Model save
Browse files- README.md +20 -21
- adapter_model.safetensors +1 -1
- all_results.json +5 -10
- runs/Jun10_01-58-39_48ddfe8e991f/events.out.tfevents.1717984740.48ddfe8e991f.6850.0 +2 -2
- train_results.json +5 -5
- trainer_state.json +232 -584
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: google/gemma-2b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: gemma2b-summarize-gemini1_5flash-32k
|
| 14 |
results: []
|
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# gemma2b-summarize-gemini1_5flash-32k
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 2.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -41,14 +40,14 @@ More information needed
|
|
| 41 |
|
| 42 |
The following hyperparameters were used during training:
|
| 43 |
- learning_rate: 0.0002
|
| 44 |
-
- train_batch_size:
|
| 45 |
-
- eval_batch_size:
|
| 46 |
- seed: 42
|
| 47 |
- distributed_type: multi-GPU
|
| 48 |
-
- num_devices:
|
| 49 |
- gradient_accumulation_steps: 2
|
| 50 |
-
- total_train_batch_size:
|
| 51 |
-
- total_eval_batch_size:
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: cosine
|
| 54 |
- lr_scheduler_warmup_ratio: 0.1
|
|
@@ -58,22 +57,22 @@ The following hyperparameters were used during training:
|
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 61 |
-
|
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
| 1.
|
| 65 |
-
| 1.
|
| 66 |
-
|
|
| 67 |
-
|
|
| 68 |
-
|
|
| 69 |
-
|
|
| 70 |
-
|
|
| 71 |
|
| 72 |
|
| 73 |
### Framework versions
|
| 74 |
|
| 75 |
- PEFT 0.11.1
|
| 76 |
-
- Transformers 4.
|
| 77 |
-
- Pytorch 2.
|
| 78 |
- Datasets 2.19.2
|
| 79 |
- Tokenizers 0.19.1
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-2b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: gemma2b-summarize-gemini1_5flash-32k
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# gemma2b-summarize-gemini1_5flash-32k
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.6817
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 0.0002
|
| 43 |
+
- train_batch_size: 16
|
| 44 |
+
- eval_batch_size: 16
|
| 45 |
- seed: 42
|
| 46 |
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 256
|
| 50 |
+
- total_eval_batch_size: 128
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: cosine
|
| 53 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 57 |
|
| 58 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 2.0459 | 1.0 | 26 | 2.8633 |
|
| 61 |
+
| 1.2724 | 2.0 | 52 | 2.7063 |
|
| 62 |
+
| 1.1654 | 3.0 | 78 | 2.6695 |
|
| 63 |
+
| 1.1056 | 4.0 | 104 | 2.6545 |
|
| 64 |
+
| 1.0615 | 5.0 | 130 | 2.6669 |
|
| 65 |
+
| 1.0347 | 6.0 | 156 | 2.6682 |
|
| 66 |
+
| 1.0216 | 7.0 | 182 | 2.6747 |
|
| 67 |
+
| 1.0204 | 8.0 | 208 | 2.6796 |
|
| 68 |
+
| 1.0071 | 9.0 | 234 | 2.6812 |
|
| 69 |
+
| 1.0088 | 10.0 | 260 | 2.6817 |
|
| 70 |
|
| 71 |
|
| 72 |
### Framework versions
|
| 73 |
|
| 74 |
- PEFT 0.11.1
|
| 75 |
+
- Transformers 4.41.2
|
| 76 |
+
- Pytorch 2.3.1+cu121
|
| 77 |
- Datasets 2.19.2
|
| 78 |
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 39256960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:536482c5985510f4f7f093fc65c047f999d635c7bf5bf55261f804b90ad98048
|
| 3 |
size 39256960
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 19.422,
|
| 7 |
-
"eval_steps_per_second": 1.942,
|
| 8 |
-
"total_flos": 4.072455240941568e+17,
|
| 9 |
-
"train_loss": 1.1447187717144305,
|
| 10 |
-
"train_runtime": 1794.4362,
|
| 11 |
"train_samples": 31676,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.092505436769485e+17,
|
| 4 |
+
"train_loss": 1.2536941931797907,
|
| 5 |
+
"train_runtime": 1396.0468,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 31676,
|
| 7 |
+
"train_samples_per_second": 47.677,
|
| 8 |
+
"train_steps_per_second": 0.186
|
| 9 |
}
|
runs/Jun10_01-58-39_48ddfe8e991f/events.out.tfevents.1717984740.48ddfe8e991f.6850.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56bdc6b260d62b9809b427415acfb0f41bbd71bf538046cbfc338740902abacb
|
| 3 |
+
size 19728
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"total_flos": 4.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 31676,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.092505436769485e+17,
|
| 4 |
+
"train_loss": 1.2536941931797907,
|
| 5 |
+
"train_runtime": 1396.0468,
|
| 6 |
"train_samples": 31676,
|
| 7 |
+
"train_samples_per_second": 47.677,
|
| 8 |
+
"train_steps_per_second": 0.186
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -3,843 +3,491 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 10.0,
|
| 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": 2.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 3.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.09615384615384616,
|
| 20 |
-
"grad_norm": 2.25,
|
| 21 |
-
"learning_rate": 1.923076923076923e-05,
|
| 22 |
-
"loss": 3.0172,
|
| 23 |
-
"step": 5
|
| 24 |
-
},
|
| 25 |
{
|
| 26 |
"epoch": 0.19230769230769232,
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 3.846153846153846e-05,
|
| 29 |
-
"loss":
|
| 30 |
-
"step":
|
| 31 |
-
},
|
| 32 |
-
{
|
| 33 |
-
"epoch": 0.28846153846153844,
|
| 34 |
-
"grad_norm": 3.28125,
|
| 35 |
-
"learning_rate": 5.769230769230769e-05,
|
| 36 |
-
"loss": 2.8517,
|
| 37 |
-
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.38461538461538464,
|
| 41 |
-
"grad_norm": 2.
|
| 42 |
"learning_rate": 7.692307692307693e-05,
|
| 43 |
-
"loss":
|
| 44 |
-
"step":
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.4807692307692308,
|
| 48 |
-
"grad_norm": 1.296875,
|
| 49 |
-
"learning_rate": 9.615384615384617e-05,
|
| 50 |
-
"loss": 2.3838,
|
| 51 |
-
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5769230769230769,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 0.00011538461538461538,
|
| 57 |
-
"loss": 2.
|
| 58 |
-
"step":
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 0.6730769230769231,
|
| 62 |
-
"grad_norm": 0.90234375,
|
| 63 |
-
"learning_rate": 0.00013461538461538464,
|
| 64 |
-
"loss": 1.9352,
|
| 65 |
-
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.7692307692307693,
|
| 69 |
-
"grad_norm": 1.
