Model save
Browse files- README.md +4 -4
- all_results.json +6 -7
- generation_config.json +1 -1
- train_results.json +6 -7
- trainer_state.json +1413 -154
README.md
CHANGED
|
@@ -27,17 +27,17 @@ print(output["generated_text"])
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/hhy-tue/huggingface/runs/
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with SFT.
|
| 34 |
|
| 35 |
### Framework versions
|
| 36 |
|
| 37 |
-
- TRL: 0.
|
| 38 |
-
- Transformers: 4.49.0
|
| 39 |
- Pytorch: 2.5.1
|
| 40 |
-
- Datasets: 3.2
|
| 41 |
- Tokenizers: 0.21.0
|
| 42 |
|
| 43 |
## Citations
|
|
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/hhy-tue/huggingface/runs/mfqpfnyh)
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with SFT.
|
| 34 |
|
| 35 |
### Framework versions
|
| 36 |
|
| 37 |
+
- TRL: 0.16.0.dev0
|
| 38 |
+
- Transformers: 4.49.0
|
| 39 |
- Pytorch: 2.5.1
|
| 40 |
+
- Datasets: 3.3.2
|
| 41 |
- Tokenizers: 0.21.0
|
| 42 |
|
| 43 |
## Citations
|
all_results.json
CHANGED
|
@@ -1,9 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"train_steps_per_second": 0.368
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 4.418757235321078e+18,
|
| 3 |
+
"train_loss": 0.5497396646150902,
|
| 4 |
+
"train_runtime": 11730.1118,
|
| 5 |
+
"train_samples": 93733,
|
| 6 |
+
"train_samples_per_second": 2.924,
|
| 7 |
+
"train_steps_per_second": 0.091
|
|
|
|
| 8 |
}
|
generation_config.json
CHANGED
|
@@ -10,5 +10,5 @@
|
|
| 10 |
"temperature": 0.7,
|
| 11 |
"top_k": 20,
|
| 12 |
"top_p": 0.8,
|
| 13 |
-
"transformers_version": "4.49.0
|
| 14 |
}
|
|
|
|
| 10 |
"temperature": 0.7,
|
| 11 |
"top_k": 20,
|
| 12 |
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.49.0"
|
| 14 |
}
|
train_results.json
CHANGED
|
@@ -1,9 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"train_steps_per_second": 0.368
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"total_flos": 4.418757235321078e+18,
|
| 3 |
+
"train_loss": 0.5497396646150902,
|
| 4 |
+
"train_runtime": 11730.1118,
|
| 5 |
+
"train_samples": 93733,
|
| 6 |
+
"train_samples_per_second": 2.924,
|
| 7 |
+
"train_steps_per_second": 0.091
|
|
|
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -2,280 +2,1539 @@
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 5
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss":
|
| 30 |
"step": 15
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 1.
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 25
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 35
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 40
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 45
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 50
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 55
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 60
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 65
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 75
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 85
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 95
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"
|
| 154 |
-
"
|
| 155 |
-
"
|
| 156 |
-
"eval_steps_per_second": 5.173,
|
| 157 |
-
"step": 100
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 0.621301775147929,
|
| 161 |
-
"grad_norm": 306.0,
|
| 162 |
-
"learning_rate": 7.545145128592009e-06,
|
| 163 |
-
"loss": 0.8314,
|
| 164 |
"step": 105
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 0.
|
| 168 |
-
"grad_norm":
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 110
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"grad_norm":
|
| 176 |
-
"learning_rate":
|
| 177 |
-
"loss": 0.
|
| 178 |
"step": 115
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 0.
|
| 182 |
-
"grad_norm":
|
| 183 |
-
"learning_rate": 4.
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 120
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 0.
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 125
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 0.
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate":
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 130
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch": 0.
|
| 203 |
-
"grad_norm":
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 135
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"grad_norm":
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 140
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate":
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 145
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch": 0.
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 150
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 0.
|
| 231 |
-
"grad_norm":
|
| 232 |
-
"learning_rate": 4.
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 155
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 160
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch": 0.
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 165
|
| 249 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 250 |
{
|
| 251 |
"epoch": 1.0,
|
| 252 |
-
"step":
|
| 253 |
-
"total_flos":
|
| 254 |
-
"train_loss": 0.
|
| 255 |
-
"train_runtime":
|
| 256 |
-
"train_samples_per_second":
|
| 257 |
-
"train_steps_per_second": 0.
|
| 258 |
}
|
| 259 |
],
|
| 260 |
"logging_steps": 5,
|
| 261 |
-
"max_steps":
|
| 262 |
"num_input_tokens_seen": 0,
|
| 263 |
"num_train_epochs": 1,
|
| 264 |
-
"save_steps":
|
| 265 |
"stateful_callbacks": {
|
| 266 |
"TrainerControl": {
|
| 267 |
"args": {
|
| 268 |
"should_epoch_stop": false,
|
| 269 |
"should_evaluate": false,
|
| 270 |
"should_log": false,
|
| 271 |
-
"should_save":
|
| 272 |
-
"should_training_stop":
|
| 273 |
},
|
| 274 |
"attributes": {}
|
| 275 |
}
|
| 276 |
},
|
| 277 |
-
"total_flos":
|
| 278 |
-
"train_batch_size":
|
| 279 |
"trial_name": null,
|
| 280 |
"trial_params": null
|
| 281 |
}
|
|
|
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1072,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0046641791044776115,
|
| 13 |
+
"grad_norm": 2.0296592712402344,
|
| 14 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 15 |
+
"loss": 0.8606,
|
| 16 |
"step": 5
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.009328358208955223,
|
| 20 |
+
"grad_norm": 1.3905454874038696,
|
| 21 |
+
"learning_rate": 9.259259259259259e-06,
|
| 22 |
+
"loss": 0.8093,
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.013992537313432836,
|
| 27 |
+
"grad_norm": 1.113548755645752,
|
| 28 |
+
"learning_rate": 1.388888888888889e-05,
|
| 29 |
+
"loss": 0.77,
|
| 30 |
"step": 15
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.018656716417910446,
|
| 34 |
+
"grad_norm": 0.6648120880126953,
|
| 35 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 36 |
+
"loss": 0.7113,
