End of training
Browse files- README.md +40 -45
- adapter_model.safetensors +1 -1
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +496 -649
README.md
CHANGED
|
@@ -4,28 +4,19 @@ library_name: peft
|
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
base_model: mistralai/Mistral-7B-v0.1
|
| 7 |
-
metrics:
|
| 8 |
-
- accuracy
|
| 9 |
model-index:
|
| 10 |
-
- name: Mistral-7B-v0.
|
| 11 |
results: []
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
|
| 17 |
-
# Mistral-7B-v0.
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss:
|
| 22 |
-
- Precision Micro: 0.8142
|
| 23 |
-
- Precision Macro: 0.7222
|
| 24 |
-
- Recall Micro: 0.8142
|
| 25 |
-
- Recall Macro: 0.7126
|
| 26 |
-
- F1 Micro: 0.8142
|
| 27 |
-
- F1 Macro: 0.7098
|
| 28 |
-
- Accuracy: 0.8142
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
|
@@ -48,46 +39,50 @@ The following hyperparameters were used during training:
|
|
| 48 |
- train_batch_size: 4
|
| 49 |
- eval_batch_size: 4
|
| 50 |
- seed: 42
|
|
|
|
|
|
|
| 51 |
- gradient_accumulation_steps: 4
|
| 52 |
-
- total_train_batch_size:
|
|
|
|
| 53 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
- lr_scheduler_type: constant
|
| 55 |
- lr_scheduler_warmup_ratio: 0.03
|
| 56 |
-
-
|
| 57 |
|
| 58 |
### Training results
|
| 59 |
|
| 60 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 61 |
-
|
| 62 |
-
|
|
| 63 |
-
|
|
| 64 |
-
|
|
| 65 |
-
|
|
| 66 |
-
|
|
| 67 |
-
|
|
| 68 |
-
|
|
| 69 |
-
|
|
| 70 |
-
|
|
| 71 |
-
|
|
| 72 |
-
|
|
| 73 |
-
|
|
| 74 |
-
|
|
| 75 |
-
|
|
| 76 |
-
|
|
| 77 |
-
|
|
| 78 |
-
|
|
| 79 |
-
|
|
| 80 |
-
|
|
| 81 |
-
|
|
| 82 |
-
|
|
| 83 |
-
|
|
| 84 |
-
|
|
| 85 |
-
|
|
| 86 |
-
|
|
| 87 |
-
|
|
| 88 |
-
|
|
| 89 |
-
|
|
| 90 |
-
|
|
|
|
|
| 91 |
|
| 92 |
|
| 93 |
### Framework versions
|
|
|
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
base_model: mistralai/Mistral-7B-v0.1
|
|
|
|
|
|
|
| 7 |
model-index:
|
| 8 |
+
- name: Mistral-7B-v0.1_caselaw
|
| 9 |
results: []
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
|
| 15 |
+
# Mistral-7B-v0.1_caselaw
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.1640
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 39 |
- train_batch_size: 4
|
| 40 |
- eval_batch_size: 4
|
| 41 |
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 4
|
| 44 |
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 64
|
| 46 |
+
- total_eval_batch_size: 16
|
| 47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
- lr_scheduler_type: constant
|
| 49 |
- lr_scheduler_warmup_ratio: 0.03
|
| 50 |
+
- num_epochs: 2.0
|
| 51 |
|
| 52 |
### Training results
|
| 53 |
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 1.2324 | 0.07 | 50 | 1.2373 |
|
| 57 |
+
| 1.2114 | 0.13 | 100 | 1.2199 |
|
| 58 |
+
| 1.1831 | 0.2 | 150 | 1.2111 |
|
| 59 |
+
| 1.2027 | 0.26 | 200 | 1.2048 |
|
| 60 |
+
| 1.1827 | 0.33 | 250 | 1.2001 |
|
| 61 |
+
| 1.1696 | 0.39 | 300 | 1.1973 |
|
| 62 |
+
| 1.2186 | 0.46 | 350 | 1.1938 |
|
| 63 |
+
| 1.1795 | 0.52 | 400 | 1.1919 |
|
| 64 |
+
| 1.2167 | 0.59 | 450 | 1.1884 |
|
| 65 |
+
| 1.1992 | 0.66 | 500 | 1.1840 |
|
| 66 |
+
| 1.2032 | 0.72 | 550 | 1.1824 |
|
| 67 |
+
| 1.1841 | 0.79 | 600 | 1.1798 |
|
| 68 |
+
| 1.166 | 0.85 | 650 | 1.1789 |
|
| 69 |
+
| 1.1641 | 0.92 | 700 | 1.1761 |
|
| 70 |
+
| 1.1859 | 0.98 | 750 | 1.1752 |
|
| 71 |
+
| 1.132 | 1.05 | 800 | 1.1736 |
|
| 72 |
+
| 1.1461 | 1.12 | 850 | 1.1724 |
|
| 73 |
+
| 1.0965 | 1.18 | 900 | 1.1726 |
|
| 74 |
+
| 1.1064 | 1.25 | 950 | 1.1724 |
|
| 75 |
+
| 1.123 | 1.31 | 1000 | 1.1729 |
|
| 76 |
+
| 1.1079 | 1.38 | 1050 | 1.1695 |
|
| 77 |
+
| 1.12 | 1.44 | 1100 | 1.1707 |
|
| 78 |
+
| 1.1288 | 1.51 | 1150 | 1.1693 |
|
| 79 |
+
| 1.133 | 1.57 | 1200 | 1.1676 |
|
| 80 |
+
| 1.1647 | 1.64 | 1250 | 1.1693 |
|
| 81 |
+
| 1.1269 | 1.71 | 1300 | 1.1658 |
|
| 82 |
+
| 1.1332 | 1.77 | 1350 | 1.1657 |
|
| 83 |
+
| 1.1276 | 1.84 | 1400 | 1.1681 |
|
| 84 |
+
| 1.1361 | 1.9 | 1450 | 1.1633 |
|
| 85 |
+
| 1.1205 | 1.97 | 1500 | 1.1640 |
|
| 86 |
|
| 87 |
|
| 88 |
### Framework versions
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 335605144
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a87aa1c22b9194aebc5bf0d5bfa563c8300e1a6549621dbf298d2728fc3e3a70
|
| 3 |
size 335605144
|
all_results.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_accuracy": 0.8141509433962264,
|
| 4 |
"eval_f1_macro": 0.7097996478763092,
|
| 5 |
"eval_f1_micro": 0.8141509433962264,
|
|
@@ -11,8 +11,8 @@
|
|
| 11 |
"eval_runtime": 66.7734,
|
| 12 |
"eval_samples_per_second": 15.875,
|
| 13 |
"eval_steps_per_second": 3.969,
|
| 14 |
-
"train_loss":
|
| 15 |
-
"train_runtime":
|
| 16 |
-
"train_samples_per_second":
|
| 17 |
-
"train_steps_per_second": 0.
|
| 18 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
"eval_accuracy": 0.8141509433962264,
|
| 4 |
"eval_f1_macro": 0.7097996478763092,
|
| 5 |
"eval_f1_micro": 0.8141509433962264,
|
|
|
|
| 11 |
"eval_runtime": 66.7734,
|
| 12 |
"eval_samples_per_second": 15.875,
|
| 13 |
"eval_steps_per_second": 3.969,
|
| 14 |
+
"train_loss": 1.1488912840840697,
|
| 15 |
+
"train_runtime": 10306.9039,
|
| 16 |
+
"train_samples_per_second": 9.463,
|
| 17 |
+
"train_steps_per_second": 0.148
|
| 18 |
}
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 1.1488912840840697,
|
| 4 |
+
"train_runtime": 10306.9039,
|
| 5 |
+
"train_samples_per_second": 9.463,
|
| 6 |
+
"train_steps_per_second": 0.148
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 50,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -11,1324 +11,1171 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
"learning_rate": 3e-05,
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
"learning_rate": 3e-05,
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
"learning_rate": 3e-05,
|
| 26 |
-
"loss":
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
"learning_rate": 3e-05,
|
| 32 |
-
"loss":
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
"learning_rate": 3e-05,
|
| 38 |
-
"loss":
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
"
|
| 46 |
-
"
|
| 47 |
-
"eval_precision_macro": 0.09801465635567799,
|
| 48 |
-
"eval_precision_micro": 0.2330188679245283,
|
| 49 |
-
"eval_recall_macro": 0.0939265996231733,
|
| 50 |
-
"eval_recall_micro": 0.2330188679245283,
|
| 51 |
-
"eval_runtime": 67.1714,
|
| 52 |
-
"eval_samples_per_second": 15.781,
|
| 53 |
-
"eval_steps_per_second": 3.945,
|
| 54 |
"step": 50
|
| 55 |
},
|
| 56 |
{
|
| 57 |
-
"epoch": 0.
|
| 58 |
"learning_rate": 3e-05,
|
| 59 |
-
"loss":
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 0.
|
| 64 |
"learning_rate": 3e-05,
|
| 65 |
-
"loss":
|
| 66 |
"step": 70
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 0.
|
| 70 |
"learning_rate": 3e-05,
|
| 71 |
-
"loss":
|
| 72 |
"step": 80
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
"learning_rate": 3e-05,
|
| 77 |
-
"loss":
|
| 78 |
"step": 90
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch": 0.
|
| 82 |
"learning_rate": 3e-05,
|
| 83 |
-
"loss":
|
| 84 |
"step": 100
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 0.
|
| 88 |
-
"
|
| 89 |
-
"
|
| 90 |
-
"
|
| 91 |
-
"
|
| 92 |
-
"eval_precision_macro": 0.19505399860785297,
|
| 93 |
-
"eval_precision_micro": 0.4075471698113208,
|
| 94 |
-
"eval_recall_macro": 0.18462138174503467,
|
| 95 |
-
"eval_recall_micro": 0.4075471698113208,
|
| 96 |
-
"eval_runtime": 67.1009,
|
| 97 |
-
"eval_samples_per_second": 15.797,
|
| 98 |
-
"eval_steps_per_second": 3.949,
|
| 99 |
"step": 100
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
"learning_rate": 3e-05,
|
| 104 |
-
"loss":
|
| 105 |
"step": 110
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
"learning_rate": 3e-05,
|
| 110 |
-
"loss":
|
| 111 |
"step": 120
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch": 0.
|
| 115 |
"learning_rate": 3e-05,
|
| 116 |
-
"loss":
|
| 117 |
"step": 130
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 0.
|
| 121 |
"learning_rate": 3e-05,
|
| 122 |
-
"loss":
|
| 123 |
"step": 140
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
"learning_rate": 3e-05,
|
| 128 |
-
"loss":
|
| 129 |
"step": 150
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"
|
| 134 |
-
"
|
| 135 |
-
"
|
| 136 |
-
"
|
| 137 |
-
"eval_precision_macro": 0.3493311966119182,
|
| 138 |
-
"eval_precision_micro": 0.5386792452830189,
|
| 139 |
-
"eval_recall_macro": 0.2871523900319283,
|
| 140 |
-
"eval_recall_micro": 0.5386792452830189,
|
| 141 |
-
"eval_runtime": 67.0903,
|
| 142 |
-
"eval_samples_per_second": 15.8,
|
| 143 |
-
"eval_steps_per_second": 3.95,
|
| 144 |
"step": 150
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 0.
