End of training
Browse files- README.md +2 -1
- all_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +1012 -130
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: training_baseline
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# training_baseline
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: training_baseline
|
|
|
|
| 16 |
|
| 17 |
# training_baseline
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/wikipedia_seed_science dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 2.
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.980891719745223,
|
| 3 |
+
"total_flos": 55447011295232.0,
|
| 4 |
+
"train_loss": 0.7642028572467657,
|
| 5 |
+
"train_runtime": 2175.16,
|
| 6 |
+
"train_samples_per_second": 6.897,
|
| 7 |
+
"train_steps_per_second": 0.072
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 2.
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.980891719745223,
|
| 3 |
+
"total_flos": 55447011295232.0,
|
| 4 |
+
"train_loss": 0.7642028572467657,
|
| 5 |
+
"train_runtime": 2175.16,
|
| 6 |
+
"train_samples_per_second": 6.897,
|
| 7 |
+
"train_steps_per_second": 0.072
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,235 +1,1117 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 2.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 3.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss":
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch":
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch":
|
| 90 |
-
"grad_norm":
|
| 91 |
"learning_rate": 7.500000000000001e-06,
|
| 92 |
-
"loss":
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss":
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch":
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss":
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch":
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch":
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch":
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"grad_norm": 1.
|
| 154 |
-
"learning_rate":
|
| 155 |
-
"loss":
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch":
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate":
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch":
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch":
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate":
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch":
|
| 202 |
-
"grad_norm":
|
| 203 |
-
"learning_rate":
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch":
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch":
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"
|
| 224 |
-
"
|
| 225 |
-
"
|
| 226 |
-
"
|
| 227 |
-
|
| 228 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 229 |
}
|
| 230 |
],
|
| 231 |
"logging_steps": 1.0,
|
| 232 |
-
"max_steps":
|
| 233 |
"num_input_tokens_seen": 0,
|
| 234 |
"num_train_epochs": 3,
|
| 235 |
"save_steps": 500,
|
|
@@ -245,7 +1127,7 @@
|
|
| 245 |
"attributes": {}
|
| 246 |
}
|
| 247 |
},
|
| 248 |
-
"total_flos":
|
| 249 |
"train_batch_size": 1,
|
| 250 |
"trial_name": null,
|
| 251 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.980891719745223,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 156,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.01910828025477707,
|
| 13 |
+
"grad_norm": 9.2407865524292,
|
| 14 |
+
"learning_rate": 6.25e-07,
|
| 15 |
+
"loss": 1.3429,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.03821656050955414,
|
| 20 |
+
"grad_norm": 9.478723526000977,
|
| 21 |
+
"learning_rate": 1.25e-06,
|
| 22 |
+
"loss": 1.3616,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.05732484076433121,
|
| 27 |
+
"grad_norm": 8.822339057922363,
|
| 28 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 29 |
+
"loss": 1.2959,
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.07643312101910828,
|
| 34 |
+
"grad_norm": 8.794820785522461,
|
| 35 |
+
"learning_rate": 2.5e-06,
|
| 36 |
+
"loss": 1.3065,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.09554140127388536,
|
| 41 |
+
"grad_norm": 8.383583068847656,
|
| 42 |
+
"learning_rate": 3.125e-06,
|
| 43 |
+
"loss": 1.3084,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.11464968152866242,
|
| 48 |
+
"grad_norm": 6.294646263122559,
|
| 49 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 50 |