|
| 70 |
"learning_rate": 0.00015384615384615385,
|
| 71 |
-
"loss":
|
| 72 |
-
"step":
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 0.8653846153846154,
|
| 76 |
-
"grad_norm": 0.65625,
|
| 77 |
-
"learning_rate": 0.0001730769230769231,
|
| 78 |
-
"loss": 1.6127,
|
| 79 |
-
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.9615384615384616,
|
| 83 |
-
"grad_norm":
|
| 84 |
"learning_rate": 0.00019230769230769233,
|
| 85 |
-
"loss":
|
| 86 |
-
"step":
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 1.0,
|
| 90 |
-
"eval_loss": 2.
|
| 91 |
-
"eval_runtime": 0.
|
| 92 |
-
"eval_samples_per_second":
|
| 93 |
-
"eval_steps_per_second":
|
| 94 |
-
"step":
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 1.0576923076923077,
|
| 98 |
-
"grad_norm": 0.37109375,
|
| 99 |
-
"learning_rate": 0.00019997972289848503,
|
| 100 |
-
"loss": 1.4008,
|
| 101 |
-
"step": 55
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.1538461538461537,
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
"learning_rate": 0.00019985583705641418,
|
| 107 |
-
"loss": 1.
|
| 108 |
-
"step":
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 1.25,
|
| 112 |
-
"grad_norm": 0.232421875,
|
| 113 |
-
"learning_rate": 0.00019961946980917456,
|
| 114 |
-
"loss": 1.2736,
|
| 115 |
-
"step": 65
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.3461538461538463,
|
| 119 |
-
"grad_norm": 0.
|
| 120 |
"learning_rate": 0.0001992708874098054,
|
| 121 |
-
"loss": 1.
|
| 122 |
-
"step":
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 1.4423076923076923,
|
| 126 |
-
"grad_norm": 0.5,
|
| 127 |
-
"learning_rate": 0.0001988104825147528,
|
| 128 |
-
"loss": 1.2168,
|
| 129 |
-
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.5384615384615383,
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
"learning_rate": 0.00019823877374156647,
|
| 135 |
-
"loss": 1.
|
| 136 |
-
"step":
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 1.6346153846153846,
|
| 140 |
-
"grad_norm": 0.921875,
|
| 141 |
-
"learning_rate": 0.00019755640508470942,
|
| 142 |
-
"loss": 1.1812,
|
| 143 |
-
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 1.7307692307692308,
|
| 147 |
-
"grad_norm": 0.
|
| 148 |
"learning_rate": 0.00019676414519013781,
|
| 149 |
-
"loss": 1.
|
| 150 |
-
"step":
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 1.8269230769230769,
|
| 154 |
-
"grad_norm": 0.82421875,
|
| 155 |
-
"learning_rate": 0.00019586288648946947,
|
| 156 |
-
"loss": 1.1671,
|
| 157 |
-
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 1.9230769230769231,
|
| 161 |
-
"grad_norm": 0.
|
| 162 |
"learning_rate": 0.00019485364419471454,
|
| 163 |
-
"loss": 1.
|
| 164 |
-
"step":
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.0,
|
| 168 |
-
"eval_loss": 2.
|
| 169 |
-
"eval_runtime": 0.
|
| 170 |
-
"eval_samples_per_second":
|
| 171 |
-
"eval_steps_per_second":
|
| 172 |
-
"step":
|
| 173 |
-
},
|
| 174 |
-
{
|
| 175 |
-
"epoch": 2.019230769230769,
|
| 176 |
-
"grad_norm": 0.30078125,
|
| 177 |
-
"learning_rate": 0.00019373755515470254,
|
| 178 |
-
"loss": 1.1342,
|
| 179 |
-
"step": 105
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.1153846153846154,
|
| 183 |
-
"grad_norm": 0.
|
| 184 |
"learning_rate": 0.00019251587657449236,
|
| 185 |
-
"loss": 1.
|
| 186 |
-
"step":
|
| 187 |
-
},
|
| 188 |
-
{
|
| 189 |
-
"epoch": 2.2115384615384617,
|
| 190 |
-
"grad_norm": 0.3515625,
|
| 191 |
-
"learning_rate": 0.00019118998459920902,
|
| 192 |
-
"loss": 1.104,
|
| 193 |
-
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 2.3076923076923075,
|
| 197 |
-
"grad_norm": 0.
|
| 198 |
"learning_rate": 0.0001897613727639014,
|
| 199 |
-
"loss": 1.
|
| 200 |
-
"step":
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 2.4038461538461537,
|
| 204 |
-
"grad_norm": 0.5625,
|
| 205 |
-
"learning_rate": 0.0001882316503111678,
|
| 206 |
-
"loss": 1.0938,
|
| 207 |
-
"step": 125
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 2.5,
|
| 211 |
-
"grad_norm": 0.
|
| 212 |
"learning_rate": 0.00018660254037844388,
|
| 213 |
-
"loss": 1.
|
| 214 |
-
"step":
|
| 215 |
-
},
|
| 216 |
-
{
|
| 217 |
-
"epoch": 2.5961538461538463,
|
| 218 |
-
"grad_norm": 0.54296875,
|
| 219 |
-
"learning_rate": 0.00018487587805699526,
|
| 220 |
-
"loss": 1.0819,
|
| 221 |
-
"step": 135
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.6923076923076925,
|
| 225 |
-
"grad_norm": 0.
|
| 226 |
"learning_rate": 0.00018305360832480117,
|
| 227 |
-
"loss": 1.
|
| 228 |
-
"step":
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 2.7884615384615383,
|
| 232 |
-
"grad_norm": 0.416015625,
|
| 233 |
-
"learning_rate": 0.00018113778385565733,
|
| 234 |
-
"loss": 1.0805,
|
| 235 |
-
"step": 145
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 2.8846153846153846,
|
| 239 |
-
"grad_norm": 0.
|
| 240 |
"learning_rate": 0.0001791305627069662,
|
| 241 |
-
"loss": 1.
|
| 242 |
-
"step":
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 2.980769230769231,
|
| 246 |
-
"grad_norm": 0.267578125,
|
| 247 |
-
"learning_rate": 0.00017703420588881946,
|
| 248 |
-
"loss": 1.0743,
|
| 249 |
-
"step": 155
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 3.0,
|
| 253 |
-
"eval_loss": 2.
|
| 254 |
-
"eval_runtime": 0.
|
| 255 |
-
"eval_samples_per_second":
|
| 256 |
-
"eval_steps_per_second":
|
| 257 |
-
"step":
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 3.076923076923077,
|
| 261 |
-
"grad_norm": 0.
|
| 262 |
"learning_rate": 0.00017485107481711012,
|
| 263 |
-
"loss": 1.