|
| 148 |
"learning_rate": 3e-05,
|
| 149 |
-
"loss":
|
| 150 |
"step": 160
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 0.
|
| 154 |
"learning_rate": 3e-05,
|
| 155 |
-
"loss":
|
| 156 |
"step": 170
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
"learning_rate": 3e-05,
|
| 161 |
-
"loss":
|
| 162 |
"step": 180
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch": 0.
|
| 166 |
"learning_rate": 3e-05,
|
| 167 |
-
"loss":
|
| 168 |
"step": 190
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch": 0.
|
| 172 |
"learning_rate": 3e-05,
|
| 173 |
-
"loss":
|
| 174 |
"step": 200
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 0.
|
| 178 |
-
"
|
| 179 |
-
"
|
| 180 |
-
"
|
| 181 |
-
"
|
| 182 |
-
"eval_precision_macro": 0.3837197141355178,
|
| 183 |
-
"eval_precision_micro": 0.620754716981132,
|
| 184 |
-
"eval_recall_macro": 0.39915842112719735,
|
| 185 |
-
"eval_recall_micro": 0.620754716981132,
|
| 186 |
-
"eval_runtime": 67.0671,
|
| 187 |
-
"eval_samples_per_second": 15.805,
|
| 188 |
-
"eval_steps_per_second": 3.951,
|
| 189 |
"step": 200
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 0.
|
| 193 |
"learning_rate": 3e-05,
|
| 194 |
-
"loss":
|
| 195 |
"step": 210
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
"learning_rate": 3e-05,
|
| 200 |
-
"loss":
|
| 201 |
"step": 220
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 0.
|
| 205 |
"learning_rate": 3e-05,
|
| 206 |
-
"loss":
|
| 207 |
"step": 230
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 0.
|
| 211 |
"learning_rate": 3e-05,
|
| 212 |
-
"loss":
|
| 213 |
"step": 240
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
"learning_rate": 3e-05,
|
| 218 |
-
"loss":
|
| 219 |
"step": 250
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"
|
| 224 |
-
"
|
| 225 |
-
"
|
| 226 |
-
"
|
| 227 |
-
"eval_precision_macro": 0.45439839834135715,
|
| 228 |
-
"eval_precision_micro": 0.6103773584905661,
|
| 229 |
-
"eval_recall_macro": 0.4312111435526721,
|
| 230 |
-
"eval_recall_micro": 0.6103773584905661,
|
| 231 |
-
"eval_runtime": 66.9345,
|
| 232 |
-
"eval_samples_per_second": 15.836,
|
| 233 |
-
"eval_steps_per_second": 3.959,
|
| 234 |
"step": 250
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
"learning_rate": 3e-05,
|
| 239 |
-
"loss":
|
| 240 |
"step": 260
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
"learning_rate": 3e-05,
|
| 245 |
-
"loss":
|
| 246 |
"step": 270
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 0.
|
| 250 |
"learning_rate": 3e-05,
|
| 251 |
-
"loss":
|
| 252 |
"step": 280
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch": 0.
|
| 256 |
"learning_rate": 3e-05,
|
| 257 |
-
"loss":
|
| 258 |
"step": 290
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch": 0.
|
| 262 |
"learning_rate": 3e-05,
|
| 263 |
-
"loss":
|
| 264 |
"step": 300
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch": 0.
|
| 268 |
-
"
|
| 269 |
-
"
|
| 270 |
-
"
|
| 271 |
-
"
|
| 272 |
-
"eval_precision_macro": 0.5649217974276287,
|
| 273 |
-
"eval_precision_micro": 0.6877358490566038,
|
| 274 |
-
"eval_recall_macro": 0.513496327466451,
|
| 275 |
-
"eval_recall_micro": 0.6877358490566038,
|
| 276 |
-
"eval_runtime": 67.1619,
|
| 277 |
-
"eval_samples_per_second": 15.783,
|
| 278 |
-
"eval_steps_per_second": 3.946,
|
| 279 |
"step": 300
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch": 0.
|
| 283 |
"learning_rate": 3e-05,
|
| 284 |
-
"loss":
|
| 285 |
"step": 310
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch": 0.
|
| 289 |
"learning_rate": 3e-05,
|
| 290 |
-
"loss":
|
| 291 |
"step": 320
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 0.
|
| 295 |
"learning_rate": 3e-05,
|
| 296 |
-
"loss":
|
| 297 |
"step": 330
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
"learning_rate": 3e-05,
|
| 302 |
-
"loss":
|
| 303 |
"step": 340
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
"learning_rate": 3e-05,
|
| 308 |
-
"loss":
|
| 309 |
"step": 350
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 0.
|
| 313 |
-
"
|
| 314 |
-
"
|
| 315 |
-
"
|
| 316 |
-
"
|
| 317 |
-
"eval_precision_macro": 0.5194445629359009,
|
| 318 |
-
"eval_precision_micro": 0.6764150943396227,
|
| 319 |
-
"eval_recall_macro": 0.5287937722322651,
|
| 320 |
-
"eval_recall_micro": 0.6764150943396227,
|
| 321 |
-
"eval_runtime": 67.2353,
|
| 322 |
-
"eval_samples_per_second": 15.766,
|
| 323 |
-
"eval_steps_per_second": 3.941,
|
| 324 |
"step": 350
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
"learning_rate": 3e-05,
|
| 329 |
-
"loss":
|
| 330 |
"step": 360
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 0.
|
| 334 |
"learning_rate": 3e-05,
|
| 335 |
-
"loss":
|
| 336 |
"step": 370
|
| 337 |
},
|
| 338 |
{
|
| 339 |
-
"epoch": 0.
|
| 340 |
"learning_rate": 3e-05,
|
| 341 |
-
"loss":
|
| 342 |
"step": 380
|
| 343 |
},
|
| 344 |
{
|
| 345 |
-
"epoch": 0.
|
| 346 |
"learning_rate": 3e-05,
|
| 347 |
-
"loss":
|
| 348 |
"step": 390
|
| 349 |
},
|
| 350 |
{
|
| 351 |
-
"epoch": 0.
|
| 352 |
"learning_rate": 3e-05,
|
| 353 |
-
"loss":
|
| 354 |
"step": 400
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 0.
|
| 358 |
-
"
|
| 359 |
-
"
|
| 360 |
-
"
|
| 361 |
-
"
|
| 362 |
-
"eval_precision_macro": 0.6193164476598846,
|
| 363 |
-
"eval_precision_micro": 0.7339622641509433,
|
| 364 |
-
"eval_recall_macro": 0.5252411264940735,
|
| 365 |
-
"eval_recall_micro": 0.7339622641509433,
|
| 366 |
-
"eval_runtime": 67.4334,
|
| 367 |
-
"eval_samples_per_second": 15.719,
|
| 368 |
-
"eval_steps_per_second": 3.93,
|
| 369 |
"step": 400
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch": 0.
|
| 373 |
"learning_rate": 3e-05,
|
| 374 |
-
"loss":
|
| 375 |
"step": 410
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 0.
|
| 379 |
"learning_rate": 3e-05,
|
| 380 |
-
"loss":
|
| 381 |
"step": 420
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch": 0.
|
| 385 |
"learning_rate": 3e-05,
|
| 386 |
-
"loss":
|
| 387 |
"step": 430
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
"learning_rate": 3e-05,
|
| 392 |
-
"loss":
|
| 393 |
"step": 440
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch": 0.
|
| 397 |
"learning_rate": 3e-05,
|
| 398 |
-
"loss":
|
| 399 |
"step": 450
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch": 0.
|
| 403 |
-
"
|
| 404 |
-
"
|
| 405 |
-
"
|
| 406 |
-
"
|
| 407 |
-
"eval_precision_macro": 0.6096698581228938,
|
| 408 |
-
"eval_precision_micro": 0.7396226415094339,
|
| 409 |
-
"eval_recall_macro": 0.5767264087198709,
|
| 410 |
-
"eval_recall_micro": 0.7396226415094339,
|
| 411 |
-
"eval_runtime": 66.8962,
|
| 412 |
-
"eval_samples_per_second": 15.845,
|
| 413 |
-
"eval_steps_per_second": 3.961,
|
| 414 |
"step": 450
|
| 415 |
},
|
| 416 |
{
|
| 417 |
-
"epoch": 0.
|
| 418 |
"learning_rate": 3e-05,
|
| 419 |
-
"loss":
|
| 420 |
"step": 460
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"epoch": 0.
|
| 424 |
"learning_rate": 3e-05,
|
| 425 |
-
"loss":
|
| 426 |
"step": 470
|
| 427 |
},
|
| 428 |
{
|
| 429 |
-
"epoch": 0.
|
| 430 |
"learning_rate": 3e-05,
|
| 431 |
-
"loss":
|
| 432 |
"step": 480
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch": 0.
|
| 436 |
"learning_rate": 3e-05,
|
| 437 |
-
"loss":
|
| 438 |
"step": 490
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch": 0.
|
| 442 |
"learning_rate": 3e-05,
|
| 443 |
-
"loss":
|
| 444 |
"step": 500
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch": 0.
|
| 448 |
-
"
|
| 449 |
-
"
|
| 450 |
-
"
|
| 451 |
-
"
|
| 452 |
-
"eval_precision_macro": 0.638418606498986,
|
| 453 |
-
"eval_precision_micro": 0.7613207547169811,
|
| 454 |
-
"eval_recall_macro": 0.6145296570629574,
|
| 455 |
-
"eval_recall_micro": 0.7613207547169811,
|
| 456 |
-
"eval_runtime": 66.9411,
|
| 457 |
-
"eval_samples_per_second": 15.835,
|
| 458 |
-
"eval_steps_per_second": 3.959,
|
| 459 |
"step": 500
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 0.
|
| 463 |
"learning_rate": 3e-05,
|
| 464 |
-
"loss":
|
| 465 |
"step": 510
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch": 0.
|
| 469 |
"learning_rate": 3e-05,
|
| 470 |
-
"loss":
|
| 471 |
"step": 520
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
"learning_rate": 3e-05,
|
| 476 |
-
"loss":
|
| 477 |
"step": 530
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch": 0.
|
| 481 |
"learning_rate": 3e-05,
|
| 482 |
-
"loss":
|
| 483 |
"step": 540
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch": 0.
|
| 487 |
"learning_rate": 3e-05,
|
| 488 |
-
"loss":
|
| 489 |
"step": 550
|
| 490 |
},
|
| 491 |
{
|
| 492 |
-
"epoch": 0.
|
| 493 |
-
"
|
| 494 |
-
"
|
| 495 |
-
"
|
| 496 |
-
"
|
| 497 |
-
"eval_precision_macro": 0.624502704252128,
|
| 498 |
-
"eval_precision_micro": 0.7509433962264151,
|
| 499 |
-
"eval_recall_macro": 0.6167049341328479,
|
| 500 |
-
"eval_recall_micro": 0.7509433962264151,
|
| 501 |
-
"eval_runtime": 67.3498,
|
| 502 |
-
"eval_samples_per_second": 15.739,
|
| 503 |
-
"eval_steps_per_second": 3.935,
|
| 504 |
"step": 550
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch": 0.