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.02332089552238806,
|
| 41 |
+
"grad_norm": 0.5477202534675598,
|
| 42 |
+
"learning_rate": 2.314814814814815e-05,
|
| 43 |
+
"loss": 0.692,
|
| 44 |
"step": 25
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.027985074626865673,
|
| 48 |
+
"grad_norm": 0.4421328008174896,
|
| 49 |
+
"learning_rate": 2.777777777777778e-05,
|
| 50 |
+
"loss": 0.6771,
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.03264925373134328,
|
| 55 |
+
"grad_norm": 0.41744956374168396,
|
| 56 |
+
"learning_rate": 3.240740740740741e-05,
|
| 57 |
+
"loss": 0.6643,
|
| 58 |
"step": 35
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.03731343283582089,
|
| 62 |
+
"grad_norm": 0.42775049805641174,
|
| 63 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 64 |
+
"loss": 0.6361,
|
| 65 |
"step": 40
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.04197761194029851,
|
| 69 |
+
"grad_norm": 0.4391756057739258,
|
| 70 |
+
"learning_rate": 4.166666666666667e-05,
|
| 71 |
+
"loss": 0.6508,
|
| 72 |
"step": 45
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.04664179104477612,
|
| 76 |
+
"grad_norm": 0.4183003604412079,
|
| 77 |
+
"learning_rate": 4.62962962962963e-05,
|
| 78 |
+
"loss": 0.6313,
|
| 79 |
"step": 50
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.051305970149253734,
|
| 83 |
+
"grad_norm": 0.37660548090934753,
|
| 84 |
+
"learning_rate": 4.999989285883431e-05,
|
| 85 |
+
"loss": 0.6322,
|
| 86 |
"step": 55
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.055970149253731345,
|
| 90 |
+
"grad_norm": 0.4807550013065338,
|
| 91 |
+
"learning_rate": 4.999614302517356e-05,
|
| 92 |
+
"loss": 0.6349,
|
| 93 |
"step": 60
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.06063432835820896,
|
| 97 |
+
"grad_norm": 0.5458044409751892,
|
| 98 |
+
"learning_rate": 4.99870371535606e-05,
|
| 99 |
+
"loss": 0.6033,
|
| 100 |
"step": 65
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.06529850746268656,
|
| 104 |
+
"grad_norm": 0.45444127917289734,
|
| 105 |
+
"learning_rate": 4.997257741198456e-05,
|
| 106 |
+
"loss": 0.6164,
|
| 107 |
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.06996268656716417,
|
| 111 |
+
"grad_norm": 0.5141887068748474,
|
| 112 |
+
"learning_rate": 4.9952767243121146e-05,
|
| 113 |
+
"loss": 0.5996,
|
| 114 |
"step": 75
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.07462686567164178,
|
| 118 |
+
"grad_norm": 0.4654217064380646,
|
| 119 |
+
"learning_rate": 4.992761136351291e-05,
|
| 120 |
+
"loss": 0.6155,
|
| 121 |
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.07929104477611941,
|
| 125 |
+
"grad_norm": 0.5317748785018921,
|
| 126 |
+
"learning_rate": 4.989711576244639e-05,
|
| 127 |
+
"loss": 0.5991,
|
| 128 |
"step": 85
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.08395522388059702,
|
| 132 |
+
"grad_norm": 0.49785587191581726,
|
| 133 |
+
"learning_rate": 4.986128770052603e-05,
|
| 134 |
+
"loss": 0.6001,
|
| 135 |
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.08861940298507463,
|
| 139 |
+
"grad_norm": 0.4556206464767456,
|
| 140 |
+
"learning_rate": 4.9820135707945634e-05,
|
| 141 |
+
"loss": 0.6039,
|
| 142 |
"step": 95
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.09328358208955224,
|
| 146 |
+
"grad_norm": 0.4123242199420929,
|
| 147 |
+
"learning_rate": 4.9773669582457364e-05,
|
| 148 |
+
"loss": 0.5986,
|
| 149 |
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.09794776119402986,
|
| 153 |
+
"grad_norm": 0.5058124661445618,
|
| 154 |
+
"learning_rate": 4.972190038703905e-05,
|
| 155 |
+
"loss": 0.5939,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 156 |
"step": 105
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.10261194029850747,
|
| 160 |
+
"grad_norm": 0.4481053948402405,
|
| 161 |
+
"learning_rate": 4.966484044726024e-05,
|
| 162 |
+
"loss": 0.5974,
|
| 163 |
"step": 110
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.10727611940298508,
|
| 167 |
+
"grad_norm": 0.4899747967720032,
|
| 168 |
+
"learning_rate": 4.9602503348347625e-05,
|
| 169 |
+
"loss": 0.6127,
|
| 170 |
"step": 115
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.11194029850746269,
|
| 174 |
+
"grad_norm": 0.44825080037117004,
|
| 175 |
+
"learning_rate": 4.953490393195063e-05,
|
| 176 |
+
"loss": 0.5927,
|
| 177 |
"step": 120
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.1166044776119403,
|
| 181 |
+
"grad_norm": 0.3983883857727051,
|
| 182 |
+
"learning_rate": 4.9462058292607735e-05,
|
| 183 |
+
"loss": 0.5907,
|
| 184 |
"step": 125
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.12126865671641791,
|
| 188 |
+
"grad_norm": 0.39468082785606384,
|
| 189 |
+
"learning_rate": 4.938398377391461e-05,
|
| 190 |
+
"loss": 0.5798,
|
| 191 |
"step": 130
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.1259328358208955,
|
| 195 |
+
"grad_norm": 0.4177343547344208,
|
| 196 |
+
"learning_rate": 4.930069896439485e-05,
|
| 197 |
+
"loss": 0.5832,
|
| 198 |
"step": 135
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.13059701492537312,
|
| 202 |
+
"grad_norm": 0.3727002739906311,
|
| 203 |
+
"learning_rate": 4.921222369307427e-05,
|
| 204 |
+
"loss": 0.5839,
|
| 205 |
"step": 140
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.13526119402985073,
|
| 209 |
+
"grad_norm": 0.4837956726551056,
|
| 210 |
+
"learning_rate": 4.9118579024759854e-05,
|
| 211 |
+
"loss": 0.5715,
|
| 212 |
"step": 145
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.13992537313432835,
|
| 216 |
+
"grad_norm": 0.41040709614753723,
|
| 217 |
+
"learning_rate": 4.901978725502454e-05,
|
| 218 |
+
"loss": 0.6,
|
| 219 |
"step": 150
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.14458955223880596,
|
| 223 |
+
"grad_norm": 0.40860089659690857,
|
| 224 |
+
"learning_rate": 4.891587190489891e-05,
|
| 225 |
+
"loss": 0.5902,
|
| 226 |
"step": 155
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.14925373134328357,
|
| 230 |
+
"grad_norm": 0.45564886927604675,
|
| 231 |
+
"learning_rate": 4.880685771527114e-05,
|
| 232 |
+
"loss": 0.5737,
|
| 233 |
"step": 160
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.15391791044776118,
|
| 237 |
+
"grad_norm": 0.349163293838501,
|
| 238 |
+
"learning_rate": 4.869277064099654e-05,
|
| 239 |
+
"loss": 0.5882,
|
| 240 |
"step": 165
|
| 241 |
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.15858208955223882,
|
| 244 |
+
"grad_norm": 0.3984043002128601,
|
| 245 |
+
"learning_rate": 4.8573637844718e-05,
|
| 246 |
+
"loss": 0.5899,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.16324626865671643,
|
| 251 |
+
"grad_norm": 0.4434589147567749,
|
| 252 |
+
"learning_rate": 4.844948769039896e-05,
|
| 253 |
+
"loss": 0.592,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.16791044776119404,