|
| 264 |
-
"step":
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 3.173076923076923,
|
| 268 |
-
"grad_norm": 0.267578125,
|
| 269 |
-
"learning_rate": 0.00017258362865354426,
|
| 270 |
-
"loss": 1.0387,
|
| 271 |
-
"step": 165
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 3.269230769230769,
|
| 275 |
-
"grad_norm": 0.
|
| 276 |
"learning_rate": 0.00017023442153554777,
|
| 277 |
-
"loss": 1.
|
| 278 |
-
"step":
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 3.3653846153846154,
|
| 282 |
-
"grad_norm": 0.361328125,
|
| 283 |
-
"learning_rate": 0.0001678060996991891,
|
| 284 |
-
"loss": 1.0387,
|
| 285 |
-
"step": 175
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 3.4615384615384617,
|
| 289 |
-
"grad_norm": 0.
|
| 290 |
"learning_rate": 0.0001653013984983585,
|
| 291 |
-
"loss": 1.
|
| 292 |
-
"step":
|
| 293 |
-
},
|
| 294 |
-
{
|
| 295 |
-
"epoch": 3.5576923076923075,
|
| 296 |
-
"grad_norm": 0.5,
|
| 297 |
-
"learning_rate": 0.00016272313932356162,
|
| 298 |
-
"loss": 1.0507,
|
| 299 |
-
"step": 185
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 3.6538461538461537,
|
| 303 |
-
"grad_norm": 0.
|
| 304 |
"learning_rate": 0.0001600742264237979,
|
| 305 |
-
"loss": 1.
|
| 306 |
-
"step":
|
| 307 |
-
},
|
| 308 |
-
{
|
| 309 |
-
"epoch": 3.75,
|
| 310 |
-
"grad_norm": 0.30859375,
|
| 311 |
-
"learning_rate": 0.0001573576436351046,
|
| 312 |
-
"loss": 1.0336,
|
| 313 |
-
"step": 195
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.8461538461538463,
|
| 317 |
-
"grad_norm": 0.
|
| 318 |
"learning_rate": 0.00015457645101945046,
|
| 319 |
-
"loss": 1.
|
| 320 |
-
"step":
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 3.9423076923076925,
|
| 324 |
-
"grad_norm": 0.40625,
|
| 325 |
-
"learning_rate": 0.00015173378141776568,
|
| 326 |
-
"loss": 1.0412,
|
| 327 |
-
"step": 205
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 4.0,
|
| 331 |
-
"eval_loss": 2.
|
| 332 |
-
"eval_runtime": 0.
|
| 333 |
-
"eval_samples_per_second":
|
| 334 |
-
"eval_steps_per_second":
|
| 335 |
-
"step":
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 4.038461538461538,
|
| 339 |
-
"grad_norm": 0.
|
| 340 |
"learning_rate": 0.00014883283692099112,
|
| 341 |
-
"loss": 1.
|
| 342 |
-
"step":
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"epoch": 4.134615384615385,
|
| 346 |
-
"grad_norm": 0.490234375,
|
| 347 |
-
"learning_rate": 0.00014587688526312143,
|
| 348 |
-
"loss": 1.0145,
|
| 349 |
-
"step": 215
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 4.230769230769231,
|
| 353 |
-
"grad_norm": 0.
|
| 354 |
"learning_rate": 0.00014286925614030542,
|
| 355 |
-
"loss": 1.
|
| 356 |
-
"step":
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 4.326923076923077,
|
| 360 |
-
"grad_norm": 0.306640625,
|
| 361 |
-
"learning_rate": 0.0001398133374601501,
|
| 362 |
-
"loss": 1.0137,
|
| 363 |
-
"step": 225
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 4.423076923076923,
|
| 367 |
-
"grad_norm": 0.
|
| 368 |
"learning_rate": 0.00013671257152545277,
|
| 369 |
-
"loss": 1.
|
| 370 |
-
"step":
|
| 371 |
-
},
|
| 372 |
-
{
|
| 373 |
-
"epoch": 4.519230769230769,
|
| 374 |
-
"grad_norm": 1.0625,
|
| 375 |
-
"learning_rate": 0.0001335704511566605,
|
| 376 |
-
"loss": 1.0112,
|
| 377 |
-
"step": 235
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 4.615384615384615,
|
| 381 |
-
"grad_norm": 0.
|
| 382 |
"learning_rate": 0.0001303905157574247,
|
| 383 |
-
"loss": 1.
|
| 384 |
-
"step":
|
| 385 |
-
},
|
| 386 |
-
{
|
| 387 |
-
"epoch": 4.711538461538462,
|
| 388 |
-
"grad_norm": 0.373046875,
|
| 389 |
-
"learning_rate": 0.00012717634732768243,
|
| 390 |
-
"loss": 1.0055,
|
| 391 |
-
"step": 245
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.8076923076923075,
|
| 395 |
-
"grad_norm": 0.
|
| 396 |
"learning_rate": 0.0001239315664287558,
|
| 397 |
-
"loss": 1.
|
| 398 |
-
"step":
|
| 399 |
-
},
|
| 400 |
-
{
|
| 401 |
-
"epoch": 4.903846153846154,
|
| 402 |
-
"grad_norm": 0.326171875,
|
| 403 |
-
"learning_rate": 0.00012065982810501404,
|
| 404 |
-
"loss": 1.0077,
|
| 405 |
-
"step": 255
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 5.0,
|
| 409 |
-
"grad_norm": 0.
|
| 410 |
"learning_rate": 0.00011736481776669306,
|
| 411 |
-
"loss": 1.
|
| 412 |
-
"step":
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 5.0,
|
| 416 |
-
"eval_loss": 2.
|
| 417 |
-
"eval_runtime": 0.
|
| 418 |
-
"eval_samples_per_second":
|
| 419 |
-
"eval_steps_per_second":
|
| 420 |
-
"step":
|
| 421 |
-
},
|
| 422 |
-
{
|
| 423 |
-
"epoch": 5.096153846153846,
|
| 424 |
-
"grad_norm": 0.44921875,
|
| 425 |
-
"learning_rate": 0.00011405024703850929,
|
| 426 |
-
"loss": 0.9883,
|
| 427 |
-
"step": 265
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 5.1923076923076925,
|
| 431 |
-
"grad_norm": 0.
|
| 432 |
"learning_rate": 0.00011071984957874479,
|
| 433 |
-
"loss":
|
| 434 |
-
"step":
|
| 435 |
-
},
|
| 436 |
-
{
|
| 437 |
-
"epoch": 5.288461538461538,
|
| 438 |
-
"grad_norm": 0.45703125,
|
| 439 |
-
"learning_rate": 0.00010737737687351284,
|
| 440 |
-
"loss": 0.9871,
|
| 441 |
-
"step": 275
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 5.384615384615385,
|
| 445 |
-
"grad_norm": 0.