|
| 508 |
"learning_rate": 3e-05,
|
| 509 |
-
"loss":
|
| 510 |
"step": 560
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch": 0.
|
| 514 |
"learning_rate": 3e-05,
|
| 515 |
-
"loss":
|
| 516 |
"step": 570
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 0.
|
| 520 |
"learning_rate": 3e-05,
|
| 521 |
-
"loss":
|
| 522 |
"step": 580
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch": 0.
|
| 526 |
"learning_rate": 3e-05,
|
| 527 |
-
"loss":
|
| 528 |
"step": 590
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch": 0.
|
| 532 |
"learning_rate": 3e-05,
|
| 533 |
-
"loss":
|
| 534 |
"step": 600
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"
|
| 539 |
-
"
|
| 540 |
-
"
|
| 541 |
-
"
|
| 542 |
-
"eval_precision_macro": 0.6499724898555727,
|
| 543 |
-
"eval_precision_micro": 0.7641509433962265,
|
| 544 |
-
"eval_recall_macro": 0.6139175086252339,
|
| 545 |
-
"eval_recall_micro": 0.7641509433962265,
|
| 546 |
-
"eval_runtime": 66.897,
|
| 547 |
-
"eval_samples_per_second": 15.845,
|
| 548 |
-
"eval_steps_per_second": 3.961,
|
| 549 |
"step": 600
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch": 0.
|
| 553 |
"learning_rate": 3e-05,
|
| 554 |
-
"loss":
|
| 555 |
"step": 610
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
"learning_rate": 3e-05,
|
| 560 |
-
"loss":
|
| 561 |
"step": 620
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch": 0.
|
| 565 |
"learning_rate": 3e-05,
|
| 566 |
-
"loss":
|
| 567 |
"step": 630
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch": 0.
|
| 571 |
"learning_rate": 3e-05,
|
| 572 |
-
"loss":
|
| 573 |
"step": 640
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch": 0.
|
| 577 |
"learning_rate": 3e-05,
|
| 578 |
-
"loss":
|
| 579 |
"step": 650
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch": 0.
|
| 583 |
-
"
|
| 584 |
-
"
|
| 585 |
-
"
|
| 586 |
-
"
|
| 587 |
-
"eval_precision_macro": 0.5941670973495327,
|
| 588 |
-
"eval_precision_micro": 0.7481132075471698,
|
| 589 |
-
"eval_recall_macro": 0.5739727328111488,
|
| 590 |
-
"eval_recall_micro": 0.7481132075471698,
|
| 591 |
-
"eval_runtime": 67.2106,
|
| 592 |
-
"eval_samples_per_second": 15.771,
|
| 593 |
-
"eval_steps_per_second": 3.943,
|
| 594 |
"step": 650
|
| 595 |
},
|
| 596 |
{
|
| 597 |
-
"epoch": 0.
|
| 598 |
"learning_rate": 3e-05,
|
| 599 |
-
"loss":
|
| 600 |
"step": 660
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"epoch": 0.
|
| 604 |
"learning_rate": 3e-05,
|
| 605 |
-
"loss":
|
| 606 |
"step": 670
|
| 607 |
},
|
| 608 |
{
|
| 609 |
-
"epoch": 0.
|
| 610 |
"learning_rate": 3e-05,
|
| 611 |
-
"loss":
|
| 612 |
"step": 680
|
| 613 |
},
|
| 614 |
{
|
| 615 |
-
"epoch": 0.
|
| 616 |
"learning_rate": 3e-05,
|
| 617 |
-
"loss":
|
| 618 |
"step": 690
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 0.
|
| 622 |
"learning_rate": 3e-05,
|
| 623 |
-
"loss":
|
| 624 |
"step": 700
|
| 625 |
},
|
| 626 |
{
|
| 627 |
-
"epoch": 0.
|
| 628 |
-
"
|
| 629 |
-
"
|
| 630 |
-
"
|
| 631 |
-
"
|
| 632 |
-
"eval_precision_macro": 0.6678204026981202,
|
| 633 |
-
"eval_precision_micro": 0.7971698113207547,
|
| 634 |
-
"eval_recall_macro": 0.6480125227888868,
|
| 635 |
-
"eval_recall_micro": 0.7971698113207547,
|
| 636 |
-
"eval_runtime": 67.3982,
|
| 637 |
-
"eval_samples_per_second": 15.727,
|
| 638 |
-
"eval_steps_per_second": 3.932,
|
| 639 |
"step": 700
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
"learning_rate": 3e-05,
|
| 644 |
-
"loss":
|
| 645 |
"step": 710
|
| 646 |
},
|
| 647 |
{
|
| 648 |
-
"epoch": 0.
|
| 649 |
"learning_rate": 3e-05,
|
| 650 |
-
"loss":
|
| 651 |
"step": 720
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"epoch": 0.
|
| 655 |
"learning_rate": 3e-05,
|
| 656 |
-
"loss":
|
| 657 |
"step": 730
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch": 0.
|
| 661 |
"learning_rate": 3e-05,
|
| 662 |
-
"loss":
|
| 663 |
"step": 740
|
| 664 |
},
|
| 665 |
{
|
| 666 |
-
"epoch": 0.
|
| 667 |
"learning_rate": 3e-05,
|
| 668 |
-
"loss":
|
| 669 |
"step": 750
|
| 670 |
},
|
| 671 |
{
|
| 672 |
-
"epoch": 0.
|
| 673 |
-
"
|
| 674 |
-
"
|
| 675 |
-
"
|
| 676 |
-
"
|
| 677 |
-
"eval_precision_macro": 0.6677975283624125,
|
| 678 |
-
"eval_precision_micro": 0.7773584905660378,
|
| 679 |
-
"eval_recall_macro": 0.6360471775658058,
|
| 680 |
-
"eval_recall_micro": 0.7773584905660378,
|
| 681 |
-
"eval_runtime": 67.7832,
|
| 682 |
-
"eval_samples_per_second": 15.638,
|
| 683 |
-
"eval_steps_per_second": 3.91,
|
| 684 |
"step": 750
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch": 0
|
| 688 |
"learning_rate": 3e-05,
|
| 689 |
-
"loss":
|
| 690 |
"step": 760
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"epoch":
|
| 694 |
"learning_rate": 3e-05,
|
| 695 |
-
"loss":
|
| 696 |
"step": 770
|
| 697 |
},
|
| 698 |
{
|
| 699 |
-
"epoch":
|
| 700 |
"learning_rate": 3e-05,
|
| 701 |
-
"loss":
|
| 702 |
"step": 780
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch":
|
| 706 |
"learning_rate": 3e-05,
|
| 707 |
-
"loss":
|
| 708 |
"step": 790
|
| 709 |
},
|
| 710 |
{
|
| 711 |
-
"epoch":
|
| 712 |
"learning_rate": 3e-05,
|
| 713 |
-
"loss":
|
| 714 |
"step": 800
|
| 715 |
},
|
| 716 |
{
|
| 717 |
-
"epoch":
|
| 718 |
-
"
|
| 719 |
-
"
|
| 720 |
-
"
|
| 721 |
-
"
|
| 722 |
-
"eval_precision_macro": 0.6435014283226803,
|
| 723 |
-
"eval_precision_micro": 0.7830188679245284,
|
| 724 |
-
"eval_recall_macro": 0.6816157451405587,
|
| 725 |
-
"eval_recall_micro": 0.7830188679245284,
|
| 726 |
-
"eval_runtime": 67.2351,
|
| 727 |
-
"eval_samples_per_second": 15.766,
|
| 728 |
-
"eval_steps_per_second": 3.941,
|
| 729 |
"step": 800
|
| 730 |
},
|
| 731 |
{
|
| 732 |
-
"epoch":
|
| 733 |
"learning_rate": 3e-05,
|
| 734 |
-
"loss":
|
| 735 |
"step": 810
|
| 736 |
},
|
| 737 |
{
|
| 738 |
-
"epoch":
|
| 739 |
"learning_rate": 3e-05,
|
| 740 |
-
"loss":
|
| 741 |
"step": 820
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
"learning_rate": 3e-05,
|
| 746 |
-
"loss":
|
| 747 |
"step": 830
|
| 748 |
},
|
| 749 |
{
|
| 750 |
-
"epoch":
|
| 751 |
"learning_rate": 3e-05,
|
| 752 |
-
"loss":
|
| 753 |
"step": 840
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"epoch":
|
| 757 |
"learning_rate": 3e-05,
|
| 758 |
-
"loss":
|
| 759 |
"step": 850
|
| 760 |
},
|
| 761 |
{
|
| 762 |
-
"epoch":
|
| 763 |
-
"
|
| 764 |
-
"
|
| 765 |
-
"
|
| 766 |
-
"
|
| 767 |
-
"eval_precision_macro": 0.7040157648486967,
|
| 768 |
-
"eval_precision_micro": 0.7971698113207547,
|
| 769 |
-
"eval_recall_macro": 0.6567342355863813,
|
| 770 |
-
"eval_recall_micro": 0.7971698113207547,
|
| 771 |
-
"eval_runtime": 67.3249,
|
| 772 |
-
"eval_samples_per_second": 15.745,
|
| 773 |
-
"eval_steps_per_second": 3.936,
|
| 774 |
"step": 850
|
| 775 |
},
|
| 776 |
{
|
| 777 |
-
"epoch":
|
| 778 |
"learning_rate": 3e-05,
|
| 779 |
-
"loss":
|
| 780 |
"step": 860
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch":
|
| 784 |
"learning_rate": 3e-05,
|
| 785 |
-
"loss":
|
| 786 |
"step": 870
|
| 787 |
},
|
| 788 |
{
|
| 789 |
-
"epoch":
|
| 790 |
"learning_rate": 3e-05,
|
| 791 |
-
"loss":
|
| 792 |
"step": 880
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"epoch":
|
| 796 |
"learning_rate": 3e-05,
|
| 797 |
-
"loss":
|
| 798 |
"step": 890
|
| 799 |
},
|
| 800 |
{
|
| 801 |
-
"epoch":
|
| 802 |
"learning_rate": 3e-05,
|
| 803 |
-
"loss":
|
| 804 |
"step": 900
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch":
|
| 808 |
-
"
|
| 809 |
-
"
|
| 810 |
-
"
|
| 811 |
-
"
|
| 812 |
-
"eval_precision_macro": 0.6519385252086033,
|
| 813 |
-
"eval_precision_micro": 0.7632075471698113,
|
| 814 |
-
"eval_recall_macro": 0.6290591814696965,
|
| 815 |
-
"eval_recall_micro": 0.7632075471698113,
|
| 816 |
-
"eval_runtime": 67.0228,
|
| 817 |
-
"eval_samples_per_second": 15.816,
|
| 818 |
-
"eval_steps_per_second": 3.954,
|
| 819 |
"step": 900
|
| 820 |
},
|
| 821 |
{
|
| 822 |
-
"epoch":
|
| 823 |
"learning_rate": 3e-05,
|
| 824 |
-
"loss":
|
| 825 |
"step": 910
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch":
|
| 829 |
