End of training
Browse files- README.md +1 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +487 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
# fahd-v1
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
- Loss: 0.1007
|
| 24 |
|
|
|
|
| 18 |
|
| 19 |
# fahd-v1
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the finetune_train dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
- Loss: 0.1007
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.1007455587387085,
|
| 4 |
+
"eval_runtime": 1.6485,
|
| 5 |
+
"eval_samples_per_second": 12.133,
|
| 6 |
+
"eval_steps_per_second": 12.133,
|
| 7 |
+
"total_flos": 3.884520534294528e+16,
|
| 8 |
+
"train_loss": 0.18234332187722127,
|
| 9 |
+
"train_runtime": 611.2552,
|
| 10 |
+
"train_samples_per_second": 31.084,
|
| 11 |
+
"train_steps_per_second": 0.491
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.1007455587387085,
|
| 4 |
+
"eval_runtime": 1.6485,
|
| 5 |
+
"eval_samples_per_second": 12.133,
|
| 6 |
+
"eval_steps_per_second": 12.133
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.884520534294528e+16,
|
| 4 |
+
"train_loss": 0.18234332187722127,
|
| 5 |
+
"train_runtime": 611.2552,
|
| 6 |
+
"train_samples_per_second": 31.084,
|
| 7 |
+
"train_steps_per_second": 0.491
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,487 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 300,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.16806722689075632,
|
| 14 |
+
"grad_norm": 5.1821064949035645,
|
| 15 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 16 |
+
"loss": 1.6929,
|
| 17 |
+
"step": 5
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.33613445378151263,
|
| 21 |
+
"grad_norm": 1.7028719186782837,
|
| 22 |
+
"learning_rate": 3e-05,
|
| 23 |
+
"loss": 1.317,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.5042016806722689,
|
| 28 |
+
"grad_norm": 0.8325338959693909,
|
| 29 |
+
"learning_rate": 4.666666666666667e-05,
|
| 30 |
+
"loss": 0.9858,
|
| 31 |
+
"step": 15
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.6722689075630253,
|
| 35 |
+
"grad_norm": 0.563953161239624,
|
| 36 |
+
"learning_rate": 6.333333333333333e-05,
|
| 37 |
+
"loss": 0.8006,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.8403361344537815,
|
| 42 |
+
"grad_norm": 0.6012160778045654,
|
| 43 |
+
"learning_rate": 8e-05,
|
| 44 |
+
"loss": 0.7177,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.0,
|
| 49 |
+
"grad_norm": 0.6171104311943054,
|
| 50 |
+
"learning_rate": 9.666666666666667e-05,
|
| 51 |
+
"loss": 0.6072,
|
| 52 |
+
"step": 30
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.1680672268907564,
|
| 56 |
+
"grad_norm": 0.5041235685348511,
|
| 57 |
+
"learning_rate": 9.994585556692624e-05,
|
| 58 |
+
"loss": 0.526,
|
| 59 |
+
"step": 35
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.3361344537815127,
|
| 63 |
+
"grad_norm": 0.5351182222366333,
|
| 64 |
+
"learning_rate": 9.972609476841367e-05,
|
| 65 |
+
"loss": 0.4891,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.504201680672269,
|
| 70 |
+
"grad_norm": 0.5860567688941956,
|
| 71 |
+
"learning_rate": 9.933807660562898e-05,
|
| 72 |
+
"loss": 0.4424,
|
| 73 |
+
"step": 45
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.6722689075630253,