+
"loss": 1.2614,
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.1337579617834395,
|
| 55 |
+
"grad_norm": 3.9098663330078125,
|
| 56 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 57 |
+
"loss": 1.177,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.15286624203821655,
|
| 62 |
+
"grad_norm": 3.4435949325561523,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 1.2101,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.17197452229299362,
|
| 69 |
+
"grad_norm": 4.044913291931152,
|
| 70 |
+
"learning_rate": 5.625e-06,
|
| 71 |
+
"loss": 1.1228,
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.1910828025477707,
|
| 76 |
+
"grad_norm": 4.243114948272705,
|
| 77 |
+
"learning_rate": 6.25e-06,
|
| 78 |
+
"loss": 1.0775,
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.21019108280254778,
|
| 83 |
+
"grad_norm": 4.074731349945068,
|
| 84 |
+
"learning_rate": 6.875e-06,
|
| 85 |
+
"loss": 1.0759,
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.22929936305732485,
|
| 90 |
+
"grad_norm": 3.4349539279937744,
|
| 91 |
"learning_rate": 7.500000000000001e-06,
|
| 92 |
+
"loss": 1.0192,
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.2484076433121019,
|
| 97 |
+
"grad_norm": 3.8267672061920166,
|
| 98 |
+
"learning_rate": 8.125000000000001e-06,
|
| 99 |
+
"loss": 1.0476,
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.267515923566879,
|
| 104 |
+
"grad_norm": 3.0685765743255615,
|
| 105 |
+
"learning_rate": 8.750000000000001e-06,
|
| 106 |
+
"loss": 1.0031,
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.28662420382165604,
|
| 111 |
+
"grad_norm": 2.2036056518554688,
|
| 112 |
+
"learning_rate": 9.375000000000001e-06,
|
| 113 |
+
"loss": 0.9644,
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.3057324840764331,
|
| 118 |
+
"grad_norm": 2.0634169578552246,
|
| 119 |
+
"learning_rate": 1e-05,
|
| 120 |
+
"loss": 0.958,
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.3248407643312102,
|
| 125 |
+
"grad_norm": 2.5293338298797607,
|
| 126 |
+
"learning_rate": 9.998741174712534e-06,
|
| 127 |
+
"loss": 0.9512,
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.34394904458598724,
|
| 132 |
+
"grad_norm": 2.4025402069091797,
|
| 133 |
+
"learning_rate": 9.994965332706574e-06,
|
| 134 |
+
"loss": 0.9528,
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.3630573248407643,
|
| 139 |
+
"grad_norm": 1.8339530229568481,
|
| 140 |
+
"learning_rate": 9.98867437523228e-06,
|
| 141 |
+
"loss": 0.9513,
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.3821656050955414,
|
| 146 |
+
"grad_norm": 1.4116219282150269,
|
| 147 |
+
"learning_rate": 9.979871469976197e-06,
|
| 148 |
+
"loss": 0.9269,
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.4012738853503185,
|
| 153 |
+
"grad_norm": 1.4383982419967651,
|
| 154 |
+
"learning_rate": 9.968561049466214e-06,
|
| 155 |
+
"loss": 0.9384,
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.42038216560509556,
|
| 160 |
+
"grad_norm": 1.4755058288574219,
|
| 161 |
+
"learning_rate": 9.954748808839675e-06,
|
| 162 |
+
"loss": 0.8952,
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.4394904458598726,
|
| 167 |
+
"grad_norm": 1.1699796915054321,
|
| 168 |
+
"learning_rate": 9.938441702975689e-06,
|
| 169 |
+
"loss": 0.8937,
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.4585987261146497,
|
| 174 |
+
"grad_norm": 1.0133947134017944,
|
| 175 |
+
"learning_rate": 9.91964794299315e-06,
|
| 176 |
+
"loss": 0.9142,
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.47770700636942676,
|
| 181 |
+
"grad_norm": 1.1321440935134888,
|
| 182 |
+
"learning_rate": 9.898376992116179e-06,
|
| 183 |
+
"loss": 0.8876,
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.4968152866242038,
|
| 188 |
+
"grad_norm": 1.1128759384155273,
|
| 189 |
+
"learning_rate": 9.874639560909118e-06,
|
| 190 |
+
"loss": 0.8914,
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.5159235668789809,
|
| 195 |
+
"grad_norm": 1.152822732925415,
|
| 196 |
+
"learning_rate": 9.848447601883436e-06,
|
| 197 |
+
"loss": 0.918,
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.535031847133758,
|
| 202 |
+
"grad_norm": 1.0430890321731567,
|
| 203 |
+
"learning_rate": 9.819814303479268e-06,
|
| 204 |
+
"loss": 0.9226,