"learning_rate": 3e-05,
|
| 830 |
-
"loss":
|
| 831 |
"step": 920
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"epoch":
|
| 835 |
"learning_rate": 3e-05,
|
| 836 |
-
"loss":
|
| 837 |
"step": 930
|
| 838 |
},
|
| 839 |
{
|
| 840 |
-
"epoch":
|
| 841 |
"learning_rate": 3e-05,
|
| 842 |
-
"loss":
|
| 843 |
"step": 940
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch":
|
| 847 |
"learning_rate": 3e-05,
|
| 848 |
-
"loss":
|
| 849 |
"step": 950
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch":
|
| 853 |
-
"
|
| 854 |
-
"
|
| 855 |
-
"
|
| 856 |
-
"
|
| 857 |
-
"eval_precision_macro": 0.67574812222591,
|
| 858 |
-
"eval_precision_micro": 0.780188679245283,
|
| 859 |
-
"eval_recall_macro": 0.6697872775950671,
|
| 860 |
-
"eval_recall_micro": 0.780188679245283,
|
| 861 |
-
"eval_runtime": 67.3483,
|
| 862 |
-
"eval_samples_per_second": 15.739,
|
| 863 |
-
"eval_steps_per_second": 3.935,
|
| 864 |
"step": 950
|
| 865 |
},
|
| 866 |
{
|
| 867 |
-
"epoch":
|
| 868 |
"learning_rate": 3e-05,
|
| 869 |
-
"loss":
|
| 870 |
"step": 960
|
| 871 |
},
|
| 872 |
{
|
| 873 |
-
"epoch":
|
| 874 |
"learning_rate": 3e-05,
|
| 875 |
-
"loss":
|
| 876 |
"step": 970
|
| 877 |
},
|
| 878 |
{
|
| 879 |
-
"epoch":
|
| 880 |
"learning_rate": 3e-05,
|
| 881 |
-
"loss":
|
| 882 |
"step": 980
|
| 883 |
},
|
| 884 |
{
|
| 885 |
-
"epoch":
|
| 886 |
"learning_rate": 3e-05,
|
| 887 |
-
"loss":
|
| 888 |
"step": 990
|
| 889 |
},
|
| 890 |
{
|
| 891 |
-
"epoch":
|
| 892 |
"learning_rate": 3e-05,
|
| 893 |
-
"loss":
|
| 894 |
"step": 1000
|
| 895 |
},
|
| 896 |
{
|
| 897 |
-
"epoch":
|
| 898 |
-
"
|
| 899 |
-
"
|
| 900 |
-
"
|
| 901 |
-
"
|
| 902 |
-
"eval_precision_macro": 0.6181852032037266,
|
| 903 |
-
"eval_precision_micro": 0.7377358490566037,
|
| 904 |
-
"eval_recall_macro": 0.6300074429793591,
|
| 905 |
-
"eval_recall_micro": 0.7377358490566037,
|
| 906 |
-
"eval_runtime": 66.8035,
|
| 907 |
-
"eval_samples_per_second": 15.867,
|
| 908 |
-
"eval_steps_per_second": 3.967,
|
| 909 |
"step": 1000
|
| 910 |
},
|
| 911 |
{
|
| 912 |
-
"epoch":
|
| 913 |
"learning_rate": 3e-05,
|
| 914 |
-
"loss":
|
| 915 |
"step": 1010
|
| 916 |
},
|
| 917 |
{
|
| 918 |
-
"epoch":
|
| 919 |
"learning_rate": 3e-05,
|
| 920 |
-
"loss":
|
| 921 |
"step": 1020
|
| 922 |
},
|
| 923 |
{
|
| 924 |
-
"epoch":
|
| 925 |
"learning_rate": 3e-05,
|
| 926 |
-
"loss":
|
| 927 |
"step": 1030
|
| 928 |
},
|
| 929 |
{
|
| 930 |
-
"epoch":
|
| 931 |
"learning_rate": 3e-05,
|
| 932 |
-
"loss":
|
| 933 |
"step": 1040
|
| 934 |
},
|
| 935 |
{
|
| 936 |
-
"epoch":
|
| 937 |
"learning_rate": 3e-05,
|
| 938 |
-
"loss":
|
| 939 |
"step": 1050
|
| 940 |
},
|
| 941 |
{
|
| 942 |
-
"epoch":
|
| 943 |
-
"
|
| 944 |
-
"
|
| 945 |
-
"
|
| 946 |
-
"
|
| 947 |
-
"eval_precision_macro": 0.6856812225733196,
|
| 948 |
-
"eval_precision_micro": 0.7981132075471699,
|
| 949 |
-
"eval_recall_macro": 0.6992476720564776,
|
| 950 |
-
"eval_recall_micro": 0.7981132075471699,
|
| 951 |
-
"eval_runtime": 66.8444,
|
| 952 |
-
"eval_samples_per_second": 15.858,
|
| 953 |
-
"eval_steps_per_second": 3.964,
|
| 954 |
"step": 1050
|
| 955 |
},
|
| 956 |
{
|
| 957 |
-
"epoch":
|
| 958 |
"learning_rate": 3e-05,
|
| 959 |
-
"loss":
|
| 960 |
"step": 1060
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch":
|
| 964 |
"learning_rate": 3e-05,
|
| 965 |
-
"loss":
|
| 966 |
"step": 1070
|
| 967 |
},
|
| 968 |
{
|
| 969 |
-
"epoch":
|
| 970 |
"learning_rate": 3e-05,
|
| 971 |
-
"loss":
|
| 972 |
"step": 1080
|
| 973 |
},
|
| 974 |
{
|
| 975 |
-
"epoch":
|
| 976 |
"learning_rate": 3e-05,
|
| 977 |
-
"loss":
|
| 978 |
"step": 1090
|
| 979 |
},
|
| 980 |
{
|
| 981 |
-
"epoch":
|
| 982 |
"learning_rate": 3e-05,
|
| 983 |
-
"loss":
|
| 984 |
"step": 1100
|
| 985 |
},
|
| 986 |
{
|
| 987 |
-
"epoch":
|
| 988 |
-
"
|
| 989 |
-
"
|
| 990 |
-
"
|
| 991 |
-
"
|
| 992 |
-
"eval_precision_macro": 0.6636349851737382,
|
| 993 |
-
"eval_precision_micro": 0.8,
|
| 994 |
-
"eval_recall_macro": 0.6697553358712118,
|
| 995 |
-
"eval_recall_micro": 0.8,
|
| 996 |
-
"eval_runtime": 66.9434,
|
| 997 |
-
"eval_samples_per_second": 15.834,
|
| 998 |
-
"eval_steps_per_second": 3.959,
|
| 999 |
"step": 1100
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
-
"epoch":
|
| 1003 |
"learning_rate": 3e-05,
|
| 1004 |
-
"loss":
|
| 1005 |
"step": 1110
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
-
"epoch":
|
| 1009 |
"learning_rate": 3e-05,
|
| 1010 |
-
"loss":
|
| 1011 |
"step": 1120
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"epoch":
|
| 1015 |
"learning_rate": 3e-05,
|
| 1016 |
-
"loss":
|
| 1017 |
"step": 1130
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
-
"epoch":
|
| 1021 |
"learning_rate": 3e-05,
|
| 1022 |
-
"loss":
|
| 1023 |
"step": 1140
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
"learning_rate": 3e-05,
|
| 1028 |
-
"loss":
|
| 1029 |
"step": 1150
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
-
"epoch":
|
| 1033 |
-
"
|
| 1034 |
-
"
|
| 1035 |
-
"
|
| 1036 |
-
"
|
| 1037 |
-
"eval_precision_macro": 0.6923924758215005,
|
| 1038 |
-
"eval_precision_micro": 0.7952830188679245,
|
| 1039 |
-
"eval_recall_macro": 0.6441492192889419,
|
| 1040 |
-
"eval_recall_micro": 0.7952830188679245,
|
| 1041 |
-
"eval_runtime": 67.1705,
|
| 1042 |
-
"eval_samples_per_second": 15.781,
|
| 1043 |
-
"eval_steps_per_second": 3.945,
|
| 1044 |
"step": 1150
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
-
"epoch":
|
| 1048 |
"learning_rate": 3e-05,
|
| 1049 |
-
"loss":
|
| 1050 |
"step": 1160
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
-
"epoch":
|
| 1054 |
"learning_rate": 3e-05,
|
| 1055 |
-
"loss":
|
| 1056 |
"step": 1170
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
-
"epoch":
|
| 1060 |
"learning_rate": 3e-05,
|
| 1061 |
-
"loss":
|
| 1062 |
"step": 1180
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
-
"epoch":
|
| 1066 |
"learning_rate": 3e-05,
|
| 1067 |
-
"loss":
|
| 1068 |
"step": 1190
|
| 1069 |
},
|
| 1070 |
{
|
| 1071 |
-
"epoch":
|
| 1072 |
"learning_rate": 3e-05,
|
| 1073 |
-
"loss":
|
| 1074 |
"step": 1200
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
-
"epoch":
|
| 1078 |
-
"
|
| 1079 |
-
"
|
| 1080 |
-
"
|
| 1081 |
-
"
|
| 1082 |
-
"eval_precision_macro": 0.6712905678869693,
|
| 1083 |
-
"eval_precision_micro": 0.7943396226415095,
|
| 1084 |
-
"eval_recall_macro": 0.6537773538776073,
|
| 1085 |
-
"eval_recall_micro": 0.7943396226415095,
|
| 1086 |
-
"eval_runtime": 66.9611,
|
| 1087 |
-
"eval_samples_per_second": 15.83,
|
| 1088 |
-
"eval_steps_per_second": 3.958,
|
| 1089 |
"step": 1200
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
-
"epoch":
|
| 1093 |
"learning_rate": 3e-05,
|
| 1094 |
-
"loss":
|
| 1095 |
"step": 1210
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
-
"epoch":
|
| 1099 |
"learning_rate": 3e-05,
|
| 1100 |
-
"loss":
|
| 1101 |
"step": 1220
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
-
"epoch":
|
| 1105 |
"learning_rate": 3e-05,
|
| 1106 |
-
"loss":
|
| 1107 |
"step": 1230
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
-
"epoch":
|
| 1111 |
"learning_rate": 3e-05,
|
| 1112 |
-
"loss":
|
| 1113 |
"step": 1240
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
-
"epoch":
|
| 1117 |
"learning_rate": 3e-05,
|
| 1118 |
-
"loss":
|
| 1119 |
"step": 1250
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
-
"epoch":
|
| 1123 |
-
"
|
| 1124 |
-
"
|
| 1125 |
-
"
|
| 1126 |
-
"
|
| 1127 |
-
"eval_precision_macro": 0.6735372413807635,
|
| 1128 |
-
"eval_precision_micro": 0.7839622641509434,
|
| 1129 |
-
"eval_recall_macro": 0.6450299050285588,
|
| 1130 |
-
"eval_recall_micro": 0.7839622641509434,
|
| 1131 |
-
"eval_runtime": 66.9053,
|
| 1132 |
-
"eval_samples_per_second": 15.843,
|
| 1133 |
-
"eval_steps_per_second": 3.961,
|
| 1134 |
"step": 1250
|
| 1135 |
},
|
| 1136 |
{
|
| 1137 |
-
"epoch": 1.
|
| 1138 |
"learning_rate": 3e-05,
|
| 1139 |
-
"loss":
|
| 1140 |
"step": 1260
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
-
"epoch": 1.