|
| 258 |
+
"grad_norm": 0.4797697365283966,
|
| 259 |
+
"learning_rate": 4.83203497365703e-05,
|
| 260 |
+
"loss": 0.5755,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.17257462686567165,
|
| 265 |
+
"grad_norm": 0.4126896262168884,
|
| 266 |
+
"learning_rate": 4.818625472929286e-05,
|
| 267 |
+
"loss": 0.5794,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.17723880597014927,
|
| 272 |
+
"grad_norm": 0.3637039363384247,
|
| 273 |
+
"learning_rate": 4.8047234594837143e-05,
|
| 274 |
+
"loss": 0.5907,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.18190298507462688,
|
| 279 |
+
"grad_norm": 0.4578976631164551,
|
| 280 |
+
"learning_rate": 4.7903322432082185e-05,
|
| 281 |
+
"loss": 0.5794,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1865671641791045,
|
| 286 |
+
"grad_norm": 0.37413689494132996,
|
| 287 |
+
"learning_rate": 4.775455250463507e-05,
|
| 288 |
+
"loss": 0.5717,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1912313432835821,
|
| 293 |
+
"grad_norm": 0.35682445764541626,
|
| 294 |
+
"learning_rate": 4.760096023267322e-05,
|
| 295 |
+
"loss": 0.5606,
|
| 296 |
+
"step": 205
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1958955223880597,
|
| 300 |
+
"grad_norm": 0.37927520275115967,
|
| 301 |
+
"learning_rate": 4.744258218451135e-05,
|
| 302 |
+
"loss": 0.5622,
|
| 303 |
+
"step": 210
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.20055970149253732,
|
| 307 |
+
"grad_norm": 0.37045177817344666,
|
| 308 |
+
"learning_rate": 4.7279456067895e-05,
|
| 309 |
+
"loss": 0.5713,
|
| 310 |
+
"step": 215
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.20522388059701493,
|
| 314 |
+
"grad_norm": 0.4183138608932495,
|
| 315 |
+
"learning_rate": 4.71116207210228e-05,
|
| 316 |
+
"loss": 0.5654,
|
| 317 |
+
"step": 220
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.20988805970149255,
|
| 321 |
+
"grad_norm": 0.3651997447013855,
|
| 322 |
+
"learning_rate": 4.6939116103299655e-05,
|
| 323 |
+
"loss": 0.5666,
|
| 324 |
+
"step": 225
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.21455223880597016,
|
| 328 |
+
"grad_norm": 0.4040168821811676,
|
| 329 |
+
"learning_rate": 4.676198328582288e-05,
|
| 330 |
+
"loss": 0.5673,
|
| 331 |
+
"step": 230
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.21921641791044777,
|
| 335 |
+
"grad_norm": 0.43385499715805054,
|
| 336 |
+
"learning_rate": 4.6580264441603724e-05,
|
| 337 |
+
"loss": 0.5687,
|
| 338 |
+
"step": 235
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.22388059701492538,
|
| 342 |
+
"grad_norm": 0.4357352554798126,
|
| 343 |
+
"learning_rate": 4.6394002835526535e-05,
|
| 344 |
+
"loss": 0.5645,
|
| 345 |
+
"step": 240
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.228544776119403,
|
| 349 |
+
"grad_norm": 0.38218751549720764,
|
| 350 |
+
"learning_rate": 4.6203242814047946e-05,
|
| 351 |
+
"loss": 0.5556,
|
| 352 |
+
"step": 245
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2332089552238806,
|
| 356 |
+
"grad_norm": 0.3532406687736511,
|
| 357 |
+
"learning_rate": 4.6008029794638596e-05,
|
| 358 |
+
"loss": 0.5796,
|
| 359 |
+
"step": 250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.23787313432835822,
|
| 363 |
+
"grad_norm": 0.36546623706817627,
|
| 364 |
+
"learning_rate": 4.580841025496974e-05,
|
| 365 |
+
"loss": 0.563,
|
| 366 |
+
"step": 255
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.24253731343283583,
|
| 370 |
+
"grad_norm": 0.3052341639995575,
|
| 371 |
+
"learning_rate": 4.560443172184763e-05,
|
| 372 |
+
"loss": 0.561,
|
| 373 |
+
"step": 260
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.24720149253731344,
|
| 377 |
+
"grad_norm": 0.34654292464256287,
|
| 378 |
+
"learning_rate": 4.539614275989793e-05,
|
| 379 |
+
"loss": 0.5704,
|
| 380 |
+
"step": 265
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.251865671641791,
|
| 384 |
+
"grad_norm": 0.41322067379951477,
|
| 385 |
+
"learning_rate": 4.5183592960003104e-05,
|
| 386 |
+
"loss": 0.5633,
|
| 387 |
+
"step": 270
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.25652985074626866,
|
| 391 |
+
"grad_norm": 0.39780569076538086,
|
| 392 |
+
"learning_rate": 4.496683292749555e-05,
|
| 393 |
+
"loss": 0.5739,
|
| 394 |
+
"step": 275
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.26119402985074625,
|
| 398 |
+
"grad_norm": 0.38549038767814636,
|
| 399 |
+
"learning_rate": 4.4745914270109055e-05,
|
| 400 |
+
"loss": 0.5583,
|
| 401 |
+
"step": 280
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2658582089552239,
|
| 405 |
+
"grad_norm": 0.39039453864097595,
|
| 406 |
+
"learning_rate": 4.4520889585691705e-05,
|
| 407 |
+
"loss": 0.565,
|
| 408 |
+
"step": 285
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.27052238805970147,
|
| 412 |
+
"grad_norm": 0.3481265604496002,
|
| 413 |
+
"learning_rate": 4.429181244968301e-05,
|
| 414 |
+
"loss": 0.5558,
|
| 415 |
+
"step": 290
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2751865671641791,
|
| 419 |
+
"grad_norm": 0.34713584184646606,
|
| 420 |
+
"learning_rate": 4.4058737402358295e-05,
|
| 421 |
+
"loss": 0.5573,
|
| 422 |
+
"step": 295
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2798507462686567,
|
| 426 |
+
"grad_norm": 0.32922685146331787,
|
| 427 |
+
"learning_rate": 4.38217199358434e-05,
|
| 428 |
+
"loss": 0.5578,
|
| 429 |
+
"step": 300
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.28451492537313433,
|
| 433 |
+
"grad_norm": 0.333051472902298,
|
| 434 |
+
"learning_rate": 4.3580816480902656e-05,
|
| 435 |
+
"loss": 0.5604,
|
| 436 |
+
"step": 305
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.2891791044776119,
|
| 440 |
+
"grad_norm": 0.2980387508869171,
|
| 441 |
+
"learning_rate": 4.3336084393503545e-05,
|
| 442 |
+
"loss": 0.5515,
|
| 443 |
+
"step": 310
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.29384328358208955,
|
| 447 |
+
"grad_norm": 0.31041234731674194,
|
| 448 |
+
"learning_rate": 4.308758194116094e-05,
|
| 449 |
+
"loss": 0.5559,
|
| 450 |
+
"step": 315
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.29850746268656714,
|
| 454 |
+
"grad_norm": 0.3864513635635376,
|
| 455 |
+
"learning_rate": 4.283536828906436e-05,
|
| 456 |
+
"loss": 0.5625,
|
| 457 |
+
"step": 320
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.3031716417910448,
|
| 461 |
+
"grad_norm": 0.3100121319293976,
|
| 462 |
+
"learning_rate": 4.2579503485991567e-05,
|
| 463 |
+
"loss": 0.5433,
|
| 464 |
+
"step": 325
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.30783582089552236,
|
| 468 |
+
"grad_norm": 0.3470633327960968,
|
| 469 |
+
"learning_rate": 4.2320048450011684e-05,
|
| 470 |
+
"loss": 0.5619,
|
| 471 |
+
"step": 330
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.3125,
|
| 475 |
+