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.554140127388535,
|
| 209 |
+
"grad_norm": 1.0511760711669922,
|
| 210 |
+
"learning_rate": 9.788754083424654e-06,
|
| 211 |
+
"loss": 0.8652,
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.5732484076433121,
|
| 216 |
+
"grad_norm": 0.9650465250015259,
|
| 217 |
+
"learning_rate": 9.755282581475769e-06,
|
| 218 |
+
"loss": 0.8734,
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.5923566878980892,
|
| 223 |
+
"grad_norm": 0.9453793168067932,
|
| 224 |
+
"learning_rate": 9.719416651541839e-06,
|
| 225 |
+
"loss": 0.8965,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6114649681528662,
|
| 230 |
+
"grad_norm": 1.0314666032791138,
|
| 231 |
+
"learning_rate": 9.681174353198687e-06,
|
| 232 |
+
"loss": 0.896,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6305732484076433,
|
| 237 |
+
"grad_norm": 1.0188624858856201,
|
| 238 |
+
"learning_rate": 9.640574942595195e-06,
|
| 239 |
+
"loss": 0.8808,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6496815286624203,
|
| 244 |
+
"grad_norm": 0.9808051586151123,
|
| 245 |
+
"learning_rate": 9.597638862757255e-06,
|
| 246 |
+
"loss": 0.8398,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.6687898089171974,
|
| 251 |
+
"grad_norm": 6.07795524597168,
|
| 252 |
+
"learning_rate": 9.552387733294081e-06,
|
| 253 |
+
"loss": 0.8645,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6878980891719745,
|
| 258 |
+
"grad_norm": 1.1291970014572144,
|
| 259 |
+
"learning_rate": 9.504844339512096e-06,
|
| 260 |
+
"loss": 0.8813,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7070063694267515,
|
| 265 |
+
"grad_norm": 1.2054293155670166,
|
| 266 |
+
"learning_rate": 9.45503262094184e-06,
|
| 267 |
+
"loss": 0.8801,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7261146496815286,
|
| 272 |
+
"grad_norm": 1.0582106113433838,
|
| 273 |
+
"learning_rate": 9.40297765928369e-06,
|
| 274 |
+
"loss": 0.8923,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.7452229299363057,
|
| 279 |
+
"grad_norm": 0.9821107983589172,
|
| 280 |
+
"learning_rate": 9.348705665778479e-06,
|
| 281 |
+
"loss": 0.856,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.7643312101910829,
|
| 286 |
+
"grad_norm": 0.9655761122703552,
|
| 287 |
+
"learning_rate": 9.292243968009332e-06,
|
| 288 |
+
"loss": 0.8046,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7834394904458599,
|
| 293 |
+
"grad_norm": 1.0355433225631714,
|
| 294 |
+
"learning_rate": 9.233620996141421e-06,
|
| 295 |
+
"loss": 0.8774,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.802547770700637,
|
| 300 |
+
"grad_norm": 0.9365308880805969,
|
| 301 |
+
"learning_rate": 9.172866268606514e-06,
|
| 302 |
+
"loss": 0.8529,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.821656050955414,
|
| 307 |
+
"grad_norm": 1.033125638961792,
|
| 308 |
+
"learning_rate": 9.110010377239552e-06,
|
| 309 |
+
"loss": 0.8878,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.8407643312101911,
|
| 314 |
+
"grad_norm": 1.0032130479812622,
|
| 315 |
+
"learning_rate": 9.045084971874738e-06,
|
| 316 |
+
"loss": 0.8456,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.8598726114649682,
|
| 321 |
+
"grad_norm": 0.9883443117141724,
|
| 322 |
+
"learning_rate": 8.978122744408905e-06,
|
| 323 |
+
"loss": 0.8591,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8789808917197452,
|
| 328 |
+
"grad_norm": 1.064292073249817,
|
| 329 |
+
"learning_rate": 8.90915741234015e-06,
|
| 330 |
+
"loss": 0.8302,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.8980891719745223,
|
| 335 |
+
"grad_norm": 0.9541292786598206,
|
| 336 |
+
"learning_rate": 8.838223701790057e-06,
|
| 337 |
+
"loss": 0.8265,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.9171974522292994,
|
| 342 |
+
"grad_norm": 1.0820716619491577,
|
| 343 |
+
"learning_rate": 8.765357330018056e-06,
|
| 344 |
+
"loss": 0.8239,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.9363057324840764,
|
| 349 |
+
"grad_norm": 1.0467376708984375,
|
| 350 |
+
"learning_rate": 8.690594987436705e-06,
|
| 351 |
+
"loss": 0.8543,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.9554140127388535,
|
| 356 |
+
"grad_norm": 0.9927631616592407,
|
| 357 |
+
"learning_rate": 8.613974319136959e-06,
|
| 358 |
+