|
| 446 |
"learning_rate": 0.00010402659401094152,
|
| 447 |
-
"loss":
|
| 448 |
-
"step":
|
| 449 |
-
},
|
| 450 |
-
{
|
| 451 |
-
"epoch": 5.480769230769231,
|
| 452 |
-
"grad_norm": 0.36328125,
|
| 453 |
-
"learning_rate": 0.00010067127544003563,
|
| 454 |
-
"loss": 0.9884,
|
| 455 |
-
"step": 285
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 5.576923076923077,
|
| 459 |
-
"grad_norm": 0.
|
| 460 |
"learning_rate": 9.73152007189939e-05,
|
| 461 |
-
"loss":
|
| 462 |
-
"step":
|
| 463 |
-
},
|
| 464 |
-
{
|
| 465 |
-
"epoch": 5.673076923076923,
|
| 466 |
-
"grad_norm": 0.33203125,
|
| 467 |
-
"learning_rate": 9.396215025777139e-05,
|
| 468 |
-
"loss": 0.9891,
|
| 469 |
-
"step": 295
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 5.769230769230769,
|
| 473 |
-
"grad_norm": 0.
|
| 474 |
"learning_rate": 9.061590105968208e-05,
|
| 475 |
-
"loss":
|
| 476 |
-
"step":
|
| 477 |
-
},
|
| 478 |
-
{
|
| 479 |
-
"epoch": 5.865384615384615,
|
| 480 |
-
"grad_norm": 0.404296875,
|
| 481 |
-
"learning_rate": 8.728022246683894e-05,
|
| 482 |
-
"loss": 0.9928,
|
| 483 |
-
"step": 305
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.961538461538462,
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
"learning_rate": 8.395887191422397e-05,
|
| 489 |
-
"loss":
|
| 490 |
-
"step":
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.0,
|
| 494 |
-
"eval_loss": 2.
|
| 495 |
-
"eval_runtime": 0.
|
| 496 |
-
"eval_samples_per_second":
|
| 497 |
-
"eval_steps_per_second":
|
| 498 |
-
"step":
|
| 499 |
-
},
|
| 500 |
-
{
|
| 501 |
-
"epoch": 6.0576923076923075,
|
| 502 |
-
"grad_norm": 0.3046875,
|
| 503 |
-
"learning_rate": 8.065559069717088e-05,
|
| 504 |
-
"loss": 0.9754,
|
| 505 |
-
"step": 315
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 6.153846153846154,
|
| 509 |
-
"grad_norm": 0.
|
| 510 |
"learning_rate": 7.73740997570278e-05,
|
| 511 |
-
"loss":
|
| 512 |
-
"step":
|
| 513 |
-
},
|
| 514 |
-
{
|
| 515 |
-
"epoch": 6.25,
|
| 516 |
-
"grad_norm": 0.41796875,
|
| 517 |
-
"learning_rate": 7.411809548974792e-05,
|
| 518 |
-
"loss": 0.9699,
|
| 519 |
-
"step": 325
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 6.346153846153846,
|
| 523 |
-
"grad_norm": 0.
|
| 524 |
"learning_rate": 7.089124558212871e-05,
|
| 525 |
-
"loss":
|
| 526 |
-
"step":
|
| 527 |
-
},
|
| 528 |
-
{
|
| 529 |
-
"epoch": 6.4423076923076925,
|
| 530 |
-
"grad_norm": 0.48046875,
|
| 531 |
-
"learning_rate": 6.769718488039023e-05,
|
| 532 |
-
"loss": 0.9686,
|
| 533 |
-
"step": 335
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 6.538461538461538,
|
| 537 |
-
"grad_norm": 0.
|
| 538 |
"learning_rate": 6.453951129574644e-05,
|
| 539 |
-
"loss":
|
| 540 |
-
"step":
|
| 541 |
-
},
|
| 542 |
-
{
|
| 543 |
-
"epoch": 6.634615384615385,
|
| 544 |
-
"grad_norm": 0.3359375,
|
| 545 |
-
"learning_rate": 6.142178175158149e-05,
|
| 546 |
-
"loss": 0.9746,
|
| 547 |
-
"step": 345
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 6.730769230769231,
|
| 551 |
-
"grad_norm": 0.
|
| 552 |
"learning_rate": 5.834750817679606e-05,
|
| 553 |
-
"loss":
|
| 554 |
-
"step":
|
| 555 |
-
},
|
| 556 |
-
{
|
| 557 |
-
"epoch": 6.826923076923077,
|
| 558 |
-
"grad_norm": 0.31640625,
|
| 559 |
-
"learning_rate": 5.5320153549837415e-05,
|
| 560 |
-
"loss": 0.9788,
|
| 561 |
-
"step": 355
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 6.923076923076923,
|
| 565 |
-
"grad_norm": 0.
|
| 566 |
"learning_rate": 5.234312799786921e-05,
|
| 567 |
-
"loss":
|
| 568 |
-
"step":
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 7.0,
|
| 572 |
-
"eval_loss": 2.
|
| 573 |
-
"eval_runtime": 0.
|
| 574 |
-
"eval_samples_per_second":
|
| 575 |
-
"eval_steps_per_second":
|
| 576 |
-
"step":
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 7.019230769230769,
|
| 580 |
-
"grad_norm": 0.353515625,
|
| 581 |
-
"learning_rate": 4.9419784955474524e-05,
|
| 582 |
-
"loss": 0.9709,
|
| 583 |
-
"step": 365
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.115384615384615,
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
"learning_rate": 4.6553417387219886e-05,
|
| 589 |
-
"loss":
|
| 590 |
-
"step":
|
| 591 |
-
},
|
| 592 |
-
{
|
| 593 |
-
"epoch": 7.211538461538462,
|
| 594 |
-
"grad_norm": 0.310546875,
|
| 595 |
-
"learning_rate": 4.374725407833532e-05,
|
| 596 |
-
"loss": 0.9601,
|
| 597 |
-
"step": 375
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.3076923076923075,
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
"learning_rate": 4.100445599768774e-05,
|
| 603 |
-
"loss":
|
| 604 |
-
"step":
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 7.403846153846154,
|
| 608 |
-
"grad_norm": 0.4140625,
|
| 609 |
-
"learning_rate": 3.832811273714569e-05,
|
| 610 |
-
"loss": 0.9669,
|
| 611 |
-
"step": 385
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.5,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
"learning_rate": 3.5721239031346066e-05,
|
| 617 |
-
"loss":
|
| 618 |
-
"step":
|
| 619 |
-
},
|
| 620 |
-
{
|
| 621 |
-
"epoch": 7.596153846153846,
|
| 622 |
-
"grad_norm": 0.328125,
|
| 623 |
-
"learning_rate": 3.318677136178228e-05,
|
| 624 |
-
"loss": 0.9557,
|
| 625 |
-
"step": 395
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.6923076923076925,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
"learning_rate": 3.072756464904006e-05,
|
| 631 |
-
"loss":
|
| 632 |
-
"step":
|
| 633 |
-
},
|
| 634 |
-
{
|
| 635 |
-
"epoch": 7.788461538461538,
|
| 636 |
-
"grad_norm": 0.341796875,
|
| 637 |
-
"learning_rate": 2.8346389036906828e-05,
|
| 638 |
-
"loss": 0.9638,
|
| 639 |
-
"step": 405
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 7.884615384615385,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
"learning_rate": 2.6045926771976303e-05,
|
| 645 |
-
"loss":
|
| 646 |
-
"step":
|
| 647 |
-
},
|
| 648 |
-
{
|
| 649 |
-
"epoch": 7.980769230769231,
|
| 650 |
-
"grad_norm": 0.310546875,
|
| 651 |
-
"learning_rate": 2.382876918226409e-05,
|
| 652 |
-
"loss": 0.9637,
|
| 653 |
-
"step": 415
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 8.0,
|
| 657 |
-
"eval_loss": 2.