|
| 77 |
+
"grad_norm": 0.6627805829048157,
|
| 78 |
+
"learning_rate": 9.878311400921072e-05,
|
| 79 |
+
"loss": 0.3993,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.8403361344537816,
|
| 84 |
+
"grad_norm": 0.6919167637825012,
|
| 85 |
+
"learning_rate": 9.806308479691595e-05,
|
| 86 |
+
"loss": 0.3435,
|
| 87 |
+
"step": 55
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 2.0,
|
| 91 |
+
"grad_norm": 0.7827972769737244,
|
| 92 |
+
"learning_rate": 9.718042531967918e-05,
|
| 93 |
+
"loss": 0.2951,
|
| 94 |
+
"step": 60
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 2.168067226890756,
|
| 98 |
+
"grad_norm": 0.6349595785140991,
|
| 99 |
+
"learning_rate": 9.613812221777212e-05,
|
| 100 |
+
"loss": 0.2419,
|
| 101 |
+
"step": 65
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.3361344537815127,
|
| 105 |
+
"grad_norm": 0.763509213924408,
|
| 106 |
+
"learning_rate": 9.493970231495835e-05,
|
| 107 |
+
"loss": 0.1926,
|
| 108 |
+
"step": 70
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.504201680672269,
|
| 112 |
+
"grad_norm": 0.706917405128479,
|
| 113 |
+
"learning_rate": 9.358922068483812e-05,
|
| 114 |
+
"loss": 0.1715,
|
| 115 |
+
"step": 75
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.6722689075630255,
|
| 119 |
+
"grad_norm": 0.6611737608909607,
|
| 120 |
+
"learning_rate": 9.209124692976287e-05,
|
| 121 |
+
"loss": 0.1689,
|
| 122 |
+
"step": 80
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.8403361344537816,
|
| 126 |
+
"grad_norm": 0.6016265749931335,
|
| 127 |
+
"learning_rate": 9.045084971874738e-05,
|
| 128 |
+
"loss": 0.148,
|
| 129 |
+
"step": 85
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 3.0,
|
| 133 |
+
"grad_norm": 0.7554067373275757,
|
| 134 |
+
"learning_rate": 8.86735796366982e-05,
|
| 135 |
+
"loss": 0.1496,
|
| 136 |
+
"step": 90
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 3.168067226890756,
|
| 140 |
+
"grad_norm": 0.7190783023834229,
|
| 141 |
+
"learning_rate": 8.676545040299145e-05,
|
| 142 |
+
"loss": 0.0929,
|
| 143 |
+
"step": 95
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.3361344537815127,
|
| 147 |
+
"grad_norm": 0.5446023941040039,
|
| 148 |
+
"learning_rate": 8.473291852294987e-05,
|
| 149 |
+
"loss": 0.0848,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 3.3361344537815127,
|
| 154 |
+
"eval_loss": 0.09926361590623856,
|
| 155 |
+
"eval_runtime": 1.6896,
|
| 156 |
+
"eval_samples_per_second": 11.837,
|
| 157 |
+
"eval_steps_per_second": 11.837,
|
| 158 |
+
"step": 100
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.504201680672269,
|
| 162 |
+
"grad_norm": 0.5175163149833679,
|
| 163 |
+
"learning_rate": 8.258286144107276e-05,
|
| 164 |
+
"loss": 0.0849,
|
| 165 |
+
"step": 105
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 3.6722689075630255,
|
| 169 |
+
"grad_norm": 0.6044480800628662,
|
| 170 |
+
"learning_rate": 8.032255426994069e-05,
|
| 171 |
+
"loss": 0.0927,
|
| 172 |
+
"step": 110
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 3.8403361344537816,
|
| 176 |
+
"grad_norm": 0.5574583411216736,
|
| 177 |
+
"learning_rate": 7.795964517353735e-05,
|
| 178 |
+
"loss": 0.0761,
|
| 179 |
+
"step": 115