"loss": 0.8486,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9745222929936306,
|
| 363 |
+
"grad_norm": 1.1420801877975464,
|
| 364 |
+
"learning_rate": 8.535533905932739e-06,
|
| 365 |
+
"loss": 0.8717,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9936305732484076,
|
| 370 |
+
"grad_norm": 0.9995855093002319,
|
| 371 |
+
"learning_rate": 8.455313244934324e-06,
|
| 372 |
+
"loss": 0.851,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.0127388535031847,
|
| 377 |
+
"grad_norm": 1.773233413696289,
|
| 378 |
+
"learning_rate": 8.373352729660373e-06,
|
| 379 |
+
"loss": 1.2804,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.0318471337579618,
|
| 384 |
+
"grad_norm": 1.146483302116394,
|
| 385 |
+
"learning_rate": 8.289693629698564e-06,
|
| 386 |
+
"loss": 0.736,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.0509554140127388,
|
| 391 |
+
"grad_norm": 0.9704205989837646,
|
| 392 |
+
"learning_rate": 8.204378069925121e-06,
|
| 393 |
+
"loss": 0.7509,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.070063694267516,
|
| 398 |
+
"grad_norm": 0.9973506331443787,
|
| 399 |
+
"learning_rate": 8.117449009293668e-06,
|
| 400 |
+
"loss": 0.7378,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.089171974522293,
|
| 405 |
+
"grad_norm": 0.9104877710342407,
|
| 406 |
+
"learning_rate": 8.0289502192041e-06,
|
| 407 |
+
"loss": 0.696,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.10828025477707,
|
| 412 |
+
"grad_norm": 1.128894567489624,
|
| 413 |
+
"learning_rate": 7.938926261462366e-06,
|
| 414 |
+
"loss": 0.7265,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.127388535031847,
|
| 419 |
+
"grad_norm": 0.9157301783561707,
|
| 420 |
+
"learning_rate": 7.84742246584226e-06,
|
| 421 |
+
"loss": 0.7366,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.1464968152866242,
|
| 426 |
+
"grad_norm": 0.9085549116134644,
|
| 427 |
+
"learning_rate": 7.754484907260513e-06,
|
| 428 |
+
"loss": 0.6838,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.1656050955414012,
|
| 433 |
+
"grad_norm": 0.9201124906539917,
|
| 434 |
+
"learning_rate": 7.660160382576683e-06,
|
| 435 |
+
"loss": 0.7569,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.1847133757961783,
|
| 440 |
+
"grad_norm": 0.9689205288887024,
|
| 441 |
+
"learning_rate": 7.564496387029532e-06,
|
| 442 |
+
"loss": 0.6864,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.2038216560509554,
|
| 447 |
+
"grad_norm": 0.9296672344207764,
|
| 448 |
+
"learning_rate": 7.467541090321735e-06,
|
| 449 |
+
"loss": 0.7402,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.2229299363057324,
|
| 454 |
+
"grad_norm": 0.965122640132904,
|
| 455 |
+
"learning_rate": 7.369343312364994e-06,
|
| 456 |
+
"loss": 0.6695,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.2420382165605095,
|
| 461 |
+
"grad_norm": 0.9942020177841187,
|
| 462 |
+
"learning_rate": 7.269952498697734e-06,
|
| 463 |
+
"loss": 0.7511,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.2611464968152866,
|
| 468 |
+
"grad_norm": 0.9647576808929443,
|
| 469 |
+
"learning_rate": 7.169418695587791e-06,
|
| 470 |
+
"loss": 0.7272,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.2802547770700636,
|
| 475 |
+
"grad_norm": 0.8429343104362488,
|
| 476 |
+
"learning_rate": 7.067792524832604e-06,
|
| 477 |
+
"loss": 0.743,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.2993630573248407,
|
| 482 |
+
"grad_norm": 0.8730988502502441,
|
| 483 |
+
"learning_rate": 6.965125158269619e-06,
|
| 484 |
+
"loss": 0.689,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.3184713375796178,
|
| 489 |
+
"grad_norm": 0.95030677318573,
|
| 490 |
+
"learning_rate": 6.8614682920097265e-06,
|
| 491 |
+
"loss": 0.7515,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.3375796178343948,
|
| 496 |
+
"grad_norm": 0.8007903099060059,
|
| 497 |
+
"learning_rate": 6.7568741204067145e-06,
|
| 498 |
+
"loss": 0.6767,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.356687898089172,
|
| 503 |
+
"grad_norm": 0.9224691390991211,
|
| 504 |
+
"learning_rate": 6.651395309775837e-06,
|
| 505 |
+
"loss": 0.7164,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.3757961783439492,
|
| 510 |
+
"grad_norm": 0.7532793283462524,
|
| 511 |
+
"learning_rate": 6.545084971874738e-06,
|
| 512 |
+