"grad_norm": 0.3305334448814392,
|
| 476 |
+
"learning_rate": 4.205706495398143e-05,
|
| 477 |
+
"loss": 0.555,
|
| 478 |
+
"step": 335
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.31716417910447764,
|
| 482 |
+
"grad_norm": 0.32060301303863525,
|
| 483 |
+
"learning_rate": 4.179061561083777e-05,
|
| 484 |
+
"loss": 0.5642,
|
| 485 |
+
"step": 340
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3218283582089552,
|
| 489 |
+
"grad_norm": 0.2954115569591522,
|
| 490 |
+
"learning_rate": 4.1520763858690644e-05,
|
| 491 |
+
"loss": 0.5454,
|
| 492 |
+
"step": 345
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.32649253731343286,
|
| 496 |
+
"grad_norm": 0.30578792095184326,
|
| 497 |
+
"learning_rate": 4.124757394571914e-05,
|
| 498 |
+
"loss": 0.5503,
|
| 499 |
+
"step": 350
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.33115671641791045,
|
| 503 |
+
"grad_norm": 0.3552355468273163,
|
| 504 |
+
"learning_rate": 4.097111091487486e-05,
|
| 505 |
+
"loss": 0.5441,
|
| 506 |
+
"step": 355
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3358208955223881,
|
| 510 |
+
"grad_norm": 0.3189811706542969,
|
| 511 |
+
"learning_rate": 4.069144058839605e-05,
|
| 512 |
+
"loss": 0.5338,
|
| 513 |
+
"step": 360
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.34048507462686567,
|
| 517 |
+
"grad_norm": 0.34771403670310974,
|
| 518 |
+
"learning_rate": 4.040862955213615e-05,
|
| 519 |
+
"loss": 0.5515,
|
| 520 |
+
"step": 365
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3451492537313433,
|
| 524 |
+
"grad_norm": 0.3125436007976532,
|
| 525 |
+
"learning_rate": 4.012274513971061e-05,
|
| 526 |
+
"loss": 0.5639,
|
| 527 |
+
"step": 370
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3498134328358209,
|
| 531 |
+
"grad_norm": 0.3404907286167145,
|
| 532 |
+
"learning_rate": 3.9833855416465624e-05,
|
| 533 |
+
"loss": 0.5442,
|
| 534 |
+
"step": 375
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.35447761194029853,
|
| 538 |
+
"grad_norm": 0.3328476548194885,
|
| 539 |
+
"learning_rate": 3.954202916327264e-05,
|
| 540 |
+
"loss": 0.5501,
|
| 541 |
+
"step": 380
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3591417910447761,
|
| 545 |
+
"grad_norm": 0.3313455581665039,
|
| 546 |
+
"learning_rate": 3.924733586015257e-05,
|
| 547 |
+
"loss": 0.5389,
|
| 548 |
+
"step": 385
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.36380597014925375,
|
| 552 |
+
"grad_norm": 0.30594876408576965,
|
| 553 |
+
"learning_rate": 3.894984566973346e-05,
|
| 554 |
+
"loss": 0.5541,
|
| 555 |
+
"step": 390
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.36847014925373134,
|
| 559 |
+
"grad_norm": 0.309085875749588,
|
| 560 |
+
"learning_rate": 3.864962942054572e-05,
|
| 561 |
+
"loss": 0.5479,
|
| 562 |
+
"step": 395
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.373134328358209,
|
| 566 |
+
"grad_norm": 0.3359319269657135,
|
| 567 |
+
"learning_rate": 3.834675859015876e-05,
|
| 568 |
+
"loss": 0.5558,
|
| 569 |
+
"step": 400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.37779850746268656,
|
| 573 |
+
"grad_norm": 0.3179395794868469,
|
| 574 |
+
"learning_rate": 3.804130528816312e-05,
|
| 575 |
+
"loss": 0.5611,
|
| 576 |
+
"step": 405
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3824626865671642,
|
| 580 |
+
"grad_norm": 0.338645875453949,
|
| 581 |
+
"learning_rate": 3.77333422390021e-05,
|
| 582 |
+
"loss": 0.5427,
|
| 583 |
+
"step": 410
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3871268656716418,
|
| 587 |
+
"grad_norm": 0.35845276713371277,
|
| 588 |
+
"learning_rate": 3.7422942764657054e-05,
|
| 589 |
+
"loss": 0.5453,
|
| 590 |
+
"step": 415
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3917910447761194,
|
| 594 |
+
"grad_norm": 0.2944061756134033,
|
| 595 |
+
"learning_rate": 3.711018076719034e-05,
|
| 596 |
+
"loss": 0.5395,
|
| 597 |
+
"step": 420
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.396455223880597,
|
| 601 |
+
"grad_norm": 0.36072415113449097,
|
| 602 |
+
"learning_rate": 3.679513071115025e-05,
|
| 603 |
+
"loss": 0.5355,
|
| 604 |
+
"step": 425
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.40111940298507465,
|
| 608 |
+
"grad_norm": 0.3697458803653717,
|
| 609 |
+
"learning_rate": 3.647786760584194e-05,
|
| 610 |
+
"loss": 0.5407,
|
| 611 |
+
"step": 430
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.40578358208955223,
|
| 615 |
+
"grad_norm": 0.3033182919025421,
|
| 616 |
+
"learning_rate": 3.615846698746869e-05,
|
| 617 |
+
"loss": 0.5456,
|
| 618 |
+
"step": 435
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.41044776119402987,
|
| 622 |
+
"grad_norm": 0.2976251244544983,
|
| 623 |
+
"learning_rate": 3.583700490114776e-05,
|
| 624 |
+
"loss": 0.5473,
|
| 625 |
+
"step": 440
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.41511194029850745,
|
| 629 |
+
"grad_norm": 0.32828009128570557,
|
| 630 |
+
"learning_rate": 3.5513557882805e-05,
|
| 631 |
+
"loss": 0.5488,
|
| 632 |
+
"step": 445
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4197761194029851,
|
| 636 |
+
"grad_norm": 0.29798197746276855,
|
| 637 |
+
"learning_rate": 3.518820294095267e-05,
|
| 638 |
+
"loss": 0.5257,
|
| 639 |
+
"step": 450
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.4244402985074627,
|
| 643 |
+
"grad_norm": 0.3206273317337036,
|
| 644 |
+
"learning_rate": 3.486101753835468e-05,
|
| 645 |
+
"loss": 0.5481,
|
| 646 |
+
"step": 455
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4291044776119403,
|
| 650 |
+
"grad_norm": 0.30574238300323486,
|
| 651 |
+
"learning_rate": 3.453207957358377e-05,
|
| 652 |
+
"loss": 0.5437,
|
| 653 |
+
"step": 460
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4337686567164179,
|
| 657 |
+
"grad_norm": 0.3001498878002167,
|
| 658 |
+
"learning_rate": 3.420146736247487e-05,
|
| 659 |
+
"loss": 0.5539,
|
| 660 |
+
"step": 465
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.43843283582089554,
|
| 664 |
+
"grad_norm": 0.33013975620269775,
|
| 665 |
+
"learning_rate": 3.386925961947906e-05,
|
| 666 |
+
"loss": 0.5405,
|
| 667 |
+
"step": 470
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.4430970149253731,
|
| 671 |
+
"grad_norm": 0.2987500727176666,
|
| 672 |
+
"learning_rate": 3.353553543892277e-05,
|
| 673 |
+
"loss": 0.5267,
|
| 674 |
+
"step": 475
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.44776119402985076,
|
| 678 |
+
"grad_norm": 0.292449951171875,
|
| 679 |
+
"learning_rate": 3.320037427617639e-05,
|
| 680 |
+
"loss": 0.5483,
|
| 681 |
+
"step": 480
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.45242537313432835,
|
| 685 |
+
"grad_norm": 0.3454206585884094,
|
| 686 |
+
"learning_rate": 3.2863855928737026e-05,
|
| 687 |
+
"loss": 0.5369,
|
| 688 |
+
"step": 485