|
| 1144 |
"learning_rate": 3e-05,
|
| 1145 |
-
"loss":
|
| 1146 |
"step": 1270
|
| 1147 |
},
|
| 1148 |
{
|
| 1149 |
-
"epoch": 1.
|
| 1150 |
"learning_rate": 3e-05,
|
| 1151 |
-
"loss":
|
| 1152 |
"step": 1280
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
-
"epoch": 1.
|
| 1156 |
"learning_rate": 3e-05,
|
| 1157 |
-
"loss":
|
| 1158 |
"step": 1290
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
-
"epoch": 1.
|
| 1162 |
"learning_rate": 3e-05,
|
| 1163 |
-
"loss":
|
| 1164 |
"step": 1300
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
-
"epoch": 1.
|
| 1168 |
-
"
|
| 1169 |
-
"
|
| 1170 |
-
"
|
| 1171 |
-
"
|
| 1172 |
-
"eval_precision_macro": 0.7109359814450084,
|
| 1173 |
-
"eval_precision_micro": 0.8047169811320755,
|
| 1174 |
-
"eval_recall_macro": 0.6387082579227776,
|
| 1175 |
-
"eval_recall_micro": 0.8047169811320755,
|
| 1176 |
-
"eval_runtime": 67.3206,
|
| 1177 |
-
"eval_samples_per_second": 15.746,
|
| 1178 |
-
"eval_steps_per_second": 3.936,
|
| 1179 |
"step": 1300
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
-
"epoch": 1.
|
| 1183 |
"learning_rate": 3e-05,
|
| 1184 |
-
"loss":
|
| 1185 |
"step": 1310
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
-
"epoch": 1.
|
| 1189 |
"learning_rate": 3e-05,
|
| 1190 |
-
"loss":
|
| 1191 |
"step": 1320
|
| 1192 |
},
|
| 1193 |
{
|
| 1194 |
-
"epoch": 1.
|
| 1195 |
"learning_rate": 3e-05,
|
| 1196 |
-
"loss":
|
| 1197 |
"step": 1330
|
| 1198 |
},
|
| 1199 |
{
|
| 1200 |
-
"epoch": 1.
|
| 1201 |
"learning_rate": 3e-05,
|
| 1202 |
-
"loss":
|
| 1203 |
"step": 1340
|
| 1204 |
},
|
| 1205 |
{
|
| 1206 |
-
"epoch": 1.
|
| 1207 |
"learning_rate": 3e-05,
|
| 1208 |
-
"loss":
|
| 1209 |
"step": 1350
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
-
"epoch": 1.
|
| 1213 |
-
"
|
| 1214 |
-
"
|
| 1215 |
-
"
|
| 1216 |
-
"
|
| 1217 |
-
"eval_precision_macro": 0.6898539099210392,
|
| 1218 |
-
"eval_precision_micro": 0.8028301886792453,
|
| 1219 |
-
"eval_recall_macro": 0.6558796396655843,
|
| 1220 |
-
"eval_recall_micro": 0.8028301886792453,
|
| 1221 |
-
"eval_runtime": 67.0656,
|
| 1222 |
-
"eval_samples_per_second": 15.805,
|
| 1223 |
-
"eval_steps_per_second": 3.951,
|
| 1224 |
"step": 1350
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
-
"epoch": 1.
|
| 1228 |
"learning_rate": 3e-05,
|
| 1229 |
-
"loss":
|
| 1230 |
"step": 1360
|
| 1231 |
},
|
| 1232 |
{
|
| 1233 |
-
"epoch": 1.
|
| 1234 |
"learning_rate": 3e-05,
|
| 1235 |
-
"loss":
|
| 1236 |
"step": 1370
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
-
"epoch": 1.
|
| 1240 |
"learning_rate": 3e-05,
|
| 1241 |
-
"loss":
|
| 1242 |
"step": 1380
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"epoch": 1.
|
| 1246 |
"learning_rate": 3e-05,
|
| 1247 |
-
"loss":
|
| 1248 |
"step": 1390
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
-
"epoch": 1.
|
| 1252 |
"learning_rate": 3e-05,
|
| 1253 |
-
"loss":
|
| 1254 |
"step": 1400
|
| 1255 |
},
|
| 1256 |
{
|
| 1257 |
-
"epoch": 1.
|
| 1258 |
-
"
|
| 1259 |
-
"
|
| 1260 |
-
"
|
| 1261 |
-
"
|
| 1262 |
-
"eval_precision_macro": 0.7114033533579757,
|
| 1263 |
-
"eval_precision_micro": 0.7962264150943397,
|
| 1264 |
-
"eval_recall_macro": 0.6942176996685531,
|
| 1265 |
-
"eval_recall_micro": 0.7962264150943397,
|
| 1266 |
-
"eval_runtime": 66.8832,
|
| 1267 |
-
"eval_samples_per_second": 15.849,
|
| 1268 |
-
"eval_steps_per_second": 3.962,
|
| 1269 |
"step": 1400
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
-
"epoch": 1.
|
| 1273 |
"learning_rate": 3e-05,
|
| 1274 |
-
"loss":
|
| 1275 |
"step": 1410
|
| 1276 |
},
|
| 1277 |
{
|
| 1278 |
-
"epoch": 1.
|
| 1279 |
"learning_rate": 3e-05,
|
| 1280 |
-
"loss":
|
| 1281 |
"step": 1420
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
-
"epoch": 1.
|
| 1285 |
"learning_rate": 3e-05,
|
| 1286 |
-
"loss":
|
| 1287 |
"step": 1430
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
-
"epoch": 1.
|
| 1291 |
"learning_rate": 3e-05,
|
| 1292 |
-
"loss":
|
| 1293 |
"step": 1440
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
-
"epoch": 1.
|
| 1297 |
"learning_rate": 3e-05,
|
| 1298 |
-
"loss":
|
| 1299 |
"step": 1450
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
-
"epoch": 1.
|
| 1303 |
-
"
|
| 1304 |
-
"
|
| 1305 |
-
"
|
| 1306 |
-
"
|
| 1307 |
-
"eval_precision_macro": 0.7222302630120379,
|
| 1308 |
-
"eval_precision_micro": 0.8141509433962264,
|
| 1309 |
-
"eval_recall_macro": 0.7125706602249756,
|
| 1310 |
-
"eval_recall_micro": 0.8141509433962264,
|
| 1311 |
-
"eval_runtime": 67.0694,
|
| 1312 |
-
"eval_samples_per_second": 15.805,
|
| 1313 |
-
"eval_steps_per_second": 3.951,
|
| 1314 |
"step": 1450
|
| 1315 |
},
|
| 1316 |
{
|
| 1317 |
-
"epoch": 1.
|
| 1318 |
-
"
|
| 1319 |
-
"
|
| 1320 |
-
"
|
| 1321 |
-
|
| 1322 |
-
|
| 1323 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1324 |
}
|
| 1325 |
],
|
| 1326 |
"logging_steps": 10,
|
| 1327 |
-
"max_steps":
|
| 1328 |
"num_input_tokens_seen": 0,
|
| 1329 |
"num_train_epochs": 2,
|
| 1330 |
-
"save_steps":
|
| 1331 |
-
"total_flos":
|
| 1332 |
"train_batch_size": 4,
|
| 1333 |
"trial_name": null,
|
| 1334 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
"eval_steps": 50,
|
| 6 |
+
"global_step": 1524,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
"learning_rate": 3e-05,
|
| 14 |
+
"loss": 1.5105,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
"learning_rate": 3e-05,
|
| 20 |
+
"loss": 1.3015,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.04,
|
| 25 |
"learning_rate": 3e-05,
|
| 26 |
+
"loss": 1.2713,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.05,
|
| 31 |
"learning_rate": 3e-05,
|
| 32 |
+
"loss": 1.2762,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.07,
|
| 37 |
"learning_rate": 3e-05,
|
| 38 |
+
"loss": 1.2324,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.07,
|
| 43 |
+
"eval_loss": 1.237317442893982,
|
| 44 |
+
"eval_runtime": 19.2113,
|
| 45 |
+
"eval_samples_per_second": 25.662,
|
| 46 |
+
"eval_steps_per_second": 1.614,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
"step": 50
|
| 48 |
},
|
| 49 |
{
|
| 50 |
+
"epoch": 0.08,
|
| 51 |
"learning_rate": 3e-05,
|
| 52 |
+
"loss": 1.2385,
|
| 53 |
"step": 60
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 0.09,
|
| 57 |
"learning_rate": 3e-05,
|
| 58 |
+
"loss": 1.2294,
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 0.1,
|
| 63 |
"learning_rate": 3e-05,
|
| 64 |
+
"loss": 1.2065,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.12,
|
| 69 |
"learning_rate": 3e-05,
|
| 70 |
+
"loss": 1.2171,
|
| 71 |
"step": 90
|
| 72 |
},
|
| 73 |
{
|
| 74 |
+
"epoch": 0.13,
|
| 75 |
"learning_rate": 3e-05,
|
| 76 |
+
"loss": 1.2114,
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 0.13,
|
| 81 |
+
"eval_loss": 1.2198712825775146,
|
| 82 |
+
"eval_runtime": 19.1971,
|
| 83 |
+
"eval_samples_per_second": 25.681,
|
| 84 |
+
"eval_steps_per_second": 1.615,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
"step": 100
|
| 86 |
},
|
| 87 |
{
|
| 88 |
+
"epoch": 0.14,
|
| 89 |
"learning_rate": 3e-05,
|
| 90 |
+
"loss": 1.1593,
|
| 91 |
"step": 110
|
| 92 |
},
|
| 93 |
{
|
| 94 |
+
"epoch": 0.16,
|
| 95 |
"learning_rate": 3e-05,
|
| 96 |
+
"loss": 1.1865,
|
| 97 |
"step": 120
|
| 98 |
},
|
| 99 |
{
|
| 100 |
+
"epoch": 0.17,
|
| 101 |
"learning_rate": 3e-05,
|
| 102 |
+
"loss": 1.1765,
|
| 103 |
"step": 130
|
| 104 |
},
|
| 105 |
{
|
| 106 |
+
"epoch": 0.18,
|
| 107 |
"learning_rate": 3e-05,
|
| 108 |
+
"loss": 1.2079,
|
| 109 |
"step": 140
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 0.2,
|
| 113 |
"learning_rate": 3e-05,
|
| 114 |
+
"loss": 1.1831,
|
| 115 |
"step": 150
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 0.2,
|
| 119 |
+
"eval_loss": 1.2110768556594849,
|
| 120 |
+
"eval_runtime": 19.4082,
|
| 121 |
+
"eval_samples_per_second": 25.402,
|
| 122 |
+
"eval_steps_per_second": 1.597,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.21,
|
| 127 |
"learning_rate": 3e-05,
|
| 128 |
+
"loss": 1.2075,
|
| 129 |
"step": 160
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.22,
|
| 133 |
"learning_rate": 3e-05,
|
| 134 |
+
"loss": 1.2124,
|
| 135 |
"step": 170
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.24,
|
| 139 |
"learning_rate": 3e-05,
|
| 140 |
+
"loss": 1.2023,
|
| 141 |
"step": 180
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 0.25,
|
| 145 |
"learning_rate": 3e-05,
|
| 146 |
+
"loss": 1.1721,
|
| 147 |
"step": 190
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.26,
|
| 151 |
"learning_rate": 3e-05,
|
| 152 |
+
"loss": 1.2027,
|
| 153 |
"step": 200
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 0.26,
|
| 157 |
+
"eval_loss": 1.2048192024230957,
|
| 158 |
+
"eval_runtime": 19.3037,
|
| 159 |
+
"eval_samples_per_second": 25.539,
|
| 160 |
+
"eval_steps_per_second": 1.606,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
"step": 200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 0.28,
|
| 165 |
"learning_rate": 3e-05,
|
| 166 |
+
"loss": 1.1674,
|
| 167 |
"step": 210
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.29,
|
| 171 |