"loss": 0.6844,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.394904458598726,
|
| 517 |
+
"grad_norm": 0.8250970244407654,
|
| 518 |
+
"learning_rate": 6.437996637160086e-06,
|
| 519 |
+
"loss": 0.7026,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.4140127388535033,
|
| 524 |
+
"grad_norm": 0.8471970558166504,
|
| 525 |
+
"learning_rate": 6.330184227833376e-06,
|
| 526 |
+
"loss": 0.6725,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.4331210191082802,
|
| 531 |
+
"grad_norm": 0.8257316946983337,
|
| 532 |
+
"learning_rate": 6.2217020306894705e-06,
|
| 533 |
+
"loss": 0.6793,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.4522292993630574,
|
| 538 |
+
"grad_norm": 0.7594157457351685,
|
| 539 |
+
"learning_rate": 6.112604669781572e-06,
|
| 540 |
+
"loss": 0.71,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.4713375796178343,
|
| 545 |
+
"grad_norm": 0.8231765627861023,
|
| 546 |
+
"learning_rate": 6.002947078916365e-06,
|
| 547 |
+
"loss": 0.7449,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.4904458598726116,
|
| 552 |
+
"grad_norm": 0.7706438899040222,
|
| 553 |
+
"learning_rate": 5.892784473993184e-06,
|
| 554 |
+
"loss": 0.6683,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.5095541401273884,
|
| 559 |
+
"grad_norm": 0.7732676863670349,
|
| 560 |
+
"learning_rate": 5.782172325201155e-06,
|
| 561 |
+
"loss": 0.6617,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.5286624203821657,
|
| 566 |
+
"grad_norm": 0.7970235347747803,
|
| 567 |
+
"learning_rate": 5.671166329088278e-06,
|
| 568 |
+
"loss": 0.7013,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.5477707006369426,
|
| 573 |
+
"grad_norm": 0.7982433438301086,
|
| 574 |
+
"learning_rate": 5.559822380516539e-06,
|
| 575 |
+
"loss": 0.6928,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.5668789808917198,
|
| 580 |
+
"grad_norm": 0.7948117852210999,
|
| 581 |
+
"learning_rate": 5.448196544517168e-06,
|
| 582 |
+
"loss": 0.7225,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.5859872611464967,
|
| 587 |
+
"grad_norm": 0.8307470083236694,
|
| 588 |
+
"learning_rate": 5.336345028060199e-06,
|
| 589 |
+
"loss": 0.7372,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.605095541401274,
|
| 594 |
+
"grad_norm": 0.7786919474601746,
|
| 595 |
+
"learning_rate": 5.224324151752575e-06,
|
| 596 |
+
"loss": 0.6999,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.6242038216560508,
|
| 601 |
+
"grad_norm": 0.8192756772041321,
|
| 602 |
+
"learning_rate": 5.112190321479026e-06,
|
| 603 |
+
"loss": 0.7043,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.643312101910828,
|
| 608 |
+
"grad_norm": 0.7583406567573547,
|
| 609 |
+
"learning_rate": 5e-06,
|
| 610 |
+
"loss": 0.6697,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.662420382165605,
|
| 615 |
+
"grad_norm": 0.8197351694107056,
|
| 616 |
+
"learning_rate": 4.887809678520976e-06,
|
| 617 |
+
"loss": 0.7563,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.6815286624203822,
|
| 622 |
+
"grad_norm": 0.8537828922271729,
|
| 623 |
+
"learning_rate": 4.775675848247427e-06,
|
| 624 |
+
"loss": 0.714,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.700636942675159,
|
| 629 |
+
"grad_norm": 0.8561527729034424,
|
| 630 |
+
"learning_rate": 4.663654971939802e-06,
|
| 631 |
+
"loss": 0.6812,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.7197452229299364,
|
| 636 |
+
"grad_norm": 0.7625755667686462,
|
| 637 |
+
"learning_rate": 4.551803455482833e-06,
|
| 638 |
+
"loss": 0.7148,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.7388535031847132,
|
| 643 |
+
"grad_norm": 0.7977814078330994,
|
| 644 |
+
"learning_rate": 4.4401776194834615e-06,
|
| 645 |
+
"loss": 0.7023,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.7579617834394905,
|
| 650 |
+
"grad_norm": 0.8178286552429199,
|
| 651 |
+
"learning_rate": 4.3288336709117246e-06,
|
| 652 |
+
"loss": 0.7438,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.7770700636942676,
|
| 657 |
+
"grad_norm": 0.8001686334609985,
|
| 658 |
+
"learning_rate": 4.217827674798845e-06,
|
| 659 |
+
"loss": 0.6842,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.7961783439490446,
|
| 664 |
+
"grad_norm": 0.754313051700592,
|
| 665 |
+
"learning_rate": 4.107215526006818e-06,
|
| 666 |
+