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.457089552238806,
|
| 692 |
+
"grad_norm": 0.3288365304470062,
|
| 693 |
+
"learning_rate": 3.252606051722972e-05,
|
| 694 |
+
"loss": 0.5547,
|
| 695 |
+
"step": 490
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.46175373134328357,
|
| 699 |
+
"grad_norm": 0.29698145389556885,
|
| 700 |
+
"learning_rate": 3.218706846633183e-05,
|
| 701 |
+
"loss": 0.5277,
|
| 702 |
+
"step": 495
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4664179104477612,
|
| 706 |
+
"grad_norm": 0.2867559790611267,
|
| 707 |
+
"learning_rate": 3.1846960485624886e-05,
|
| 708 |
+
"loss": 0.5431,
|
| 709 |
+
"step": 500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4710820895522388,
|
| 713 |
+
"grad_norm": 0.2875625789165497,
|
| 714 |
+
"learning_rate": 3.150581755037877e-05,
|
| 715 |
+
"loss": 0.5293,
|
| 716 |
+
"step": 505
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.47574626865671643,
|
| 720 |
+
"grad_norm": 0.3133365511894226,
|
| 721 |
+
"learning_rate": 3.1163720882272516e-05,
|
| 722 |
+
"loss": 0.544,
|
| 723 |
+
"step": 510
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.480410447761194,
|
| 727 |
+
"grad_norm": 0.31600460410118103,
|
| 728 |
+
"learning_rate": 3.08207519300565e-05,
|
| 729 |
+
"loss": 0.5527,
|
| 730 |
+
"step": 515
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.48507462686567165,
|
| 734 |
+
"grad_norm": 0.3297214210033417,
|
| 735 |
+
"learning_rate": 3.047699235016056e-05,
|
| 736 |
+
"loss": 0.5476,
|
| 737 |
+
"step": 520
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.48973880597014924,
|
| 741 |
+
"grad_norm": 0.3018137812614441,
|
| 742 |
+
"learning_rate": 3.0132523987252658e-05,
|
| 743 |
+
"loss": 0.5303,
|
| 744 |
+
"step": 525
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.4944029850746269,
|
| 748 |
+
"grad_norm": 0.32776540517807007,
|
| 749 |
+
"learning_rate": 2.9787428854752736e-05,
|
| 750 |
+
"loss": 0.5346,
|
| 751 |
+
"step": 530
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.49906716417910446,
|
| 755 |
+
"grad_norm": 0.3238219618797302,
|
| 756 |
+
"learning_rate": 2.9441789115306402e-05,
|
| 757 |
+
"loss": 0.5257,
|
| 758 |
+
"step": 535
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.503731343283582,
|
| 762 |
+
"grad_norm": 0.38174256682395935,
|
| 763 |
+
"learning_rate": 2.9095687061223058e-05,
|
| 764 |
+
"loss": 0.5345,
|
| 765 |
+
"step": 540
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5083955223880597,
|
| 769 |
+
"grad_norm": 0.2929813265800476,
|
| 770 |
+
"learning_rate": 2.874920509488319e-05,
|
| 771 |
+
"loss": 0.5337,
|
| 772 |
+
"step": 545
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5130597014925373,
|
| 776 |
+
"grad_norm": 0.2884758710861206,
|
| 777 |
+
"learning_rate": 2.8402425709119435e-05,
|
| 778 |
+
"loss": 0.5286,
|
| 779 |
+
"step": 550
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5177238805970149,
|
| 783 |
+
"grad_norm": 0.3295556902885437,
|
| 784 |
+
"learning_rate": 2.8055431467576106e-05,
|
| 785 |
+
"loss": 0.5405,
|
| 786 |
+
"step": 555
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5223880597014925,
|
| 790 |
+
"grad_norm": 0.28779491782188416,
|
| 791 |
+
"learning_rate": 2.7708304985051868e-05,
|
| 792 |
+
"loss": 0.5321,
|
| 793 |
+
"step": 560
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5270522388059702,
|
| 797 |
+
"grad_norm": 0.2856288254261017,
|
| 798 |
+
"learning_rate": 2.7361128907830253e-05,
|
| 799 |
+
"loss": 0.5345,
|
| 800 |
+
"step": 565
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5317164179104478,
|
| 804 |
+
"grad_norm": 0.2876598536968231,
|
| 805 |
+
"learning_rate": 2.7013985894002623e-05,
|
| 806 |
+
"loss": 0.5247,
|
| 807 |
+
"step": 570
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.5363805970149254,
|
| 811 |
+
"grad_norm": 0.30823373794555664,
|
| 812 |
+
"learning_rate": 2.6666958593788405e-05,
|
| 813 |
+
"loss": 0.5263,
|
| 814 |
+
"step": 575
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.5410447761194029,
|
| 818 |
+
"grad_norm": 0.2999833822250366,
|
| 819 |
+
"learning_rate": 2.6320129629857093e-05,
|
| 820 |
+
"loss": 0.525,
|
| 821 |
+
"step": 580
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5457089552238806,
|
| 825 |
+
"grad_norm": 0.2924771010875702,
|
| 826 |
+
"learning_rate": 2.597358157765692e-05,
|
| 827 |
+
"loss": 0.5275,
|
| 828 |
+
"step": 585
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5503731343283582,
|
| 832 |
+
"grad_norm": 0.2781645357608795,
|
| 833 |
+
"learning_rate": 2.56273969457547e-05,
|
| 834 |
+
"loss": 0.5307,
|
| 835 |
+
"step": 590
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5550373134328358,
|
| 839 |
+
"grad_norm": 0.31060436367988586,
|
| 840 |
+
"learning_rate": 2.528165815619162e-05,
|
| 841 |
+
"loss": 0.525,
|
| 842 |
+
"step": 595
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5597014925373134,
|
| 846 |
+
"grad_norm": 0.2785758674144745,
|
| 847 |
+
"learning_rate": 2.4936447524859625e-05,
|
| 848 |
+
"loss": 0.5414,
|
| 849 |
+
"step": 600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5643656716417911,
|
| 853 |
+
"grad_norm": 0.2810611426830292,
|
| 854 |
+
"learning_rate": 2.459184724190308e-05,
|
| 855 |
+
"loss": 0.5305,
|
| 856 |
+
"step": 605
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5690298507462687,
|
| 860 |
+
"grad_norm": 0.310566246509552,
|
| 861 |
+
"learning_rate": 2.4247939352150386e-05,
|
| 862 |
+
"loss": 0.5308,
|
| 863 |
+
"step": 610
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5736940298507462,
|
| 867 |
+
"grad_norm": 0.2734331786632538,
|
| 868 |
+
"learning_rate": 2.390480573558012e-05,
|
| 869 |
+
"loss": 0.5191,
|
| 870 |
+
"step": 615
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5783582089552238,
|
| 874 |
+
"grad_norm": 0.28134024143218994,
|
| 875 |
+
"learning_rate": 2.3562528087826573e-05,
|
| 876 |
+
"loss": 0.5218,
|
| 877 |
+
"step": 620
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5830223880597015,
|
| 881 |
+
"grad_norm": 0.28448307514190674,
|
| 882 |
+
"learning_rate": 2.3221187900729003e-05,
|
| 883 |
+
"loss": 0.5245,
|
| 884 |
+
"step": 625
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5876865671641791,
|
| 888 |
+
"grad_norm": 0.2660384476184845,
|
| 889 |
+
"learning_rate": 2.2880866442929544e-05,
|
| 890 |
+
"loss": 0.5387,
|
| 891 |
+
"step": 630
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5923507462686567,
|
| 895 |
+
"grad_norm": 0.28395891189575195,
|
| 896 |
+
"learning_rate": 2.254164474052416e-05,
|
| 897 |
+
"loss": 0.5342,
|
| 898 |
+
"step": 635
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5970149253731343,
|
| 902 |
+
"grad_norm": 0.2656669616699219,
|
| 903 |
+
"learning_rate": 2.2203603557771447e-05,
|