"learning_rate": 3e-05,
|
| 172 |
+
"loss": 1.1882,
|
| 173 |
"step": 220
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 0.3,
|
| 177 |
"learning_rate": 3e-05,
|
| 178 |
+
"loss": 1.2099,
|
| 179 |
"step": 230
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 0.31,
|
| 183 |
"learning_rate": 3e-05,
|
| 184 |
+
"loss": 1.1988,
|
| 185 |
"step": 240
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 0.33,
|
| 189 |
"learning_rate": 3e-05,
|
| 190 |
+
"loss": 1.1827,
|
| 191 |
"step": 250
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.33,
|
| 195 |
+
"eval_loss": 1.2000844478607178,
|
| 196 |
+
"eval_runtime": 19.2851,
|
| 197 |
+
"eval_samples_per_second": 25.564,
|
| 198 |
+
"eval_steps_per_second": 1.607,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 199 |
"step": 250
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 0.34,
|
| 203 |
"learning_rate": 3e-05,
|
| 204 |
+
"loss": 1.1452,
|
| 205 |
"step": 260
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.35,
|
| 209 |
"learning_rate": 3e-05,
|
| 210 |
+
"loss": 1.185,
|
| 211 |
"step": 270
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 0.37,
|
| 215 |
"learning_rate": 3e-05,
|
| 216 |
+
"loss": 1.1979,
|
| 217 |
"step": 280
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"epoch": 0.38,
|
| 221 |
"learning_rate": 3e-05,
|
| 222 |
+
"loss": 1.2155,
|
| 223 |
"step": 290
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 0.39,
|
| 227 |
"learning_rate": 3e-05,
|
| 228 |
+
"loss": 1.1696,
|
| 229 |
"step": 300
|
| 230 |
},
|
| 231 |
{
|
| 232 |
+
"epoch": 0.39,
|
| 233 |
+
"eval_loss": 1.1973345279693604,
|
| 234 |
+
"eval_runtime": 18.7078,
|
| 235 |
+
"eval_samples_per_second": 26.353,
|
| 236 |
+
"eval_steps_per_second": 1.657,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 0.41,
|
| 241 |
"learning_rate": 3e-05,
|
| 242 |
+
"loss": 1.1426,
|
| 243 |
"step": 310
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 0.42,
|
| 247 |
"learning_rate": 3e-05,
|
| 248 |
+
"loss": 1.1691,
|
| 249 |
"step": 320
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 0.43,
|
| 253 |
"learning_rate": 3e-05,
|
| 254 |
+
"loss": 1.1991,
|
| 255 |
"step": 330
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 0.45,
|
| 259 |
"learning_rate": 3e-05,
|
| 260 |
+
"loss": 1.1992,
|
| 261 |
"step": 340
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.46,
|
| 265 |
"learning_rate": 3e-05,
|
| 266 |
+
"loss": 1.2186,
|
| 267 |
"step": 350
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 0.46,
|
| 271 |
+
"eval_loss": 1.193804383277893,
|
| 272 |
+
"eval_runtime": 18.8972,
|
| 273 |
+
"eval_samples_per_second": 26.088,
|
| 274 |
+
"eval_steps_per_second": 1.64,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 275 |
"step": 350
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.47,
|
| 279 |
"learning_rate": 3e-05,
|
| 280 |
+
"loss": 1.1691,
|
| 281 |
"step": 360
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.49,
|
| 285 |
"learning_rate": 3e-05,
|
| 286 |
+
"loss": 1.1595,
|
| 287 |
"step": 370
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 0.5,
|
| 291 |
"learning_rate": 3e-05,
|
| 292 |
+
"loss": 1.1494,
|
| 293 |
"step": 380
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 0.51,
|
| 297 |
"learning_rate": 3e-05,
|
| 298 |
+
"loss": 1.1985,
|
| 299 |
"step": 390
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 0.52,
|
| 303 |
"learning_rate": 3e-05,
|
| 304 |
+
"loss": 1.1795,
|
| 305 |
"step": 400
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.52,
|
| 309 |
+
"eval_loss": 1.1919257640838623,
|
| 310 |
+
"eval_runtime": 19.3777,
|
| 311 |
+
"eval_samples_per_second": 25.442,
|
| 312 |
+
"eval_steps_per_second": 1.6,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 313 |
"step": 400
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"epoch": 0.54,
|
| 317 |
"learning_rate": 3e-05,
|
| 318 |
+
"loss": 1.1254,
|
| 319 |
"step": 410
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 0.55,
|
| 323 |
"learning_rate": 3e-05,
|
| 324 |
+
"loss": 1.1772,
|
| 325 |
"step": 420
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 0.56,
|
| 329 |
"learning_rate": 3e-05,
|
| 330 |
+
"loss": 1.1956,
|
| 331 |
"step": 430
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.58,
|
| 335 |
"learning_rate": 3e-05,
|
| 336 |
+
"loss": 1.1959,
|
| 337 |
"step": 440
|
| 338 |
},
|
| 339 |
{
|
| 340 |
+
"epoch": 0.59,
|
| 341 |
"learning_rate": 3e-05,
|
| 342 |
+
"loss": 1.2167,
|
| 343 |
"step": 450
|
| 344 |
},
|
| 345 |
{
|
| 346 |
+
"epoch": 0.59,
|
| 347 |
+
"eval_loss": 1.188421607017517,
|
| 348 |
+
"eval_runtime": 18.7028,
|
| 349 |
+
"eval_samples_per_second": 26.36,
|
| 350 |
+
"eval_steps_per_second": 1.658,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 351 |
"step": 450
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 0.6,
|
| 355 |
"learning_rate": 3e-05,
|
| 356 |
+
"loss": 1.1625,
|
| 357 |
"step": 460
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 0.62,
|
| 361 |
"learning_rate": 3e-05,
|
| 362 |
+
"loss": 1.1979,
|
| 363 |
"step": 470
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 0.63,
|
| 367 |
"learning_rate": 3e-05,
|
| 368 |
+
"loss": 1.1705,
|
| 369 |
"step": 480
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 0.64,
|
| 373 |
"learning_rate": 3e-05,
|
| 374 |
+
"loss": 1.1998,
|
| 375 |
"step": 490
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 0.66,
|
| 379 |
"learning_rate": 3e-05,
|
| 380 |
+
"loss": 1.1992,
|
| 381 |
"step": 500
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.66,
|
| 385 |
+
"eval_loss": 1.1840450763702393,
|
| 386 |
+
"eval_runtime": 19.5434,
|
| 387 |
+
"eval_samples_per_second": 25.226,
|
| 388 |
+
"eval_steps_per_second": 1.586,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 389 |
"step": 500
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.67,
|
| 393 |
"learning_rate": 3e-05,
|
| 394 |
+
"loss": 1.1842,
|
| 395 |
"step": 510
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.68,
|
| 399 |
"learning_rate": 3e-05,
|
| 400 |
+
"loss": 1.1598,
|
| 401 |
"step": 520
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.7,
|
| 405 |
"learning_rate": 3e-05,
|
| 406 |
+
"loss": 1.1538,
|
| 407 |
"step": 530
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 0.71,
|
| 411 |
"learning_rate": 3e-05,
|
| 412 |
+
"loss": 1.1506,
|
| 413 |
"step": 540
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.72,
|
| 417 |
"learning_rate": 3e-05,
|
| 418 |
+
"loss": 1.2032,
|
| 419 |
"step": 550
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 0.72,
|
| 423 |
+
"eval_loss": 1.1824493408203125,
|
| 424 |
+
"eval_runtime": 18.4972,
|
| 425 |
+
"eval_samples_per_second": 26.653,
|
| 426 |
+
"eval_steps_per_second": 1.676,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 427 |
"step": 550
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 0.73,
|
| 431 |
"learning_rate": 3e-05,
|
| 432 |
+
"loss": 1.1795,
|
| 433 |
"step": 560
|
| 434 |
},
|
| 435 |
{
|
| 436 |
+
"epoch": 0.75,
|
| 437 |
"learning_rate": 3e-05,
|
| 438 |
+
"loss": 1.1604,
|
| 439 |
"step": 570
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 0.76,
|
| 443 |
"learning_rate": 3e-05,
|
| 444 |
+
"loss": 1.1548,
|
| 445 |
"step": 580
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 0.77,
|
| 449 |
"learning_rate": 3e-05,
|
| 450 |
+
"loss": 1.1876,
|
| 451 |
"step": 590
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 0.79,
|
| 455 |
"learning_rate": 3e-05,
|
| 456 |
+
"loss": 1.1841,
|
| 457 |
"step": 600
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.79,
|
| 461 |
+
"eval_loss": 1.1797986030578613,
|
| 462 |
+
"eval_runtime": 19.5627,
|
| 463 |
+
"eval_samples_per_second": 25.201,
|
| 464 |
+
"eval_steps_per_second": 1.585,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 465 |
"step": 600
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 0.8,
|
| 469 |
"learning_rate": 3e-05,
|
| 470 |
+
"loss": 1.1579,
|
| 471 |
"step": 610
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.81,
|
| 475 |
"learning_rate": 3e-05,
|
| 476 |
+
"loss": 1.1858,
|
| 477 |
"step": 620
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 0.83,
|
| 481 |
"learning_rate": 3e-05,
|
| 482 |
+
"loss": 1.1994,
|
| 483 |
"step": 630
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 0.84,
|
| 487 |
"learning_rate": 3e-05,
|
| 488 |
+
"loss": 1.1712,
|
| 489 |
"step": 640
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 0.85,
|
| 493 |
"learning_rate": 3e-05,
|
| 494 |
+
"loss": 1.166,
|
| 495 |
"step": 650
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 0.85,
|
| 499 |
+
"eval_loss": 1.1789214611053467,
|
| 500 |
+
"eval_runtime": 19.1568,
|
| 501 |
+
"eval_samples_per_second": 25.735,
|
| 502 |
+
"eval_steps_per_second": 1.618,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 503 |
"step": 650
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 0.87,
|
| 507 |
"learning_rate": 3e-05,
|
| 508 |
+
"loss": 1.1426,
|
| 509 |
"step": 660
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 0.88,
|
| 513 |
"learning_rate": 3e-05,
|