"loss": 0.6574,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.8152866242038217,
|
| 671 |
+
"grad_norm": 0.7473746538162231,
|
| 672 |
+
"learning_rate": 3.997052921083637e-06,
|
| 673 |
+
"loss": 0.7421,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.8343949044585988,
|
| 678 |
+
"grad_norm": 0.7881169319152832,
|
| 679 |
+
"learning_rate": 3.887395330218429e-06,
|
| 680 |
+
"loss": 0.7061,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.8535031847133758,
|
| 685 |
+
"grad_norm": 0.7832265496253967,
|
| 686 |
+
"learning_rate": 3.778297969310529e-06,
|
| 687 |
+
"loss": 0.6998,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.872611464968153,
|
| 692 |
+
"grad_norm": 0.8894858956336975,
|
| 693 |
+
"learning_rate": 3.669815772166625e-06,
|
| 694 |
+
"loss": 0.7187,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.89171974522293,
|
| 699 |
+
"grad_norm": 0.7932900190353394,
|
| 700 |
+
"learning_rate": 3.562003362839914e-06,
|
| 701 |
+
"loss": 0.6835,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.910828025477707,
|
| 706 |
+
"grad_norm": 0.8342806696891785,
|
| 707 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 708 |
+
"loss": 0.6395,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.929936305732484,
|
| 713 |
+
"grad_norm": 0.8320187926292419,
|
| 714 |
+
"learning_rate": 3.3486046902241663e-06,
|
| 715 |
+
"loss": 0.7582,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.9490445859872612,
|
| 720 |
+
"grad_norm": 0.7808068990707397,
|
| 721 |
+
"learning_rate": 3.2431258795932863e-06,
|
| 722 |
+
"loss": 0.6748,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.9681528662420382,
|
| 727 |
+
"grad_norm": 0.6990087032318115,
|
| 728 |
+
"learning_rate": 3.1385317079902743e-06,
|
| 729 |
+
"loss": 0.6935,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.9872611464968153,
|
| 734 |
+
"grad_norm": 0.8083034753799438,
|
| 735 |
+
"learning_rate": 3.0348748417303826e-06,
|
| 736 |
+
"loss": 0.6842,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.0063694267515926,
|
| 741 |
+
"grad_norm": 1.6020653247833252,
|
| 742 |
+
"learning_rate": 2.932207475167398e-06,
|
| 743 |
+
"loss": 1.1038,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.0254777070063694,
|
| 748 |
+
"grad_norm": 0.8427936434745789,
|
| 749 |
+
"learning_rate": 2.83058130441221e-06,
|
| 750 |
+
"loss": 0.6547,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.0445859872611467,
|
| 755 |
+
"grad_norm": 0.8315389156341553,
|
| 756 |
+
"learning_rate": 2.7300475013022666e-06,
|
| 757 |
+
"loss": 0.565,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.0636942675159236,
|
| 762 |
+
"grad_norm": 0.7301693558692932,
|
| 763 |
+
"learning_rate": 2.6306566876350072e-06,
|
| 764 |
+
"loss": 0.5997,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.082802547770701,
|
| 769 |
+
"grad_norm": 0.6864070892333984,
|
| 770 |
+
"learning_rate": 2.532458909678266e-06,
|
| 771 |
+
"loss": 0.5757,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.1019108280254777,
|
| 776 |
+
"grad_norm": 0.7325605750083923,
|
| 777 |
+
"learning_rate": 2.43550361297047e-06,
|
| 778 |
+
"loss": 0.6274,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.121019108280255,
|
| 783 |
+
"grad_norm": 0.7794485092163086,
|
| 784 |
+
"learning_rate": 2.339839617423318e-06,
|
| 785 |
+
"loss": 0.6346,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.140127388535032,
|
| 790 |
+
"grad_norm": 0.7400060892105103,
|
| 791 |
+
"learning_rate": 2.245515092739488e-06,
|
| 792 |
+
"loss": 0.6,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.159235668789809,
|
| 797 |
+
"grad_norm": 0.8290343284606934,
|
| 798 |
+
"learning_rate": 2.1525775341577404e-06,
|
| 799 |
+
"loss": 0.6427,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.178343949044586,
|
| 804 |
+
"grad_norm": 0.7199001312255859,
|
| 805 |
+
"learning_rate": 2.061073738537635e-06,
|
| 806 |
+
"loss": 0.5617,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.1974522292993632,
|
| 811 |
+
"grad_norm": 0.7009884715080261,
|
| 812 |
+
"learning_rate": 1.971049780795901e-06,
|
| 813 |
+
"loss": 0.5894,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.21656050955414,
|
| 818 |
+
"grad_norm": 0.7035958766937256,
|
| 819 |
+
"learning_rate": 1.8825509907063328e-06,
|
| 820 |
+