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 4.0,
|
| 183 |
+
"grad_norm": 0.5631000995635986,
|
| 184 |
+
"learning_rate": 7.550212948828377e-05,
|
| 185 |
+
"loss": 0.0711,
|
| 186 |
+
"step": 120
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 4.168067226890757,
|
| 190 |
+
"grad_norm": 0.49856850504875183,
|
| 191 |
+
"learning_rate": 7.295832266935059e-05,
|
| 192 |
+
"loss": 0.0512,
|
| 193 |
+
"step": 125
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 4.336134453781512,
|
| 197 |
+
"grad_norm": 0.49295711517333984,
|
| 198 |
+
"learning_rate": 7.033683215379002e-05,
|
| 199 |
+
"loss": 0.0482,
|
| 200 |
+
"step": 130
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.504201680672269,
|
| 204 |
+
"grad_norm": 0.4027378559112549,
|
| 205 |
+
"learning_rate": 6.764652823569344e-05,
|
| 206 |
+
"loss": 0.0448,
|
| 207 |
+
"step": 135
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 4.6722689075630255,
|
| 211 |
+
"grad_norm": 0.328042209148407,
|
| 212 |
+
"learning_rate": 6.48965140519241e-05,
|
| 213 |
+
"loss": 0.0433,
|
| 214 |
+
"step": 140
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 4.840336134453781,
|
| 218 |
+
"grad_norm": 0.500453770160675,
|
| 219 |
+
"learning_rate": 6.209609477998338e-05,
|
| 220 |
+
"loss": 0.0524,
|
| 221 |
+
"step": 145
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 5.0,
|
| 225 |
+
"grad_norm": 0.4435368776321411,
|
| 226 |
+
"learning_rate": 5.925474615223573e-05,
|
| 227 |
+
"loss": 0.0431,
|
| 228 |
+
"step": 150
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 5.168067226890757,
|
| 232 |
+
"grad_norm": 0.4105546772480011,
|
| 233 |
+
"learning_rate": 5.6382082393029746e-05,
|
| 234 |
+
"loss": 0.0273,
|
| 235 |
+
"step": 155
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 5.336134453781512,
|
| 239 |
+
"grad_norm": 0.4708898067474365,
|
| 240 |
+
"learning_rate": 5.348782368720626e-05,
|
| 241 |
+
"loss": 0.0299,
|
| 242 |
+
"step": 160
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 5.504201680672269,
|
| 246 |
+
"grad_norm": 0.32217878103256226,
|
| 247 |
+
"learning_rate": 5.0581763290069865e-05,
|
| 248 |
+
"loss": 0.0252,
|
| 249 |
+
"step": 165
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 5.6722689075630255,
|
| 253 |
+
"grad_norm": 0.33504194021224976,
|
| 254 |
+
"learning_rate": 4.767373439011267e-05,
|
| 255 |
+
"loss": 0.0273,
|
| 256 |
+
"step": 170
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 5.840336134453781,
|
| 260 |
+
"grad_norm": 0.453778475522995,
|
| 261 |
+
"learning_rate": 4.477357683661734e-05,
|
| 262 |
+
"loss": 0.0324,
|
| 263 |
+
"step": 175
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 6.0,
|
| 267 |
+
"grad_norm": 0.42859596014022827,
|
| 268 |
+
"learning_rate": 4.1891103844721636e-05,
|
| 269 |
+
"loss": 0.0315,
|
| 270 |
+
"step": 180
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 6.168067226890757,
|
| 274 |
+
"grad_norm": 0.21872906386852264,
|
| 275 |
+
"learning_rate": 3.903606879060483e-05,
|
| 276 |
+
"loss": 0.0178,
|
| 277 |
+
"step": 185
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.336134453781512,
|
| 281 |
+
"grad_norm": 0.2727561295032501,
|
| 282 |
+
"learning_rate": 3.6218132209150045e-05,
|
| 283 |
+
"loss": 0.0172,