| 904 |
+
"loss": 0.5337,
|
| 905 |
+
"step": 640
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.601679104477612,
|
| 909 |
+
"grad_norm": 0.2691911458969116,
|
| 910 |
+
"learning_rate": 2.186682337786365e-05,
|
| 911 |
+
"loss": 0.5273,
|
| 912 |
+
"step": 645
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6063432835820896,
|
| 916 |
+
"grad_norm": 0.2647033929824829,
|
| 917 |
+
"learning_rate": 2.153138438376473e-05,
|
| 918 |
+
"loss": 0.5311,
|
| 919 |
+
"step": 650
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6110074626865671,
|
| 923 |
+
"grad_norm": 0.2803230881690979,
|
| 924 |
+
"learning_rate": 2.119736643911979e-05,
|
| 925 |
+
"loss": 0.5196,
|
| 926 |
+
"step": 655
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6156716417910447,
|
| 930 |
+
"grad_norm": 0.2749485373497009,
|
| 931 |
+
"learning_rate": 2.0864849069240645e-05,
|
| 932 |
+
"loss": 0.5412,
|
| 933 |
+
"step": 660
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6203358208955224,
|
| 937 |
+
"grad_norm": 0.27285492420196533,
|
| 938 |
+
"learning_rate": 2.0533911442171805e-05,
|
| 939 |
+
"loss": 0.523,
|
| 940 |
+
"step": 665
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.625,
|
| 944 |
+
"grad_norm": 0.2703399956226349,
|
| 945 |
+
"learning_rate": 2.0204632349841667e-05,
|
| 946 |
+
"loss": 0.5156,
|
| 947 |
+
"step": 670
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.6296641791044776,
|
| 951 |
+
"grad_norm": 0.3498228192329407,
|
| 952 |
+
"learning_rate": 1.9877090189303182e-05,
|
| 953 |
+
"loss": 0.5282,
|
| 954 |
+
"step": 675
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.6343283582089553,
|
| 958 |
+
"grad_norm": 0.3206793963909149,
|
| 959 |
+
"learning_rate": 1.9551362944068462e-05,
|
| 960 |
+
"loss": 0.5193,
|
| 961 |
+
"step": 680
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.6389925373134329,
|
| 965 |
+
"grad_norm": 0.28730154037475586,
|
| 966 |
+
"learning_rate": 1.922752816554204e-05,
|
| 967 |
+
"loss": 0.5345,
|
| 968 |
+
"step": 685
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.6436567164179104,
|
| 972 |
+
"grad_norm": 0.25482338666915894,
|
| 973 |
+
"learning_rate": 1.890566295455678e-05,
|
| 974 |
+
"loss": 0.5324,
|
| 975 |
+
"step": 690
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.648320895522388,
|
| 979 |
+
"grad_norm": 0.2529654800891876,
|
| 980 |
+
"learning_rate": 1.858584394301728e-05,
|
| 981 |
+
"loss": 0.5175,
|
| 982 |
+
"step": 695
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.6529850746268657,
|
| 986 |
+
"grad_norm": 0.2732594907283783,
|
| 987 |
+
"learning_rate": 1.8268147275654707e-05,
|
| 988 |
+
"loss": 0.5206,
|
| 989 |
+
"step": 700
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6576492537313433,
|
| 993 |
+
"grad_norm": 0.26254940032958984,
|
| 994 |
+
"learning_rate": 1.7952648591897858e-05,
|
| 995 |
+
"loss": 0.5156,
|
| 996 |
+
"step": 705
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.6623134328358209,
|
| 1000 |
+
"grad_norm": 0.2806949019432068,
|
| 1001 |
+
"learning_rate": 1.7639423007864252e-05,
|
| 1002 |
+
"loss": 0.5325,
|
| 1003 |
+
"step": 710
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6669776119402985,
|
| 1007 |
+
"grad_norm": 0.2589339315891266,
|
| 1008 |
+
"learning_rate": 1.7328545098476106e-05,
|
| 1009 |
+
"loss": 0.5264,
|
| 1010 |
+
"step": 715
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6716417910447762,
|
| 1014 |
+
"grad_norm": 0.2677186131477356,
|
| 1015 |
+
"learning_rate": 1.702008887970491e-05,
|
| 1016 |
+
"loss": 0.5286,
|
| 1017 |
+
"step": 720
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6763059701492538,
|
| 1021 |
+
"grad_norm": 0.25420376658439636,
|
| 1022 |
+
"learning_rate": 1.671412779094926e-05,
|
| 1023 |
+
"loss": 0.5164,
|
| 1024 |
+
"step": 725
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6809701492537313,
|
| 1028 |
+
"grad_norm": 0.2698172330856323,
|
| 1029 |
+
"learning_rate": 1.6410734677549872e-05,
|
| 1030 |
+
"loss": 0.5332,
|
| 1031 |
+
"step": 730
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6856343283582089,
|
| 1035 |
+
"grad_norm": 0.26080092787742615,
|
| 1036 |
+
"learning_rate": 1.6109981773446036e-05,
|
| 1037 |
+
"loss": 0.5315,
|
| 1038 |
+
"step": 735
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6902985074626866,
|
| 1042 |
+
"grad_norm": 0.24189084768295288,
|
| 1043 |
+
"learning_rate": 1.58119406839777e-05,
|
| 1044 |
+
"loss": 0.5102,
|
| 1045 |
+
"step": 740
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6949626865671642,
|
| 1049 |
+
"grad_norm": 0.24392467737197876,
|
| 1050 |
+
"learning_rate": 1.5516682368837133e-05,
|
| 1051 |
+
"loss": 0.5104,
|
| 1052 |
+
"step": 745
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6996268656716418,
|
| 1056 |
+
"grad_norm": 0.23985299468040466,
|
| 1057 |
+
"learning_rate": 1.5224277125174388e-05,
|
| 1058 |
+
"loss": 0.5193,
|
| 1059 |
+
"step": 750
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7042910447761194,
|
| 1063 |
+
"grad_norm": 0.2562742829322815,
|
| 1064 |
+
"learning_rate": 1.4934794570860416e-05,
|
| 1065 |
+
"loss": 0.5218,
|
| 1066 |
+
"step": 755
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7089552238805971,
|
| 1070 |
+
"grad_norm": 0.2763282060623169,
|
| 1071 |
+
"learning_rate": 1.464830362791204e-05,
|
| 1072 |
+
"loss": 0.5296,
|
| 1073 |
+
"step": 760
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7136194029850746,
|
| 1077 |
+
"grad_norm": 0.24829770624637604,
|
| 1078 |
+
"learning_rate": 1.4364872506082425e-05,
|
| 1079 |
+
"loss": 0.5325,
|
| 1080 |
+
"step": 765
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.7182835820895522,
|
| 1084 |
+
"grad_norm": 0.256150484085083,
|
| 1085 |
+
"learning_rate": 1.4084568686621314e-05,
|
| 1086 |
+
"loss": 0.5161,
|
| 1087 |
+
"step": 770
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.7229477611940298,
|
| 1091 |
+
"grad_norm": 0.2693561017513275,
|
| 1092 |
+
"learning_rate": 1.3807458906208546e-05,
|
| 1093 |
+
"loss": 0.5315,
|
| 1094 |
+
"step": 775
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.7276119402985075,
|
| 1098 |
+
"grad_norm": 0.2647959887981415,
|
| 1099 |
+
"learning_rate": 1.3533609141065008e-05,
|
| 1100 |
+
"loss": 0.5149,
|
| 1101 |
+
"step": 780
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.7322761194029851,
|
| 1105 |
+
"grad_norm": 0.24539123475551605,
|
| 1106 |
+
"learning_rate": 1.326308459124447e-05,
|
| 1107 |
+
"loss": 0.5156,
|
| 1108 |
+
"step": 785
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.7369402985074627,
|
| 1112 |
+
"grad_norm": 0.2749805152416229,
|
| 1113 |
+
"learning_rate": 1.299594966511038e-05,
|
| 1114 |
+
"loss": 0.5323,
|
| 1115 |
+
"step": 790
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.7416044776119403,