"loss": 0.5989,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.2356687898089174,
|
| 825 |
+
"grad_norm": 0.7354803085327148,
|
| 826 |
+
"learning_rate": 1.7956219300748796e-06,
|
| 827 |
+
"loss": 0.6264,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.254777070063694,
|
| 832 |
+
"grad_norm": 0.6822336316108704,
|
| 833 |
+
"learning_rate": 1.7103063703014372e-06,
|
| 834 |
+
"loss": 0.5717,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.2738853503184715,
|
| 839 |
+
"grad_norm": 0.7196558117866516,
|
| 840 |
+
"learning_rate": 1.6266472703396286e-06,
|
| 841 |
+
"loss": 0.608,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.2929936305732483,
|
| 846 |
+
"grad_norm": 0.7429059743881226,
|
| 847 |
+
"learning_rate": 1.544686755065677e-06,
|
| 848 |
+
"loss": 0.6242,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.3121019108280256,
|
| 853 |
+
"grad_norm": 0.7424628138542175,
|
| 854 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 855 |
+
"loss": 0.5927,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.3312101910828025,
|
| 860 |
+
"grad_norm": 0.7479440569877625,
|
| 861 |
+
"learning_rate": 1.3860256808630429e-06,
|
| 862 |
+
"loss": 0.5857,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.3503184713375798,
|
| 867 |
+
"grad_norm": 0.7213776707649231,
|
| 868 |
+
"learning_rate": 1.3094050125632973e-06,
|
| 869 |
+
"loss": 0.6115,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.3694267515923566,
|
| 874 |
+
"grad_norm": 0.6591588258743286,
|
| 875 |
+
"learning_rate": 1.234642669981946e-06,
|
| 876 |
+
"loss": 0.5559,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.388535031847134,
|
| 881 |
+
"grad_norm": 0.7499435544013977,
|
| 882 |
+
"learning_rate": 1.1617762982099446e-06,
|
| 883 |
+
"loss": 0.6358,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.4076433121019107,
|
| 888 |
+
"grad_norm": 0.6854066848754883,
|
| 889 |
+
"learning_rate": 1.0908425876598512e-06,
|
| 890 |
+
"loss": 0.5673,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.426751592356688,
|
| 895 |
+
"grad_norm": 0.7064222693443298,
|
| 896 |
+
"learning_rate": 1.0218772555910955e-06,
|
| 897 |
+
"loss": 0.603,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.445859872611465,
|
| 902 |
+
"grad_norm": 0.7362521290779114,
|
| 903 |
+
"learning_rate": 9.549150281252633e-07,
|
| 904 |
+
"loss": 0.6113,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.464968152866242,
|
| 909 |
+
"grad_norm": 0.742847740650177,
|
| 910 |
+
"learning_rate": 8.899896227604509e-07,
|
| 911 |
+
"loss": 0.601,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.484076433121019,
|
| 916 |
+
"grad_norm": 0.6763571500778198,
|
| 917 |
+
"learning_rate": 8.271337313934869e-07,
|
| 918 |
+
"loss": 0.5505,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.5031847133757963,
|
| 923 |
+
"grad_norm": 0.7390684485435486,
|
| 924 |
+
"learning_rate": 7.663790038585794e-07,
|
| 925 |
+
"loss": 0.6827,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.522292993630573,
|
| 930 |
+
"grad_norm": 0.6655154228210449,
|
| 931 |
+
"learning_rate": 7.077560319906696e-07,
|
| 932 |
+
"loss": 0.5709,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.5414012738853504,
|
| 937 |
+
"grad_norm": 0.7252643704414368,
|
| 938 |
+
"learning_rate": 6.512943342215234e-07,
|
| 939 |
+
"loss": 0.6091,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.5605095541401273,
|
| 944 |
+
"grad_norm": 0.6566873788833618,
|
| 945 |
+
"learning_rate": 5.9702234071631e-07,
|
| 946 |
+
"loss": 0.5609,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.5796178343949046,
|
| 951 |
+
"grad_norm": 0.6787551641464233,
|
| 952 |
+
"learning_rate": 5.449673790581611e-07,
|
| 953 |
+
"loss": 0.5865,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.5987261146496814,
|
| 958 |
+
"grad_norm": 0.6773931980133057,
|
| 959 |
+
"learning_rate": 4.951556604879049e-07,
|
| 960 |
+
"loss": 0.5847,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.6178343949044587,
|
| 965 |
+
"grad_norm": 0.7028627395629883,
|
| 966 |
+
"learning_rate": 4.4761226670592074e-07,
|
| 967 |
+
"loss": 0.5941,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.6369426751592355,
|
| 972 |
+
"grad_norm": 0.6910837888717651,
|
| 973 |
+
"learning_rate": 4.0236113724274716e-07,
|
| 974 |
+