|
| 284 |
+
"step": 190
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.504201680672269,
|
| 288 |
+
"grad_norm": 0.259501188993454,
|
| 289 |
+
"learning_rate": 3.34468291057521e-05,
|
| 290 |
+
"loss": 0.0191,
|
| 291 |
+
"step": 195
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 6.6722689075630255,
|
| 295 |
+
"grad_norm": 0.2755616307258606,
|
| 296 |
+
"learning_rate": 3.073153669287759e-05,
|
| 297 |
+
"loss": 0.0182,
|
| 298 |
+
"step": 200
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 6.6722689075630255,
|
| 302 |
+
"eval_loss": 0.09819571673870087,
|
| 303 |
+
"eval_runtime": 1.6764,
|
| 304 |
+
"eval_samples_per_second": 11.93,
|
| 305 |
+
"eval_steps_per_second": 11.93,
|
| 306 |
+
"step": 200
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 6.840336134453781,
|
| 310 |
+
"grad_norm": 0.30396485328674316,
|
| 311 |
+
"learning_rate": 2.8081442660546125e-05,
|
| 312 |
+
"loss": 0.0185,
|
| 313 |
+
"step": 205
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 7.0,
|
| 317 |
+
"grad_norm": 0.3928929269313812,
|
| 318 |
+
"learning_rate": 2.5505514088095655e-05,
|
| 319 |
+
"loss": 0.0184,
|
| 320 |
+
"step": 210
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 7.168067226890757,
|
| 324 |
+
"grad_norm": 0.22996097803115845,
|
| 325 |
+
"learning_rate": 2.3012467102424373e-05,
|
| 326 |
+
"loss": 0.0121,
|
| 327 |
+
"step": 215
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 7.336134453781512,
|
| 331 |
+
"grad_norm": 0.19302956759929657,
|
| 332 |
+
"learning_rate": 2.061073738537635e-05,
|
| 333 |
+
"loss": 0.0122,
|
| 334 |
+
"step": 220
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 7.504201680672269,
|
| 338 |
+
"grad_norm": 0.1985304355621338,
|
| 339 |
+
"learning_rate": 1.8308451630064484e-05,
|
| 340 |
+
"loss": 0.0134,
|
| 341 |
+
"step": 225
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 7.6722689075630255,
|
| 345 |
+
"grad_norm": 0.13862721621990204,
|
| 346 |
+
"learning_rate": 1.611340004271339e-05,
|
| 347 |
+
"loss": 0.0123,
|
| 348 |
+
"step": 230
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 7.840336134453781,
|
| 352 |
+
"grad_norm": 0.1474885195493698,
|
| 353 |
+
"learning_rate": 1.4033009983067452e-05,
|
| 354 |
+
"loss": 0.0117,
|
| 355 |
+
"step": 235
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 8.0,
|
| 359 |
+
"grad_norm": 0.28437599539756775,
|
| 360 |
+
"learning_rate": 1.2074320832556556e-05,
|
| 361 |
+
"loss": 0.0113,
|
| 362 |
+
"step": 240
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 8.168067226890756,
|
| 366 |
+
"grad_norm": 0.12733525037765503,
|
| 367 |
+
"learning_rate": 1.0243960175257606e-05,
|
| 368 |
+
"loss": 0.0096,
|
| 369 |
+
"step": 245
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 8.336134453781513,
|
| 373 |
+
"grad_norm": 0.17934471368789673,
|
| 374 |
+
"learning_rate": 8.548121372247918e-06,
|
| 375 |
+
"loss": 0.0098,
|
| 376 |
+
"step": 250
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 8.504201680672269,
|
| 380 |
+
"grad_norm": 0.20203900337219238,
|
| 381 |
+
"learning_rate": 6.992542605231739e-06,
|
| 382 |
+
"loss": 0.0097,
|
| 383 |
+
"step": 255
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 8.672268907563025,
