewards/rejected": -0.46313780546188354,
|
| 669 |
"step": 430
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 0.9212248102590945,
|
| 673 |
+
"grad_norm": 3.7152553962599533,
|
| 674 |
+
"learning_rate": 9.12094829893642e-09,
|
| 675 |
+
"logits/chosen": 6062.4091796875,
|
| 676 |
+
"logits/rejected": 5768.86572265625,
|
| 677 |
+
"logps/chosen": -247.6633758544922,
|
| 678 |
+
"logps/rejected": -270.07781982421875,
|
| 679 |
+
"loss": 0.4831,
|
| 680 |
+
"rewards/accuracies": 0.71875,
|
| 681 |
+
"rewards/chosen": -0.125531867146492,
|
| 682 |
+
"rewards/margins": 0.2730921804904938,
|
| 683 |
+
"rewards/rejected": -0.39862409234046936,
|
| 684 |
"step": 440
|
| 685 |
},
|
| 686 |
{
|
| 687 |
+
"epoch": 0.942161737764983,
|
| 688 |
+
"grad_norm": 3.385435845957556,
|
| 689 |
+
"learning_rate": 4.8708793644441086e-09,
|
| 690 |
+
"logits/chosen": 6208.9853515625,
|
| 691 |
+
"logits/rejected": 5350.4443359375,
|
| 692 |
+
"logps/chosen": -253.6076202392578,
|
| 693 |
+
"logps/rejected": -252.9407196044922,
|
| 694 |
+
"loss": 0.4804,
|
| 695 |
+
"rewards/accuracies": 0.753125011920929,
|
| 696 |
+
"rewards/chosen": -0.10648094117641449,
|
| 697 |
+
"rewards/margins": 0.2990611791610718,
|
| 698 |
+
"rewards/rejected": -0.40554213523864746,
|
| 699 |
"step": 450
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 0.9630986652708715,
|
| 703 |
+
"grad_norm": 3.849103407950397,
|
| 704 |
+
"learning_rate": 1.9347820230782295e-09,
|
| 705 |
+
"logits/chosen": 6395.10302734375,
|
| 706 |
+
"logits/rejected": 5409.55078125,
|
| 707 |
+
"logps/chosen": -276.77349853515625,
|
| 708 |
+
"logps/rejected": -277.45819091796875,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 709 |
"loss": 0.4832,
|
| 710 |
+
"rewards/accuracies": 0.7093750238418579,
|
| 711 |
+
"rewards/chosen": -0.13324801623821259,
|
| 712 |
+
"rewards/margins": 0.28933390974998474,
|
| 713 |
+
"rewards/rejected": -0.42258191108703613,
|
| 714 |
+
"step": 460
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 715 |
},
|
| 716 |
{
|
| 717 |
+
"epoch": 0.98403559277676,
|
| 718 |
+
"grad_norm": 3.425749585050412,
|
| 719 |
+
"learning_rate": 3.2839470889836627e-10,
|
| 720 |
+
"logits/chosen": 5962.32763671875,
|
| 721 |
+
"logits/rejected": 5300.6298828125,
|
| 722 |
+
"logps/chosen": -245.10165405273438,
|
| 723 |
+
"logps/rejected": -248.36233520507812,
|
| 724 |
"loss": 0.481,
|
| 725 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 726 |
+
"rewards/chosen": -0.1263255625963211,
|
| 727 |
+
"rewards/margins": 0.26447486877441406,
|
| 728 |
+
"rewards/rejected": -0.390800416469574,
|
| 729 |
+
"step": 470
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 730 |
},
|
| 731 |
{
|
| 732 |
+
"epoch": 0.998691442030882,
|
| 733 |
+
"step": 477,
|
| 734 |
"total_flos": 0.0,
|
| 735 |
+
"train_loss": 0.4856958399278813,
|
| 736 |
+
"train_runtime": 14922.9586,
|
| 737 |
+
"train_samples_per_second": 4.097,
|
| 738 |
+
"train_steps_per_second": 0.032
|
| 739 |
}
|
| 740 |
],
|
| 741 |
"logging_steps": 10,
|
| 742 |
+
"max_steps": 477,
|
| 743 |
"num_input_tokens_seen": 0,
|
| 744 |
"num_train_epochs": 1,
|
| 745 |
"save_steps": 1000,
|
|
|
|
| 749 |
"should_epoch_stop": false,
|
| 750 |
"should_evaluate": false,
|
| 751 |
"should_log": false,
|
| 752 |
+
"should_save": false,
|
| 753 |
"should_training_stop": false
|
| 754 |
},
|
| 755 |
"attributes": {}
|
| 756 |
}
|
| 757 |
},
|
| 758 |
"total_flos": 0.0,
|
| 759 |
+
"train_batch_size": 4,
|
| 760 |
"trial_name": null,
|
| 761 |
"trial_params": null
|
| 762 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7288
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:23b58b0ae630332eb83798c14135937fb5e53e833fb25ba6c89ad41ca9d5cdf3
|
| 3 |
size 7288
|