"loss": 0.6,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.656050955414013,
|
| 979 |
+
"grad_norm": 0.6671729683876038,
|
| 980 |
+
"learning_rate": 3.5942505740480583e-07,
|
| 981 |
+
"loss": 0.5928,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.6751592356687897,
|
| 986 |
+
"grad_norm": 0.6896730065345764,
|
| 987 |
+
"learning_rate": 3.18825646801314e-07,
|
| 988 |
+
"loss": 0.5801,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.694267515923567,
|
| 993 |
+
"grad_norm": 0.6889122128486633,
|
| 994 |
+
"learning_rate": 2.8058334845816214e-07,
|
| 995 |
+
"loss": 0.6199,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.713375796178344,
|
| 1000 |
+
"grad_norm": 0.691997230052948,
|
| 1001 |
+
"learning_rate": 2.447174185242324e-07,
|
| 1002 |
+
"loss": 0.6229,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.732484076433121,
|
| 1007 |
+
"grad_norm": 0.6809545755386353,
|
| 1008 |
+
"learning_rate": 2.1124591657534776e-07,
|
| 1009 |
+
"loss": 0.6053,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.7515923566878984,
|
| 1014 |
+
"grad_norm": 0.670366108417511,
|
| 1015 |
+
"learning_rate": 1.801856965207338e-07,
|
| 1016 |
+
"loss": 0.5444,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.770700636942675,
|
| 1021 |
+
"grad_norm": 0.664279580116272,
|
| 1022 |
+
"learning_rate": 1.5155239811656562e-07,
|
| 1023 |
+
"loss": 0.5756,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.789808917197452,
|
| 1028 |
+
"grad_norm": 0.6993831396102905,
|
| 1029 |
+
"learning_rate": 1.253604390908819e-07,
|
| 1030 |
+
"loss": 0.6188,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.8089171974522293,
|
| 1035 |
+
"grad_norm": 0.7156822085380554,
|
| 1036 |
+
"learning_rate": 1.0162300788382263e-07,
|
| 1037 |
+
"loss": 0.6433,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.8280254777070066,
|
| 1042 |
+
"grad_norm": 0.6507427096366882,
|
| 1043 |
+
"learning_rate": 8.035205700685167e-08,
|
| 1044 |
+
"loss": 0.5587,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.8471337579617835,
|
| 1049 |
+
"grad_norm": 0.624755322933197,
|
| 1050 |
+
"learning_rate": 6.15582970243117e-08,
|
| 1051 |
+
"loss": 0.5927,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.8662420382165603,
|
| 1056 |
+
"grad_norm": 0.6648584604263306,
|
| 1057 |
+
"learning_rate": 4.52511911603265e-08,
|
| 1058 |
+
"loss": 0.5883,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.8853503184713376,
|
| 1063 |
+
"grad_norm": 0.6581620573997498,
|
| 1064 |
+
"learning_rate": 3.143895053378698e-08,
|
| 1065 |
+
"loss": 0.5976,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.904458598726115,
|
| 1070 |
+
"grad_norm": 0.699501633644104,
|
| 1071 |
+
"learning_rate": 2.012853002380466e-08,
|
| 1072 |
+
"loss": 0.5732,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.9235668789808917,
|
| 1077 |
+
"grad_norm": 0.7073370814323425,
|
| 1078 |
+
"learning_rate": 1.132562476771959e-08,
|
| 1079 |
+
"loss": 0.6537,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.9426751592356686,
|
| 1084 |
+
"grad_norm": 0.6681355834007263,
|
| 1085 |
+
"learning_rate": 5.034667293427053e-09,
|
| 1086 |
+
"loss": 0.5774,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.961783439490446,
|
| 1091 |
+
"grad_norm": 0.6787852048873901,
|
| 1092 |
+
"learning_rate": 1.2588252874673469e-09,
|
| 1093 |
+
"loss": 0.58,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.980891719745223,
|
| 1098 |
+
"grad_norm": 0.6549314856529236,
|
| 1099 |
+
"learning_rate": 0.0,
|
| 1100 |
+
"loss": 0.6071,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.980891719745223,
|
| 1105 |
+
"step": 156,
|
| 1106 |
+
"total_flos": 55447011295232.0,
|
| 1107 |
+
"train_loss": 0.7642028572467657,
|
| 1108 |
+
"train_runtime": 2175.16,
|
| 1109 |
+
"train_samples_per_second": 6.897,
|
| 1110 |
+
"train_steps_per_second": 0.072
|
| 1111 |
}
|
| 1112 |
],
|
| 1113 |
"logging_steps": 1.0,
|
| 1114 |
+
"max_steps": 156,
|
| 1115 |
"num_input_tokens_seen": 0,
|
| 1116 |
"num_train_epochs": 3,
|
| 1117 |
"save_steps": 500,
|
|
|
|
| 1127 |
"attributes": {}
|
| 1128 |
}
|
| 1129 |
},
|
| 1130 |
+
"total_flos": 55447011295232.0,
|
| 1131 |
"train_batch_size": 1,
|
| 1132 |
"trial_name": null,
|
| 1133 |
"trial_params": null
|
training_loss.png
CHANGED
|
|