|
| 658 |
-
"eval_runtime": 0.
|
| 659 |
-
"eval_samples_per_second":
|
| 660 |
-
"eval_steps_per_second":
|
| 661 |
-
"step":
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 8.076923076923077,
|
| 665 |
-
"grad_norm": 0.
|
| 666 |
"learning_rate": 2.1697413758237784e-05,
|
| 667 |
-
"loss":
|
| 668 |
-
"step":
|
| 669 |
-
},
|
| 670 |
-
{
|
| 671 |
-
"epoch": 8.173076923076923,
|
| 672 |
-
"grad_norm": 0.30859375,
|
| 673 |
-
"learning_rate": 1.965426133954854e-05,
|
| 674 |
-
"loss": 0.9599,
|
| 675 |
-
"step": 425
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 8.26923076923077,
|
| 679 |
-
"grad_norm": 0.
|
| 680 |
"learning_rate": 1.7701613410634365e-05,
|
| 681 |
-
"loss":
|
| 682 |
-
"step":
|
| 683 |
-
},
|
| 684 |
-
{
|
| 685 |
-
"epoch": 8.365384615384615,
|
| 686 |
-
"grad_norm": 0.314453125,
|
| 687 |
-
"learning_rate": 1.584166950824061e-05,
|
| 688 |
-
"loss": 0.9586,
|
| 689 |
-
"step": 435
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 8.461538461538462,
|
| 693 |
-
"grad_norm": 0.
|
| 694 |
"learning_rate": 1.4076524743778319e-05,
|
| 695 |
-
"loss":
|
| 696 |
-
"step":
|
| 697 |
-
},
|
| 698 |
-
{
|
| 699 |
-
"epoch": 8.557692307692308,
|
| 700 |
-
"grad_norm": 0.29296875,
|
| 701 |
-
"learning_rate": 1.2408167443311214e-05,
|
| 702 |
-
"loss": 0.9549,
|
| 703 |
-
"step": 445
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 8.653846153846153,
|
| 707 |
-
"grad_norm": 0.
|
| 708 |
"learning_rate": 1.083847690782972e-05,
|
| 709 |
-
"loss":
|
| 710 |
-
"step":
|
| 711 |
-
},
|
| 712 |
-
{
|
| 713 |
-
"epoch": 8.75,
|
| 714 |
-
"grad_norm": 0.326171875,
|
| 715 |
-
"learning_rate": 9.369221296335006e-06,
|
| 716 |
-
"loss": 0.9558,
|
| 717 |
-
"step": 455
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 8.846153846153847,
|
| 721 |
-
"grad_norm": 0.
|
| 722 |
"learning_rate": 8.002055634117578e-06,
|
| 723 |
-
"loss":
|
| 724 |
-
"step":
|
| 725 |
-
},
|
| 726 |
-
{
|
| 727 |
-
"epoch": 8.942307692307692,
|
| 728 |
-
"grad_norm": 0.298828125,
|
| 729 |
-
"learning_rate": 6.738519948473976e-06,
|
| 730 |
-
"loss": 0.9604,
|
| 731 |
-
"step": 465
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 9.0,
|
| 735 |
-
"eval_loss": 2.
|
| 736 |
-
"eval_runtime": 0.
|
| 737 |
-
"eval_samples_per_second":
|
| 738 |
-
"eval_steps_per_second":
|
| 739 |
-
"step":
|
| 740 |
},
|
| 741 |
{
|
| 742 |
"epoch": 9.038461538461538,
|
| 743 |
-
"grad_norm": 0.
|
| 744 |
"learning_rate": 5.580037533961546e-06,
|
| 745 |
-
"loss":
|
| 746 |
-
"step":
|
| 747 |
-
},
|
| 748 |
-
{
|
| 749 |
-
"epoch": 9.134615384615385,
|
| 750 |
-
"grad_norm": 0.294921875,
|
| 751 |
-
"learning_rate": 4.527913349145441e-06,
|
| 752 |
-
"loss": 0.9503,
|
| 753 |
-
"step": 475
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 9.23076923076923,
|
| 757 |
-
"grad_norm": 0.
|
| 758 |
"learning_rate": 3.5833325466437694e-06,
|
| 759 |
-
"loss":
|
| 760 |
-
"step":
|
| 761 |
-
},
|
| 762 |
-
{
|
| 763 |
-
"epoch": 9.326923076923077,
|
| 764 |
-
"grad_norm": 0.3046875,
|
| 765 |
-
"learning_rate": 2.7473591381266708e-06,
|
| 766 |
-
"loss": 0.9571,
|
| 767 |
-
"step": 485
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 9.423076923076923,
|
| 771 |
-
"grad_norm": 0.
|
| 772 |
"learning_rate": 2.0209347957732328e-06,
|
| 773 |
-
"loss":
|
| 774 |
-
"step":
|
| 775 |
-
},
|
| 776 |
-
{
|
| 777 |
-
"epoch": 9.51923076923077,
|
| 778 |
-
"grad_norm": 0.30859375,
|
| 779 |
-
"learning_rate": 1.404877791536141e-06,
|
| 780 |
-
"loss": 0.9599,
|
| 781 |
-
"step": 495
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 9.615384615384615,
|
| 785 |
-
"grad_norm": 0.