|
| 1119 |
+
"grad_norm": 0.2615833580493927,
|
| 1120 |
+
"learning_rate": 1.2732267964001033e-05,
|
| 1121 |
+
"loss": 0.5284,
|
| 1122 |
+
"step": 795
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.746268656716418,
|
| 1126 |
+
"grad_norm": 0.25320667028427124,
|
| 1127 |
+
"learning_rate": 1.2472102267086904e-05,
|
| 1128 |
+
"loss": 0.5185,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.7509328358208955,
|
| 1133 |
+
"grad_norm": 0.261538028717041,
|
| 1134 |
+
"learning_rate": 1.2215514516423813e-05,
|
| 1135 |
+
"loss": 0.5222,
|
| 1136 |
+
"step": 805
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.7555970149253731,
|
| 1140 |
+
"grad_norm": 0.25271981954574585,
|
| 1141 |
+
"learning_rate": 1.1962565802205255e-05,
|
| 1142 |
+
"loss": 0.5202,
|
| 1143 |
+
"step": 810
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.7602611940298507,
|
| 1147 |
+
"grad_norm": 0.24026913940906525,
|
| 1148 |
+
"learning_rate": 1.1713316348217673e-05,
|
| 1149 |
+
"loss": 0.5095,
|
| 1150 |
+
"step": 815
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.7649253731343284,
|
| 1154 |
+
"grad_norm": 0.24491222202777863,
|
| 1155 |
+
"learning_rate": 1.1467825497501954e-05,
|
| 1156 |
+
"loss": 0.5137,
|
| 1157 |
+
"step": 820
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.769589552238806,
|
| 1161 |
+
"grad_norm": 0.2599160969257355,
|
| 1162 |
+
"learning_rate": 1.1226151698224597e-05,
|
| 1163 |
+
"loss": 0.5271,
|
| 1164 |
+
"step": 825
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.7742537313432836,
|
| 1168 |
+
"grad_norm": 0.2577415108680725,
|
| 1169 |
+
"learning_rate": 1.0988352489762006e-05,
|
| 1170 |
+
"loss": 0.526,
|
| 1171 |
+
"step": 830
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.7789179104477612,
|
| 1175 |
+
"grad_norm": 0.23763109743595123,
|
| 1176 |
+
"learning_rate": 1.0754484489001085e-05,
|
| 1177 |
+
"loss": 0.5025,
|
| 1178 |
+
"step": 835
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.7835820895522388,
|
| 1182 |
+
"grad_norm": 0.24150507152080536,
|
| 1183 |
+
"learning_rate": 1.052460337685951e-05,
|
| 1184 |
+
"loss": 0.518,
|
| 1185 |
+
"step": 840
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.7882462686567164,
|
| 1189 |
+
"grad_norm": 0.2546503245830536,
|
| 1190 |
+
"learning_rate": 1.0298763885028839e-05,
|
| 1191 |
+
"loss": 0.521,
|
| 1192 |
+
"step": 845
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.792910447761194,
|
| 1196 |
+
"grad_norm": 0.2274450659751892,
|
| 1197 |
+
"learning_rate": 1.0077019782943584e-05,
|
| 1198 |
+
"loss": 0.5169,
|
| 1199 |
+
"step": 850
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7975746268656716,
|
| 1203 |
+
"grad_norm": 0.23724405467510223,
|
| 1204 |
+
"learning_rate": 9.859423864979441e-06,
|
| 1205 |
+
"loss": 0.5298,
|
| 1206 |
+
"step": 855
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.8022388059701493,
|
| 1210 |
+
"grad_norm": 0.24516107141971588,
|
| 1211 |
+
"learning_rate": 9.646027937883622e-06,
|
| 1212 |
+
"loss": 0.5117,
|
| 1213 |
+
"step": 860
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8069029850746269,
|
| 1217 |
+
"grad_norm": 0.2544306516647339,
|
| 1218 |
+
"learning_rate": 9.436882808440334e-06,
|
| 1219 |
+
"loss": 0.5131,
|
| 1220 |
+
"step": 865
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.8115671641791045,
|
| 1224 |
+
"grad_norm": 0.25421103835105896,
|
| 1225 |
+
"learning_rate": 9.232038271374377e-06,
|
| 1226 |
+
"loss": 0.5168,
|
| 1227 |
+
"step": 870
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.816231343283582,
|
| 1231 |
+
"grad_norm": 0.24098147451877594,
|
| 1232 |
+
"learning_rate": 9.031543097495638e-06,
|
| 1233 |
+
"loss": 0.5127,
|
| 1234 |
+
"step": 875
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.8208955223880597,
|
| 1238 |
+
"grad_norm": 0.23913885653018951,
|
| 1239 |
+
"learning_rate": 8.835445022087426e-06,
|
| 1240 |
+
"loss": 0.5214,
|
| 1241 |
+
"step": 880
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.8255597014925373,
|
| 1245 |
+
"grad_norm": 0.24211934208869934,
|
| 1246 |
+
"learning_rate": 8.6437907335413e-06,
|
| 1247 |
+
"loss": 0.5254,
|
| 1248 |
+
"step": 885
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.8302238805970149,
|
| 1252 |
+
"grad_norm": 0.23977182805538177,
|
| 1253 |
+
"learning_rate": 8.456625862241193e-06,
|
| 1254 |
+
"loss": 0.5241,
|
| 1255 |
+
"step": 890
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.8348880597014925,
|
| 1259 |
+
"grad_norm": 0.23448112607002258,
|
| 1260 |
+
"learning_rate": 8.273994969699394e-06,
|
| 1261 |
+
"loss": 0.5127,
|
| 1262 |
+
"step": 895
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.8395522388059702,
|
| 1266 |
+
"grad_norm": 0.2343115657567978,
|
| 1267 |
+
"learning_rate": 8.095941537947057e-06,
|
| 1268 |
+
"loss": 0.5135,
|
| 1269 |
+
"step": 900
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.8442164179104478,
|
| 1273 |
+
"grad_norm": 0.23262952268123627,
|
| 1274 |
+
"learning_rate": 7.922507959181673e-06,
|
| 1275 |
+
"loss": 0.5342,
|
| 1276 |
+
"step": 905
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.8488805970149254,
|
| 1280 |
+
"grad_norm": 0.24626874923706055,
|
| 1281 |
+
"learning_rate": 7.753735525674059e-06,
|
| 1282 |
+
"loss": 0.5305,
|
| 1283 |
+
"step": 910
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.8535447761194029,
|
| 1287 |
+
"grad_norm": 0.24683545529842377,
|
| 1288 |
+
"learning_rate": 7.58966441993719e-06,
|
| 1289 |
+
"loss": 0.5297,
|
| 1290 |
+
"step": 915
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.8582089552238806,
|
| 1294 |
+
"grad_norm": 0.23083548247814178,
|
| 1295 |
+
"learning_rate": 7.430333705159286e-06,
|
| 1296 |
+
"loss": 0.5243,
|
| 1297 |
+
"step": 920
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.8628731343283582,
|
| 1301 |
+
"grad_norm": 0.24129389226436615,
|
| 1302 |
+
"learning_rate": 7.275781315903374e-06,
|
| 1303 |
+
"loss": 0.5109,
|
| 1304 |
+
"step": 925
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.8675373134328358,
|
| 1308 |
+
"grad_norm": 0.25144246220588684,
|
| 1309 |
+
"learning_rate": 7.126044049075548e-06,
|
| 1310 |
+
"loss": 0.5152,
|
| 1311 |
+
"step": 930
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.8722014925373134,
|
| 1315 |
+
"grad_norm": 0.24448096752166748,
|
| 1316 |
+
"learning_rate": 6.9811575551641224e-06,
|
| 1317 |
+
"loss": 0.5037,
|
| 1318 |
+
"step": 935
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.8768656716417911,
|
| 1322 |
+
"grad_norm": 0.23240606486797333,
|
| 1323 |
+
"learning_rate": 6.8411563297516995e-06,
|
| 1324 |
+
"loss": 0.5062,
|
| 1325 |
+
"step": 940
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.8815298507462687,
|
| 1329 |