|
| 387 |
+
"grad_norm": 0.14116302132606506,
|
| 388 |
+
"learning_rate": 5.582487460349805e-06,
|
| 389 |
+
"loss": 0.0099,
|
| 390 |
+
"step": 260
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 8.840336134453782,
|
| 394 |
+
"grad_norm": 0.17721185088157654,
|
| 395 |
+
"learning_rate": 4.322727117869951e-06,
|
| 396 |
+
"loss": 0.0095,
|
| 397 |
+
"step": 265
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 9.0,
|
| 401 |
+
"grad_norm": 0.14386144280433655,
|
| 402 |
+
"learning_rate": 3.2175242080234313e-06,
|
| 403 |
+
"loss": 0.0092,
|
| 404 |
+
"step": 270
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 9.168067226890756,
|
| 408 |
+
"grad_norm": 0.13240846991539001,
|
| 409 |
+
"learning_rate": 2.2706183876134045e-06,
|
| 410 |
+
"loss": 0.0083,
|
| 411 |
+
"step": 275
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 9.336134453781513,
|
| 415 |
+
"grad_norm": 0.12592285871505737,
|
| 416 |
+
"learning_rate": 1.4852136862001764e-06,
|
| 417 |
+
"loss": 0.0092,
|
| 418 |
+
"step": 280
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 9.504201680672269,
|
| 422 |
+
"grad_norm": 0.10958818346261978,
|
| 423 |
+
"learning_rate": 8.639676646793382e-07,
|
| 424 |
+
"loss": 0.0076,
|
| 425 |
+
"step": 285
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 9.672268907563025,
|
| 429 |
+
"grad_norm": 0.11822710186243057,
|
| 430 |
+
"learning_rate": 4.089824229369155e-07,
|
| 431 |
+
"loss": 0.0093,
|
| 432 |
+
"step": 290
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 9.840336134453782,
|
| 436 |
+
"grad_norm": 0.14177332818508148,
|
| 437 |
+
"learning_rate": 1.2179748700879012e-07,
|
| 438 |
+
"loss": 0.0087,
|
| 439 |
+
"step": 295
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 10.0,
|
| 443 |
+
"grad_norm": 0.15116795897483826,
|
| 444 |
+
"learning_rate": 3.384599811889766e-09,
|
| 445 |
+
"loss": 0.0095,
|
| 446 |
+
"step": 300
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 10.0,
|
| 450 |
+
"eval_loss": 0.1007455587387085,
|
| 451 |
+
"eval_runtime": 1.6871,
|
| 452 |
+
"eval_samples_per_second": 11.855,
|
| 453 |
+
"eval_steps_per_second": 11.855,
|
| 454 |
+
"step": 300
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 10.0,
|
| 458 |
+
"step": 300,
|
| 459 |
+
"total_flos": 3.884520534294528e+16,
|
| 460 |
+
"train_loss": 0.18234332187722127,
|
| 461 |
+
"train_runtime": 611.2552,
|
| 462 |
+
"train_samples_per_second": 31.084,
|
| 463 |
+
"train_steps_per_second": 0.491
|
| 464 |
+
}
|
| 465 |
+
],
|
| 466 |
+
"logging_steps": 5,
|
| 467 |
+
"max_steps": 300,
|
| 468 |
+
"num_input_tokens_seen": 0,
|
| 469 |
+
"num_train_epochs": 10,
|
| 470 |
+
"save_steps": 100,
|
| 471 |
+
"stateful_callbacks": {
|
| 472 |
+
"TrainerControl": {
|
| 473 |
+
"args": {
|
| 474 |
+
"should_epoch_stop": false,
|
| 475 |
+
"should_evaluate": false,
|
| 476 |
+
"should_log": false,
|
| 477 |
+
"should_save": true,
|
| 478 |
+
"should_training_stop": true
|
| 479 |
+
},
|
| 480 |
+
"attributes": {}
|
| 481 |
+
}
|
| 482 |
+
},
|
| 483 |
+
"total_flos": 3.884520534294528e+16,
|
| 484 |
+
"train_batch_size": 16,
|
| 485 |
+
"trial_name": null,
|
| 486 |
+
"trial_params": null
|
| 487 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|