|
| 786 |
"learning_rate": 8.998820754091531e-07,
|
| 787 |
-
"loss":
|
| 788 |
-
"step":
|
| 789 |
-
},
|
| 790 |
-
{
|
| 791 |
-
"epoch": 9.711538461538462,
|
| 792 |
-
"grad_norm": 0.30078125,
|
| 793 |
-
"learning_rate": 5.065164937354428e-07,
|
| 794 |
-
"loss": 0.9578,
|
| 795 |
-
"step": 505
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 9.807692307692308,
|
| 799 |
-
"grad_norm": 0.
|
| 800 |
"learning_rate": 2.2522414843748618e-07,
|
| 801 |
-
"loss":
|
| 802 |
-
"step":
|
| 803 |
-
},
|
| 804 |
-
{
|
| 805 |
-
"epoch": 9.903846153846153,
|
| 806 |
-
"grad_norm": 0.29296875,
|
| 807 |
-
"learning_rate": 5.632189789027687e-08,
|
| 808 |
-
"loss": 0.9593,
|
| 809 |
-
"step": 515
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 10.0,
|
| 813 |
-
"grad_norm": 0.
|
| 814 |
"learning_rate": 0.0,
|
| 815 |
-
"loss":
|
| 816 |
-
"step":
|
| 817 |
},
|
| 818 |
{
|
| 819 |
"epoch": 10.0,
|
| 820 |
-
"eval_loss": 2.
|
| 821 |
-
"eval_runtime": 0.
|
| 822 |
-
"eval_samples_per_second":
|
| 823 |
-
"eval_steps_per_second":
|
| 824 |
-
"step":
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 10.0,
|
| 828 |
-
"step":
|
| 829 |
-
"total_flos": 4.
|
| 830 |
-
"train_loss": 1.
|
| 831 |
-
"train_runtime":
|
| 832 |
-
"train_samples_per_second":
|
| 833 |
-
"train_steps_per_second": 0.
|
| 834 |
}
|
| 835 |
],
|
| 836 |
"logging_steps": 5,
|
| 837 |
-
"max_steps":
|
| 838 |
"num_input_tokens_seen": 0,
|
| 839 |
"num_train_epochs": 10,
|
| 840 |
"save_steps": 100,
|
| 841 |
-
"
|
| 842 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 843 |
"trial_name": null,
|
| 844 |
"trial_params": null
|
| 845 |
}
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 260,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.038461538461538464,
|
| 13 |
+
"grad_norm": 2.890625,
|
| 14 |
+
"learning_rate": 7.692307692307694e-06,
|
| 15 |
+
"loss": 3.3093,
|
| 16 |
"step": 1
|
| 17 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
{
|
| 19 |
"epoch": 0.19230769230769232,
|
| 20 |
+
"grad_norm": 4.90625,
|
| 21 |
"learning_rate": 3.846153846153846e-05,
|
| 22 |
+
"loss": 3.317,
|
| 23 |
+
"step": 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.38461538461538464,
|
| 27 |
+
"grad_norm": 2.25,
|
| 28 |
"learning_rate": 7.692307692307693e-05,
|
| 29 |
+
"loss": 3.1254,
|
| 30 |
+
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.5769230769230769,
|
| 34 |
+
"grad_norm": 3.328125,
|
| 35 |
"learning_rate": 0.00011538461538461538,
|
| 36 |
+
"loss": 2.6178,
|
| 37 |
+
"step": 15
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.7692307692307693,
|
| 41 |
+
"grad_norm": 1.5390625,
|
| 42 |
"learning_rate": 0.00015384615384615385,
|
| 43 |
+
"loss": 2.3151,
|
| 44 |
+
"step": 20
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.9615384615384616,
|
| 48 |
+
"grad_norm": 0.98828125,
|
| 49 |
"learning_rate": 0.00019230769230769233,
|
| 50 |
+
"loss": 2.0459,
|
| 51 |
+
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 1.0,
|
| 55 |
+
"eval_loss": 2.863295793533325,
|
| 56 |
+
"eval_runtime": 0.4843,
|
| 57 |
+
"eval_samples_per_second": 43.364,
|
| 58 |
+
"eval_steps_per_second": 2.065,
|
| 59 |
+
"step": 26
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 1.1538461538461537,
|
| 63 |
+
"grad_norm": 0.546875,
|
| 64 |
"learning_rate": 0.00019985583705641418,
|
| 65 |
+
"loss": 1.6915,
|
| 66 |
+
"step": 30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.3461538461538463,
|
| 70 |
+
"grad_norm": 0.400390625,
|
| 71 |
"learning_rate": 0.0001992708874098054,
|
| 72 |
+
"loss": 1.4892,
|
| 73 |
+
"step": 35
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 1.5384615384615383,
|
| 77 |
+
"grad_norm": 0.515625,
|
| 78 |
"learning_rate": 0.00019823877374156647,
|
| 79 |
+
"loss": 1.3906,
|
| 80 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 1.7307692307692308,
|
| 84 |
+
"grad_norm": 0.73046875,
|
| 85 |
"learning_rate": 0.00019676414519013781,
|
| 86 |
+
"loss": 1.3301,
|
| 87 |
+
"step": 45
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 1.9230769230769231,
|
| 91 |
+
"grad_norm": 0.310546875,
|
| 92 |
"learning_rate": 0.00019485364419471454,
|
| 93 |
+
"loss": 1.2724,
|
| 94 |
+
"step": 50
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 2.0,
|
| 98 |
+
"eval_loss": 2.706272602081299,
|
| 99 |
+
"eval_runtime": 0.4877,
|
| 100 |
+
"eval_samples_per_second": 43.056,
|
| 101 |
+
"eval_steps_per_second": 2.05,
|
| 102 |
+
"step": 52
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 2.1153846153846154,
|
| 106 |
+
"grad_norm": 0.2373046875,
|
| 107 |
"learning_rate": 0.00019251587657449236,
|
| 108 |
+
"loss": 1.2358,
|
| 109 |
+
"step": 55
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 2.3076923076923075,
|
| 113 |
+
"grad_norm": 0.427734375,
|
| 114 |
"learning_rate": 0.0001897613727639014,
|
| 115 |
+
"loss": 1.2118,
|
| 116 |
+
"step": 60
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 2.5,
|
| 120 |
+
"grad_norm": 0.412109375,
|
| 121 |
"learning_rate": 0.00018660254037844388,
|
| 122 |
+
"loss": 1.1899,
|
| 123 |
+
"step": 65
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.6923076923076925,
|
| 127 |
+
"grad_norm": 0.7109375,
|
| 128 |
"learning_rate": 0.00018305360832480117,
|
| 129 |
+
"loss": 1.1662,
|
| 130 |
+
"step": 70
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 2.8846153846153846,
|
| 134 |
+
"grad_norm": 0.29296875,
|
| 135 |
"learning_rate": 0.0001791305627069662,