| 514 |
+
"loss": 1.1291,
|
| 515 |
"step": 670
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 0.89,
|
| 519 |
"learning_rate": 3e-05,
|
| 520 |
+
"loss": 1.1825,
|
| 521 |
"step": 680
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 0.91,
|
| 525 |
"learning_rate": 3e-05,
|
| 526 |
+
"loss": 1.1214,
|
| 527 |
"step": 690
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.92,
|
| 531 |
"learning_rate": 3e-05,
|
| 532 |
+
"loss": 1.1641,
|
| 533 |
"step": 700
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 0.92,
|
| 537 |
+
"eval_loss": 1.1761133670806885,
|
| 538 |
+
"eval_runtime": 18.1966,
|
| 539 |
+
"eval_samples_per_second": 27.093,
|
| 540 |
+
"eval_steps_per_second": 1.704,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 541 |
"step": 700
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.93,
|
| 545 |
"learning_rate": 3e-05,
|
| 546 |
+
"loss": 1.1069,
|
| 547 |
"step": 710
|
| 548 |
},
|
| 549 |
{
|
| 550 |
+
"epoch": 0.94,
|
| 551 |
"learning_rate": 3e-05,
|
| 552 |
+
"loss": 1.1267,
|
| 553 |
"step": 720
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 0.96,
|
| 557 |
"learning_rate": 3e-05,
|
| 558 |
+
"loss": 1.1472,
|
| 559 |
"step": 730
|
| 560 |
},
|
| 561 |
{
|
| 562 |
+
"epoch": 0.97,
|
| 563 |
"learning_rate": 3e-05,
|
| 564 |
+
"loss": 1.2204,
|
| 565 |
"step": 740
|
| 566 |
},
|
| 567 |
{
|
| 568 |
+
"epoch": 0.98,
|
| 569 |
"learning_rate": 3e-05,
|
| 570 |
+
"loss": 1.1859,
|
| 571 |
"step": 750
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"epoch": 0.98,
|
| 575 |
+
"eval_loss": 1.1751502752304077,
|
| 576 |
+
"eval_runtime": 18.7143,
|
| 577 |
+
"eval_samples_per_second": 26.343,
|
| 578 |
+
"eval_steps_per_second": 1.656,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 579 |
"step": 750
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 1.0,
|
| 583 |
"learning_rate": 3e-05,
|
| 584 |
+
"loss": 1.1314,
|
| 585 |
"step": 760
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 1.01,
|
| 589 |
"learning_rate": 3e-05,
|
| 590 |
+
"loss": 1.12,
|
| 591 |
"step": 770
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 1.02,
|
| 595 |
"learning_rate": 3e-05,
|
| 596 |
+
"loss": 1.1007,
|
| 597 |
"step": 780
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 1.04,
|
| 601 |
"learning_rate": 3e-05,
|
| 602 |
+
"loss": 1.0822,
|
| 603 |
"step": 790
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 1.05,
|
| 607 |
"learning_rate": 3e-05,
|
| 608 |
+
"loss": 1.132,
|
| 609 |
"step": 800
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 1.05,
|
| 613 |
+
"eval_loss": 1.1736373901367188,
|
| 614 |
+
"eval_runtime": 19.2149,
|
| 615 |
+
"eval_samples_per_second": 25.657,
|
| 616 |
+
"eval_steps_per_second": 1.613,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 617 |
"step": 800
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 1.06,
|
| 621 |
"learning_rate": 3e-05,
|
| 622 |
+
"loss": 1.1076,
|
| 623 |
"step": 810
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"epoch": 1.08,
|
| 627 |
"learning_rate": 3e-05,
|
| 628 |
+
"loss": 1.1007,
|
| 629 |
"step": 820
|
| 630 |
},
|
| 631 |
{
|
| 632 |
+
"epoch": 1.09,
|
| 633 |
"learning_rate": 3e-05,
|
| 634 |
+
"loss": 1.1215,
|
| 635 |
"step": 830
|
| 636 |
},
|
| 637 |
{
|
| 638 |
+
"epoch": 1.1,
|
| 639 |
"learning_rate": 3e-05,
|
| 640 |
+
"loss": 1.0956,
|
| 641 |
"step": 840
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 1.12,
|
| 645 |
"learning_rate": 3e-05,
|
| 646 |
+
"loss": 1.1461,
|
| 647 |
"step": 850
|
| 648 |
},
|
| 649 |
{
|
| 650 |
+
"epoch": 1.12,
|
| 651 |
+
"eval_loss": 1.1723910570144653,
|
| 652 |
+
"eval_runtime": 18.6093,
|
| 653 |
+
"eval_samples_per_second": 26.492,
|
| 654 |
+
"eval_steps_per_second": 1.666,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 655 |
"step": 850
|
| 656 |
},
|
| 657 |
{
|
| 658 |
+
"epoch": 1.13,
|
| 659 |
"learning_rate": 3e-05,
|
| 660 |
+
"loss": 1.0818,
|
| 661 |
"step": 860
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"epoch": 1.14,
|
| 665 |
"learning_rate": 3e-05,
|
| 666 |
+
"loss": 1.0959,
|
| 667 |
"step": 870
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 1.15,
|
| 671 |
"learning_rate": 3e-05,
|
| 672 |
+
"loss": 1.0948,
|
| 673 |
"step": 880
|
| 674 |
},
|
| 675 |
{
|
| 676 |
+
"epoch": 1.17,
|
| 677 |
"learning_rate": 3e-05,
|
| 678 |
+
"loss": 1.1246,
|
| 679 |
"step": 890
|
| 680 |
},
|
| 681 |
{
|
| 682 |
+
"epoch": 1.18,
|
| 683 |
"learning_rate": 3e-05,
|
| 684 |
+
"loss": 1.0965,
|
| 685 |
"step": 900
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 1.18,
|
| 689 |
+
"eval_loss": 1.172638177871704,
|
| 690 |
+
"eval_runtime": 18.6729,
|
| 691 |
+
"eval_samples_per_second": 26.402,
|
| 692 |
+
"eval_steps_per_second": 1.66,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 693 |
"step": 900
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"epoch": 1.19,
|
| 697 |
"learning_rate": 3e-05,
|
| 698 |
+
"loss": 1.114,
|
| 699 |
"step": 910
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 1.21,
|
| 703 |
"learning_rate": 3e-05,
|
| 704 |
+
"loss": 1.1152,
|
| 705 |
"step": 920
|
| 706 |
},
|
| 707 |
{
|
| 708 |
+
"epoch": 1.22,
|
| 709 |
"learning_rate": 3e-05,
|
| 710 |
+
"loss": 1.0885,
|
| 711 |
"step": 930
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 1.23,
|
| 715 |
"learning_rate": 3e-05,
|
| 716 |
+
"loss": 1.1221,
|
| 717 |
"step": 940
|
| 718 |
},
|
| 719 |
{
|
| 720 |
+
"epoch": 1.25,
|
| 721 |
"learning_rate": 3e-05,
|
| 722 |
+
"loss": 1.1064,
|
| 723 |
"step": 950
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 1.25,
|
| 727 |
+
"eval_loss": 1.172351598739624,
|
| 728 |
+
"eval_runtime": 19.5348,
|
| 729 |
+
"eval_samples_per_second": 25.237,
|
| 730 |
+
"eval_steps_per_second": 1.587,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 731 |
"step": 950
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 1.26,
|
| 735 |
"learning_rate": 3e-05,
|
| 736 |
+
"loss": 1.0518,
|
| 737 |
"step": 960
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 1.27,
|
| 741 |
"learning_rate": 3e-05,
|
| 742 |
+
"loss": 1.0938,
|
| 743 |
"step": 970
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 1.29,
|
| 747 |
"learning_rate": 3e-05,
|
| 748 |
+
"loss": 1.1184,
|
| 749 |
"step": 980
|
| 750 |
},
|
| 751 |
{
|
| 752 |
+
"epoch": 1.3,
|
| 753 |
"learning_rate": 3e-05,
|
| 754 |
+
"loss": 1.09,
|
| 755 |
"step": 990
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 1.31,
|
| 759 |
"learning_rate": 3e-05,
|
| 760 |
+
"loss": 1.123,
|
| 761 |
"step": 1000
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"epoch": 1.31,
|
| 765 |
+
"eval_loss": 1.1728639602661133,
|
| 766 |
+
"eval_runtime": 18.8741,
|
| 767 |
+
"eval_samples_per_second": 26.12,
|
| 768 |
+
"eval_steps_per_second": 1.642,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 769 |
"step": 1000
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 1.33,
|
| 773 |
"learning_rate": 3e-05,
|
| 774 |
+
"loss": 1.1059,
|
| 775 |
"step": 1010
|
| 776 |
},
|
| 777 |
{
|
| 778 |
+
"epoch": 1.34,
|
| 779 |
"learning_rate": 3e-05,
|
| 780 |
+
"loss": 1.1061,
|
| 781 |
"step": 1020
|
| 782 |
},
|
| 783 |
{
|
| 784 |
+
"epoch": 1.35,
|
| 785 |
"learning_rate": 3e-05,
|
| 786 |
+
"loss": 1.1147,
|
| 787 |
"step": 1030
|
| 788 |
},
|
| 789 |
{
|
| 790 |
+
"epoch": 1.36,
|
| 791 |
"learning_rate": 3e-05,
|
| 792 |
+
"loss": 1.1322,
|
| 793 |
"step": 1040
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"epoch": 1.38,
|
| 797 |
"learning_rate": 3e-05,
|
| 798 |
+
"loss": 1.1079,
|
| 799 |
"step": 1050
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 1.38,
|
| 803 |
+
"eval_loss": 1.1694797277450562,
|
| 804 |
+
"eval_runtime": 18.842,
|
| 805 |
+
"eval_samples_per_second": 26.165,
|
| 806 |
+
"eval_steps_per_second": 1.645,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 807 |
"step": 1050
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 1.39,
|
| 811 |
"learning_rate": 3e-05,
|
| 812 |
+
"loss": 1.0826,
|
| 813 |
"step": 1060
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 1.4,
|
| 817 |
"learning_rate": 3e-05,
|
| 818 |
+
"loss": 1.1194,
|
| 819 |
"step": 1070
|
| 820 |
},
|
| 821 |
{
|
| 822 |
+
"epoch": 1.42,
|
| 823 |
"learning_rate": 3e-05,
|
| 824 |
+
"loss": 1.1398,
|
| 825 |
"step": 1080
|
| 826 |
},
|
| 827 |
{
|
| 828 |
+
"epoch": 1.43,
|
| 829 |
"learning_rate": 3e-05,
|
| 830 |
+
"loss": 1.11,
|
| 831 |
"step": 1090
|
| 832 |
},
|
| 833 |
{
|
| 834 |
+
"epoch": 1.44,
|
| 835 |
"learning_rate": 3e-05,
|
| 836 |
+
"loss": 1.12,
|
| 837 |
"step": 1100
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 1.44,
|
| 841 |
+
"eval_loss": 1.1707435846328735,
|
| 842 |
+
"eval_runtime": 19.0362,
|
| 843 |
+
"eval_samples_per_second": 25.898,
|
| 844 |
+
"eval_steps_per_second": 1.628,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 845 |
"step": 1100
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 1.46,
|
| 849 |
"learning_rate": 3e-05,
|
| 850 |
+
"loss": 1.0891,
|
| 851 |
"step": 1110
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"epoch": 1.47,
|
| 855 |
"learning_rate": 3e-05,