+
"grad_norm": 0.2471693903207779,
|
| 1330 |
+
"learning_rate": 6.706073705302254e-06,
|
| 1331 |
+
"loss": 0.516,
|
| 1332 |
+
"step": 945
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.8861940298507462,
|
| 1336 |
+
"grad_norm": 0.25190070271492004,
|
| 1337 |
+
"learning_rate": 6.575941843225068e-06,
|
| 1338 |
+
"loss": 0.522,
|
| 1339 |
+
"step": 950
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.8908582089552238,
|
| 1343 |
+
"grad_norm": 0.25332111120224,
|
| 1344 |
+
"learning_rate": 6.450791726217538e-06,
|
| 1345 |
+
"loss": 0.5173,
|
| 1346 |
+
"step": 955
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.8955223880597015,
|
| 1350 |
+
"grad_norm": 0.23512767255306244,
|
| 1351 |
+
"learning_rate": 6.330653150888617e-06,
|
| 1352 |
+
"loss": 0.5172,
|
| 1353 |
+
"step": 960
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.9001865671641791,
|
| 1357 |
+
"grad_norm": 0.2349744737148285,
|
| 1358 |
+
"learning_rate": 6.215554720664598e-06,
|
| 1359 |
+
"loss": 0.5134,
|
| 1360 |
+
"step": 965
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.9048507462686567,
|
| 1364 |
+
"grad_norm": 0.2327372282743454,
|
| 1365 |
+
"learning_rate": 6.105523838979022e-06,
|
| 1366 |
+
"loss": 0.5214,
|
| 1367 |
+
"step": 970
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.9095149253731343,
|
| 1371 |
+
"grad_norm": 0.23093824088573456,
|
| 1372 |
+
"learning_rate": 6.000586702748301e-06,
|
| 1373 |
+
"loss": 0.5011,
|
| 1374 |
+
"step": 975
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.914179104477612,
|
| 1378 |
+
"grad_norm": 0.2209603190422058,
|
| 1379 |
+
"learning_rate": 5.900768296134551e-06,
|
| 1380 |
+
"loss": 0.5048,
|
| 1381 |
+
"step": 980
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.9188432835820896,
|
| 1385 |
+
"grad_norm": 0.2304598093032837,
|
| 1386 |
+
"learning_rate": 5.8060923845971825e-06,
|
| 1387 |
+
"loss": 0.5192,
|
| 1388 |
+
"step": 985
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.9235074626865671,
|
| 1392 |
+
"grad_norm": 0.23641374707221985,
|
| 1393 |
+
"learning_rate": 5.7165815092346825e-06,
|
| 1394 |
+
"loss": 0.5014,
|
| 1395 |
+
"step": 990
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.9281716417910447,
|
| 1399 |
+
"grad_norm": 0.25322774052619934,
|
| 1400 |
+
"learning_rate": 5.632256981417845e-06,
|
| 1401 |
+
"loss": 0.5247,
|
| 1402 |
+
"step": 995
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.9328358208955224,
|
| 1406 |
+
"grad_norm": 0.226620152592659,
|
| 1407 |
+
"learning_rate": 5.553138877715833e-06,
|
| 1408 |
+
"loss": 0.5086,
|
| 1409 |
+
"step": 1000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.9375,
|
| 1413 |
+
"grad_norm": 0.23227237164974213,
|
| 1414 |
+
"learning_rate": 5.479246035116201e-06,
|
| 1415 |
+
"loss": 0.5078,
|
| 1416 |
+
"step": 1005
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.9421641791044776,
|
| 1420 |
+
"grad_norm": 0.23838132619857788,
|
| 1421 |
+
"learning_rate": 5.410596046540051e-06,
|
| 1422 |
+
"loss": 0.5263,
|
| 1423 |
+
"step": 1010
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.9468283582089553,
|
| 1427 |
+
"grad_norm": 0.2337532490491867,
|
| 1428 |
+
"learning_rate": 5.347205256653387e-06,
|
| 1429 |
+
"loss": 0.5168,
|
| 1430 |
+
"step": 1015
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.9514925373134329,
|
| 1434 |
+
"grad_norm": 0.24512432515621185,
|
| 1435 |
+
"learning_rate": 5.28908875797568e-06,
|
| 1436 |
+
"loss": 0.5092,
|
| 1437 |
+
"step": 1020
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.9561567164179104,
|
| 1441 |
+
"grad_norm": 0.24788177013397217,
|
| 1442 |
+
"learning_rate": 5.236260387286509e-06,
|
| 1443 |
+
"loss": 0.5251,
|
| 1444 |
+
"step": 1025
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.960820895522388,
|
| 1448 |
+
"grad_norm": 0.24356013536453247,
|
| 1449 |
+
"learning_rate": 5.1887327223312296e-06,
|
| 1450 |
+
"loss": 0.5109,
|
| 1451 |
+
"step": 1030
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.9654850746268657,
|
| 1455 |
+
"grad_norm": 0.2298063486814499,
|
| 1456 |
+
"learning_rate": 5.1465170788263595e-06,
|
| 1457 |
+
"loss": 0.5288,
|
| 1458 |
+
"step": 1035
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.9701492537313433,
|
| 1462 |
+
"grad_norm": 0.21959343552589417,
|
| 1463 |
+
"learning_rate": 5.109623507765466e-06,
|
| 1464 |
+
"loss": 0.506,
|
| 1465 |
+
"step": 1040
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.9748134328358209,
|
| 1469 |
+
"grad_norm": 0.22912734746932983,
|
| 1470 |
+
"learning_rate": 5.07806079302615e-06,
|
| 1471 |
+
"loss": 0.508,
|
| 1472 |
+
"step": 1045
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.9794776119402985,
|
| 1476 |
+
"grad_norm": 0.2325100600719452,
|
| 1477 |
+
"learning_rate": 5.051836449278715e-06,
|
| 1478 |
+
"loss": 0.5001,
|
| 1479 |
+
"step": 1050
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.9841417910447762,
|
| 1483 |
+
"grad_norm": 0.2450973242521286,
|
| 1484 |
+
"learning_rate": 5.030956720197035e-06,
|
| 1485 |
+
"loss": 0.5107,
|
| 1486 |
+
"step": 1055
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.9888059701492538,
|
| 1490 |
+
"grad_norm": 0.22343529760837555,
|
| 1491 |
+
"learning_rate": 5.015426576972003e-06,
|
| 1492 |
+
"loss": 0.509,
|
| 1493 |
+
"step": 1060
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.9934701492537313,
|
| 1497 |
+
"grad_norm": 0.23454374074935913,
|
| 1498 |
+
"learning_rate": 5.005249717127964e-06,
|
| 1499 |
+
"loss": 0.5137,
|
| 1500 |
+
"step": 1065
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.9981343283582089,
|
| 1504 |
+
"grad_norm": 0.23174691200256348,
|
| 1505 |
+
"learning_rate": 5.000428563642382e-06,
|
| 1506 |
+
"loss": 0.5,
|
| 1507 |
+
"step": 1070
|
| 1508 |
+
},
|
| 1509 |
{
|
| 1510 |
"epoch": 1.0,
|
| 1511 |
+
"step": 1072,
|
| 1512 |
+
"total_flos": 4.418757235321078e+18,
|
| 1513 |
+
"train_loss": 0.5497396646150902,
|
| 1514 |
+
"train_runtime": 11730.1118,
|
| 1515 |
+
"train_samples_per_second": 2.924,
|
| 1516 |
+
"train_steps_per_second": 0.091
|
| 1517 |
}
|
| 1518 |
],
|
| 1519 |
"logging_steps": 5,
|
| 1520 |
+
"max_steps": 1072,
|
| 1521 |
"num_input_tokens_seen": 0,
|
| 1522 |
"num_train_epochs": 1,
|
| 1523 |
+
"save_steps": 100,
|
| 1524 |
"stateful_callbacks": {
|
| 1525 |
"TrainerControl": {
|
| 1526 |
"args": {
|
| 1527 |
"should_epoch_stop": false,
|
| 1528 |
"should_evaluate": false,
|
| 1529 |
"should_log": false,
|
| 1530 |
+
"should_save": true,
|
| 1531 |
+
"should_training_stop": true
|
| 1532 |
},
|
| 1533 |
"attributes": {}
|
| 1534 |
}
|
| 1535 |
},
|
| 1536 |
+
"total_flos": 4.418757235321078e+18,
|
| 1537 |
+
"train_batch_size": 16,
|
| 1538 |
"trial_name": null,
|
| 1539 |
"trial_params": null
|
| 1540 |
}
|