|
| 136 |
+
"loss": 1.1654,
|
| 137 |
+
"step": 75
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 3.0,
|
| 141 |
+
"eval_loss": 2.669499397277832,
|
| 142 |
+
"eval_runtime": 0.492,
|
| 143 |
+
"eval_samples_per_second": 42.682,
|
| 144 |
+
"eval_steps_per_second": 2.032,
|
| 145 |
+
"step": 78
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 3.076923076923077,
|
| 149 |
+
"grad_norm": 0.322265625,
|
| 150 |
"learning_rate": 0.00017485107481711012,
|
| 151 |
+
"loss": 1.1435,
|
| 152 |
+
"step": 80
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 3.269230769230769,
|
| 156 |
+
"grad_norm": 0.3359375,
|
| 157 |
"learning_rate": 0.00017023442153554777,
|
| 158 |
+
"loss": 1.1188,
|
| 159 |
+
"step": 85
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 3.4615384615384617,
|
| 163 |
+
"grad_norm": 0.25,
|
| 164 |
"learning_rate": 0.0001653013984983585,
|
| 165 |
+
"loss": 1.124,
|
| 166 |
+
"step": 90
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 3.6538461538461537,
|
| 170 |
+
"grad_norm": 0.41015625,
|
| 171 |
"learning_rate": 0.0001600742264237979,
|
| 172 |
+
"loss": 1.108,
|
| 173 |
+
"step": 95
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 3.8461538461538463,
|
| 177 |
+
"grad_norm": 0.5546875,
|
| 178 |
"learning_rate": 0.00015457645101945046,
|
| 179 |
+
"loss": 1.1056,
|
| 180 |
+
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 4.0,
|
| 184 |
+
"eval_loss": 2.654463291168213,
|
| 185 |
+
"eval_runtime": 0.4845,
|
| 186 |
+
"eval_samples_per_second": 43.339,
|
| 187 |
+
"eval_steps_per_second": 2.064,
|
| 188 |
+
"step": 104
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 4.038461538461538,
|
| 192 |
+
"grad_norm": 0.33984375,
|
| 193 |
"learning_rate": 0.00014883283692099112,
|
| 194 |
+
"loss": 1.0918,
|
| 195 |
+
"step": 105
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 4.230769230769231,
|
| 199 |
+
"grad_norm": 0.2373046875,
|
| 200 |
"learning_rate": 0.00014286925614030542,
|
| 201 |
+
"loss": 1.0814,
|
| 202 |
+
"step": 110
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 4.423076923076923,
|
| 206 |
+
"grad_norm": 0.236328125,
|
| 207 |
"learning_rate": 0.00013671257152545277,
|
| 208 |
+
"loss": 1.0778,
|
| 209 |
+
"step": 115
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 4.615384615384615,
|
| 213 |
+
"grad_norm": 0.2275390625,
|
| 214 |
"learning_rate": 0.0001303905157574247,
|
| 215 |
+
"loss": 1.0717,
|
| 216 |
+
"step": 120
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 4.8076923076923075,
|
| 220 |
+
"grad_norm": 0.46875,
|
| 221 |
"learning_rate": 0.0001239315664287558,
|
| 222 |
+
"loss": 1.0643,
|
| 223 |
+
"step": 125
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 5.0,
|
| 227 |
+
"grad_norm": 0.224609375,
|
| 228 |
"learning_rate": 0.00011736481776669306,
|
| 229 |
+
"loss": 1.0615,
|
| 230 |
+
"step": 130
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 5.0,
|
| 234 |
+
"eval_loss": 2.6669278144836426,
|
| 235 |
+
"eval_runtime": 0.4883,
|
| 236 |
+
"eval_samples_per_second": 43.009,
|
| 237 |
+
"eval_steps_per_second": 2.048,
|
| 238 |
+
"step": 130
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 5.1923076923076925,
|
| 242 |
+
"grad_norm": 0.208984375,
|
| 243 |
"learning_rate": 0.00011071984957874479,
|
| 244 |
+
"loss": 1.0503,
|
| 245 |
+
"step": 135
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 5.384615384615385,
|
| 249 |
+
"grad_norm": 0.2421875,
|
| 250 |
"learning_rate": 0.00010402659401094152,
|
| 251 |
+
"loss": 1.05,
|
| 252 |
+
"step": 140
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 5.576923076923077,
|
| 256 |
+
"grad_norm": 0.2578125,
|
| 257 |
"learning_rate": 9.73152007189939e-05,
|
| 258 |
+
"loss": 1.0449,
|
| 259 |
+
"step": 145
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 5.769230769230769,
|
| 263 |
+
"grad_norm": 0.22265625,
|
| 264 |
"learning_rate": 9.061590105968208e-05,
|
| 265 |
+
"loss": 1.0451,
|
| 266 |
+
"step": 150
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 5.961538461538462,
|
| 270 |
+
"grad_norm": 0.337890625,
|
| 271 |
"learning_rate": 8.395887191422397e-05,
|
| 272 |
+
"loss": 1.0347,
|
| 273 |
+
"step": 155
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 6.0,
|
| 277 |
+
"eval_loss": 2.668196439743042,
|
| 278 |
+
"eval_runtime": 0.4907,
|
| 279 |
+
"eval_samples_per_second": 42.799,
|
| 280 |
+
"eval_steps_per_second": 2.038,
|
| 281 |
+
"step": 156
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 6.153846153846154,
|
| 285 |
+
"grad_norm": 0.42578125,
|
| 286 |
"learning_rate": 7.73740997570278e-05,
|
| 287 |
+
"loss": 1.0339,
|
| 288 |
+
"step": 160
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 6.346153846153846,
|
| 292 |
+
"grad_norm": 0.28515625,
|
| 293 |
"learning_rate": 7.089124558212871e-05,
|
| 294 |
+
"loss": 1.0281,
|
| 295 |
+
"step": 165
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 6.538461538461538,
|
| 299 |
+
"grad_norm": 0.41796875,
|
| 300 |
"learning_rate": 6.453951129574644e-05,
|
| 301 |
+
"loss": 1.0306,
|
| 302 |
+
"step": 170
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 6.730769230769231,
|
| 306 |
+
"grad_norm": 0.279296875,
|
| 307 |
"learning_rate": 5.834750817679606e-05,
|
| 308 |
+
"loss": 1.0268,
|
| 309 |
+
"step": 175
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 6.923076923076923,
|
| 313 |
+
"grad_norm": 0.2119140625,
|
| 314 |
"learning_rate": 5.234312799786921e-05,