|
| 856 |
+
"loss": 1.1216,
|
| 857 |
"step": 1120
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 1.48,
|
| 861 |
"learning_rate": 3e-05,
|
| 862 |
+
"loss": 1.1122,
|
| 863 |
"step": 1130
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 1.5,
|
| 867 |
"learning_rate": 3e-05,
|
| 868 |
+
"loss": 1.1065,
|
| 869 |
"step": 1140
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 1.51,
|
| 873 |
"learning_rate": 3e-05,
|
| 874 |
+
"loss": 1.1288,
|
| 875 |
"step": 1150
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 1.51,
|
| 879 |
+
"eval_loss": 1.1693464517593384,
|
| 880 |
+
"eval_runtime": 19.1181,
|
| 881 |
+
"eval_samples_per_second": 25.787,
|
| 882 |
+
"eval_steps_per_second": 1.622,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 883 |
"step": 1150
|
| 884 |
},
|
| 885 |
{
|
| 886 |
+
"epoch": 1.52,
|
| 887 |
"learning_rate": 3e-05,
|
| 888 |
+
"loss": 1.1145,
|
| 889 |
"step": 1160
|
| 890 |
},
|
| 891 |
{
|
| 892 |
+
"epoch": 1.54,
|
| 893 |
"learning_rate": 3e-05,
|
| 894 |
+
"loss": 1.0812,
|
| 895 |
"step": 1170
|
| 896 |
},
|
| 897 |
{
|
| 898 |
+
"epoch": 1.55,
|
| 899 |
"learning_rate": 3e-05,
|
| 900 |
+
"loss": 1.1291,
|
| 901 |
"step": 1180
|
| 902 |
},
|
| 903 |
{
|
| 904 |
+
"epoch": 1.56,
|
| 905 |
"learning_rate": 3e-05,
|
| 906 |
+
"loss": 1.1114,
|
| 907 |
"step": 1190
|
| 908 |
},
|
| 909 |
{
|
| 910 |
+
"epoch": 1.57,
|
| 911 |
"learning_rate": 3e-05,
|
| 912 |
+
"loss": 1.133,
|
| 913 |
"step": 1200
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 1.57,
|
| 917 |
+
"eval_loss": 1.1675716638565063,
|
| 918 |
+
"eval_runtime": 19.1116,
|
| 919 |
+
"eval_samples_per_second": 25.796,
|
| 920 |
+
"eval_steps_per_second": 1.622,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 921 |
"step": 1200
|
| 922 |
},
|
| 923 |
{
|
| 924 |
+
"epoch": 1.59,
|
| 925 |
"learning_rate": 3e-05,
|
| 926 |
+
"loss": 1.0918,
|
| 927 |
"step": 1210
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 1.6,
|
| 931 |
"learning_rate": 3e-05,
|
| 932 |
+
"loss": 1.1009,
|
| 933 |
"step": 1220
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 1.61,
|
| 937 |
"learning_rate": 3e-05,
|
| 938 |
+
"loss": 1.1279,
|
| 939 |
"step": 1230
|
| 940 |
},
|
| 941 |
{
|
| 942 |
+
"epoch": 1.63,
|
| 943 |
"learning_rate": 3e-05,
|
| 944 |
+
"loss": 1.1314,
|
| 945 |
"step": 1240
|
| 946 |
},
|
| 947 |
{
|
| 948 |
+
"epoch": 1.64,
|
| 949 |
"learning_rate": 3e-05,
|
| 950 |
+
"loss": 1.1647,
|
| 951 |
"step": 1250
|
| 952 |
},
|
| 953 |
{
|
| 954 |
+
"epoch": 1.64,
|
| 955 |
+
"eval_loss": 1.1693305969238281,
|
| 956 |
+
"eval_runtime": 18.9258,
|
| 957 |
+
"eval_samples_per_second": 26.049,
|
| 958 |
+
"eval_steps_per_second": 1.638,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 959 |
"step": 1250
|
| 960 |
},
|
| 961 |
{
|
| 962 |
+
"epoch": 1.65,
|
| 963 |
"learning_rate": 3e-05,
|
| 964 |
+
"loss": 1.0633,
|
| 965 |
"step": 1260
|
| 966 |
},
|
| 967 |
{
|
| 968 |
+
"epoch": 1.67,
|
| 969 |
"learning_rate": 3e-05,
|
| 970 |
+
"loss": 1.0961,
|
| 971 |
"step": 1270
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 1.68,
|
| 975 |
"learning_rate": 3e-05,
|
| 976 |
+
"loss": 1.1106,
|
| 977 |
"step": 1280
|
| 978 |
},
|
| 979 |
{
|
| 980 |
+
"epoch": 1.69,
|
| 981 |
"learning_rate": 3e-05,
|
| 982 |
+
"loss": 1.1233,
|
| 983 |
"step": 1290
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 1.71,
|
| 987 |
"learning_rate": 3e-05,
|
| 988 |
+
"loss": 1.1269,
|
| 989 |
"step": 1300
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 1.71,
|
| 993 |
+
"eval_loss": 1.1658315658569336,
|
| 994 |
+
"eval_runtime": 18.972,
|
| 995 |
+
"eval_samples_per_second": 25.986,
|
| 996 |
+
"eval_steps_per_second": 1.634,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 997 |
"step": 1300
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
+
"epoch": 1.72,
|
| 1001 |
"learning_rate": 3e-05,
|
| 1002 |
+
"loss": 1.0683,
|
| 1003 |
"step": 1310
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
+
"epoch": 1.73,
|
| 1007 |
"learning_rate": 3e-05,
|
| 1008 |
+
"loss": 1.1079,
|
| 1009 |
"step": 1320
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
+
"epoch": 1.75,
|
| 1013 |
"learning_rate": 3e-05,
|
| 1014 |
+
"loss": 1.1367,
|
| 1015 |
"step": 1330
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
+
"epoch": 1.76,
|
| 1019 |
"learning_rate": 3e-05,
|
| 1020 |
+
"loss": 1.1077,
|
| 1021 |
"step": 1340
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
+
"epoch": 1.77,
|
| 1025 |
"learning_rate": 3e-05,
|
| 1026 |
+
"loss": 1.1332,
|
| 1027 |
"step": 1350
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
+
"epoch": 1.77,
|
| 1031 |
+
"eval_loss": 1.1656816005706787,
|
| 1032 |
+
"eval_runtime": 19.0244,
|
| 1033 |
+
"eval_samples_per_second": 25.914,
|
| 1034 |
+
"eval_steps_per_second": 1.629,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1035 |
"step": 1350
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
+
"epoch": 1.78,
|
| 1039 |
"learning_rate": 3e-05,
|
| 1040 |
+
"loss": 1.0921,
|
| 1041 |
"step": 1360
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
+
"epoch": 1.8,
|
| 1045 |
"learning_rate": 3e-05,
|
| 1046 |
+
"loss": 1.0669,
|
| 1047 |
"step": 1370
|
| 1048 |
},
|
| 1049 |
{
|
| 1050 |
+
"epoch": 1.81,
|
| 1051 |
"learning_rate": 3e-05,
|
| 1052 |
+
"loss": 1.1185,
|
| 1053 |
"step": 1380
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
+
"epoch": 1.82,
|
| 1057 |
"learning_rate": 3e-05,
|
| 1058 |
+
"loss": 1.108,
|
| 1059 |
"step": 1390
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
+
"epoch": 1.84,
|
| 1063 |
"learning_rate": 3e-05,
|
| 1064 |
+
"loss": 1.1276,
|
| 1065 |
"step": 1400
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
+
"epoch": 1.84,
|
| 1069 |
+
"eval_loss": 1.1681002378463745,
|
| 1070 |
+
"eval_runtime": 18.4913,
|
| 1071 |
+
"eval_samples_per_second": 26.661,
|
| 1072 |
+
"eval_steps_per_second": 1.676,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1073 |
"step": 1400
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"epoch": 1.85,
|
| 1077 |
"learning_rate": 3e-05,
|
| 1078 |
+
"loss": 1.0666,
|
| 1079 |
"step": 1410
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
+
"epoch": 1.86,
|
| 1083 |
"learning_rate": 3e-05,
|
| 1084 |
+
"loss": 1.1286,
|
| 1085 |
"step": 1420
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
+
"epoch": 1.88,
|
| 1089 |
"learning_rate": 3e-05,
|
| 1090 |
+
"loss": 1.1286,
|
| 1091 |
"step": 1430
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
+
"epoch": 1.89,
|
| 1095 |
"learning_rate": 3e-05,
|
| 1096 |
+
"loss": 1.0967,
|
| 1097 |
"step": 1440
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
+
"epoch": 1.9,
|
| 1101 |
"learning_rate": 3e-05,
|
| 1102 |
+
"loss": 1.1361,
|
| 1103 |
"step": 1450
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
+
"epoch": 1.9,
|
| 1107 |
+
"eval_loss": 1.1633367538452148,
|
| 1108 |
+
"eval_runtime": 18.9278,
|
| 1109 |
+
"eval_samples_per_second": 26.046,
|
| 1110 |
+
"eval_steps_per_second": 1.638,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1111 |
"step": 1450
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
+
"epoch": 1.92,
|
| 1115 |
+
"learning_rate": 3e-05,
|
| 1116 |
+
"loss": 1.0907,
|
| 1117 |
+
"step": 1460
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 1.93,
|
| 1121 |
+
"learning_rate": 3e-05,
|
| 1122 |
+
"loss": 1.1137,
|
| 1123 |
+
"step": 1470
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.94,
|
| 1127 |
+
"learning_rate": 3e-05,
|
| 1128 |
+
"loss": 1.125,
|
| 1129 |
+
"step": 1480
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.96,
|
| 1133 |
+
"learning_rate": 3e-05,
|
| 1134 |
+
"loss": 1.1047,
|
| 1135 |
+
"step": 1490
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 1.97,
|
| 1139 |
+
"learning_rate": 3e-05,
|
| 1140 |
+
"loss": 1.1205,
|
| 1141 |
+
"step": 1500
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 1.97,
|
| 1145 |
+
"eval_loss": 1.1639732122421265,
|
| 1146 |
+
"eval_runtime": 19.1908,
|
| 1147 |
+
"eval_samples_per_second": 25.689,
|
| 1148 |
+
"eval_steps_per_second": 1.615,
|
| 1149 |
+
"step": 1500
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 1.98,
|
| 1153 |
+
"learning_rate": 3e-05,
|
| 1154 |
+
"loss": 1.0902,
|
| 1155 |
+
"step": 1510
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 1.99,
|
| 1159 |
+
"learning_rate": 3e-05,
|
| 1160 |
+
"loss": 1.0794,
|
| 1161 |
+
"step": 1520
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 2.0,
|
| 1165 |
+
"step": 1524,
|
| 1166 |
+
"total_flos": 2.2001611985471406e+18,
|
| 1167 |
+
"train_loss": 1.1488912840840697,
|
| 1168 |
+
"train_runtime": 10306.9039,
|
| 1169 |
+
"train_samples_per_second": 9.463,
|
| 1170 |
+
"train_steps_per_second": 0.148
|
| 1171 |
}
|
| 1172 |
],
|
| 1173 |
"logging_steps": 10,
|
| 1174 |
+
"max_steps": 1524,
|
| 1175 |
"num_input_tokens_seen": 0,
|
| 1176 |
"num_train_epochs": 2,
|
| 1177 |
+
"save_steps": 100,
|
| 1178 |
+
"total_flos": 2.2001611985471406e+18,
|
| 1179 |
"train_batch_size": 4,
|
| 1180 |
"trial_name": null,
|
| 1181 |
"trial_params": null
|