|
| 315 |
+
"loss": 1.0216,
|
| 316 |
+
"step": 180
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 7.0,
|
| 320 |
+
"eval_loss": 2.674699544906616,
|
| 321 |
+
"eval_runtime": 0.4901,
|
| 322 |
+
"eval_samples_per_second": 42.844,
|
| 323 |
+
"eval_steps_per_second": 2.04,
|
| 324 |
+
"step": 182
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 7.115384615384615,
|
| 328 |
+
"grad_norm": 0.205078125,
|
| 329 |
"learning_rate": 4.6553417387219886e-05,
|
| 330 |
+
"loss": 1.0217,
|
| 331 |
+
"step": 185
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 7.3076923076923075,
|
| 335 |
+
"grad_norm": 0.205078125,
|
| 336 |
"learning_rate": 4.100445599768774e-05,
|
| 337 |
+
"loss": 1.0184,
|
| 338 |
+
"step": 190
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 7.5,
|
| 342 |
+
"grad_norm": 0.251953125,
|
| 343 |
"learning_rate": 3.5721239031346066e-05,
|
| 344 |
+
"loss": 1.0163,
|
| 345 |
+
"step": 195
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 7.6923076923076925,
|
| 349 |
+
"grad_norm": 0.2080078125,
|
| 350 |
"learning_rate": 3.072756464904006e-05,
|
| 351 |
+
"loss": 1.0133,
|
| 352 |
+
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 7.884615384615385,
|
| 356 |
+
"grad_norm": 0.2099609375,
|
| 357 |
"learning_rate": 2.6045926771976303e-05,
|
| 358 |
+
"loss": 1.0204,
|
| 359 |
+
"step": 205
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 8.0,
|
| 363 |
+
"eval_loss": 2.679577112197876,
|
| 364 |
+
"eval_runtime": 0.4883,
|
| 365 |
+
"eval_samples_per_second": 43.003,
|
| 366 |
+
"eval_steps_per_second": 2.048,
|
| 367 |
+
"step": 208
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 8.076923076923077,
|
| 371 |
+
"grad_norm": 0.19921875,
|
| 372 |
"learning_rate": 2.1697413758237784e-05,
|
| 373 |
+
"loss": 1.0097,
|
| 374 |
+
"step": 210
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 8.26923076923077,
|
| 378 |
+
"grad_norm": 0.208984375,
|
| 379 |
"learning_rate": 1.7701613410634365e-05,
|
| 380 |
+
"loss": 1.0168,
|
| 381 |
+
"step": 215
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 8.461538461538462,
|
| 385 |
+
"grad_norm": 0.2041015625,
|
| 386 |
"learning_rate": 1.4076524743778319e-05,
|
| 387 |
+
"loss": 1.0141,
|
| 388 |
+
"step": 220
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 8.653846153846153,
|
| 392 |
+
"grad_norm": 0.1962890625,
|
| 393 |
"learning_rate": 1.083847690782972e-05,
|
| 394 |
+
"loss": 1.0125,
|
| 395 |
+
"step": 225
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 8.846153846153847,
|
| 399 |
+
"grad_norm": 0.193359375,
|
| 400 |
"learning_rate": 8.002055634117578e-06,
|
| 401 |
+
"loss": 1.0071,
|
| 402 |
+
"step": 230
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 9.0,
|
| 406 |
+
"eval_loss": 2.6812119483947754,
|
| 407 |
+
"eval_runtime": 0.4891,
|
| 408 |
+
"eval_samples_per_second": 42.934,
|
| 409 |
+
"eval_steps_per_second": 2.044,
|
| 410 |
+
"step": 234
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 9.038461538461538,
|
| 414 |
+
"grad_norm": 0.1884765625,
|
| 415 |
"learning_rate": 5.580037533961546e-06,
|
| 416 |
+
"loss": 1.0137,
|
| 417 |
+
"step": 235
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 9.23076923076923,
|
| 421 |
+
"grad_norm": 0.1953125,
|
| 422 |
"learning_rate": 3.5833325466437694e-06,
|
| 423 |
+
"loss": 1.0116,
|
| 424 |
+
"step": 240
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 9.423076923076923,
|
| 428 |
+
"grad_norm": 0.1865234375,
|
| 429 |
"learning_rate": 2.0209347957732328e-06,
|
| 430 |
+
"loss": 1.0093,
|
| 431 |
+
"step": 245
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 9.615384615384615,
|
| 435 |
+
"grad_norm": 0.185546875,
|
| 436 |
"learning_rate": 8.998820754091531e-07,
|
| 437 |
+
"loss": 1.0136,
|
| 438 |
+
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 9.807692307692308,
|
| 442 |
+
"grad_norm": 0.1923828125,
|
| 443 |
"learning_rate": 2.2522414843748618e-07,
|
| 444 |
+
"loss": 1.008,
|
| 445 |
+
"step": 255
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 10.0,
|
| 449 |
+
"grad_norm": 0.1875,
|
| 450 |
"learning_rate": 0.0,
|
| 451 |
+
"loss": 1.0088,
|
| 452 |
+
"step": 260
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 10.0,
|
| 456 |
+
"eval_loss": 2.681701183319092,
|
| 457 |
+
"eval_runtime": 0.4897,
|
| 458 |
+
"eval_samples_per_second": 42.88,
|
| 459 |
+
"eval_steps_per_second": 2.042,
|
| 460 |
+
"step": 260
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 10.0,
|
| 464 |
+
"step": 260,
|
| 465 |
+
"total_flos": 4.092505436769485e+17,
|
| 466 |
+
"train_loss": 1.2536941931797907,
|
| 467 |
+
"train_runtime": 1396.0468,
|
| 468 |
+
"train_samples_per_second": 47.677,
|
| 469 |
+
"train_steps_per_second": 0.186
|
| 470 |
}
|
| 471 |
],
|
| 472 |
"logging_steps": 5,
|
| 473 |
+
"max_steps": 260,
|
| 474 |
"num_input_tokens_seen": 0,
|
| 475 |
"num_train_epochs": 10,
|
| 476 |
"save_steps": 100,
|
| 477 |
+
"stateful_callbacks": {
|
| 478 |
+
"TrainerControl": {
|
| 479 |
+
"args": {
|
| 480 |
+
"should_epoch_stop": false,
|
| 481 |
+
"should_evaluate": false,
|
| 482 |
+
"should_log": false,
|
| 483 |
+
"should_save": true,
|
| 484 |
+
"should_training_stop": false
|
| 485 |
+
},
|
| 486 |
+
"attributes": {}
|
| 487 |
+
}
|
| 488 |
+
},
|
| 489 |
+
"total_flos": 4.092505436769485e+17,
|
| 490 |
+
"train_batch_size": 16,
|
| 491 |
"trial_name": null,
|
| 492 |
"trial_params": null
|
| 493 |
}
|