End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +780 -0
README.md
CHANGED
|
@@ -33,7 +33,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
- Accuracy: 0.8897
|
| 38 |
|
| 39 |
## Model description
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2836
|
| 37 |
- Accuracy: 0.8897
|
| 38 |
|
| 39 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.15384615384615,
|
| 3 |
+
"eval_accuracy": 0.8897058823529411,
|
| 4 |
+
"eval_loss": 0.28356170654296875,
|
| 5 |
+
"eval_runtime": 34.2358,
|
| 6 |
+
"eval_samples_per_second": 3.972,
|
| 7 |
+
"eval_steps_per_second": 0.146,
|
| 8 |
+
"total_flos": 1.4197680152537334e+18,
|
| 9 |
+
"train_loss": 0.6011993959214952,
|
| 10 |
+
"train_runtime": 12614.4578,
|
| 11 |
+
"train_samples_per_second": 4.852,
|
| 12 |
+
"train_steps_per_second": 0.036
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.15384615384615,
|
| 3 |
+
"eval_accuracy": 0.8897058823529411,
|
| 4 |
+
"eval_loss": 0.28356170654296875,
|
| 5 |
+
"eval_runtime": 34.2358,
|
| 6 |
+
"eval_samples_per_second": 3.972,
|
| 7 |
+
"eval_steps_per_second": 0.146
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.15384615384615,
|
| 3 |
+
"total_flos": 1.4197680152537334e+18,
|
| 4 |
+
"train_loss": 0.6011993959214952,
|
| 5 |
+
"train_runtime": 12614.4578,
|
| 6 |
+
"train_samples_per_second": 4.852,
|
| 7 |
+
"train_steps_per_second": 0.036
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,780 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8897058823529411,
|
| 3 |
+
"best_model_checkpoint": "convnext-tiny-224-finetuned-papsmear/checkpoint-399",
|
| 4 |
+
"epoch": 46.15384615384615,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 450,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9230769230769231,
|
| 13 |
+
"eval_accuracy": 0.16911764705882354,
|
| 14 |
+
"eval_loss": 1.780842900276184,
|
| 15 |
+
"eval_runtime": 28.9972,
|
| 16 |
+
"eval_samples_per_second": 4.69,
|
| 17 |
+
"eval_steps_per_second": 0.172,
|
| 18 |
+
"step": 9
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0256410256410255,
|
| 22 |
+
"grad_norm": 2.4460816383361816,
|
| 23 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 24 |
+
"loss": 1.8057,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.9487179487179487,
|
| 29 |
+
"eval_accuracy": 0.33088235294117646,
|
| 30 |
+
"eval_loss": 1.680767297744751,
|
| 31 |
+
"eval_runtime": 28.8937,
|
| 32 |
+
"eval_samples_per_second": 4.707,
|
| 33 |
+
"eval_steps_per_second": 0.173,
|
| 34 |
+
"step": 19
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.051282051282051,
|
| 38 |
+
"grad_norm": 2.7383179664611816,
|
| 39 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 40 |
+
"loss": 1.7394,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.9743589743589745,
|
| 45 |
+
"eval_accuracy": 0.3382352941176471,
|
| 46 |
+
"eval_loss": 1.5824888944625854,
|
| 47 |
+
"eval_runtime": 29.5183,
|
| 48 |
+
"eval_samples_per_second": 4.607,
|
| 49 |
+
"eval_steps_per_second": 0.169,
|
| 50 |
+
"step": 29
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.076923076923077,
|
| 54 |
+
"grad_norm": 2.6751656532287598,
|
| 55 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 56 |
+
"loss": 1.6408,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"eval_accuracy": 0.375,
|
| 62 |
+
"eval_loss": 1.4575930833816528,
|
| 63 |
+
"eval_runtime": 29.917,
|
| 64 |
+
"eval_samples_per_second": 4.546,
|
| 65 |
+
"eval_steps_per_second": 0.167,
|
| 66 |
+
"step": 39
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.102564102564102,
|
| 70 |
+
"grad_norm": 2.3050172328948975,
|
| 71 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 72 |
+
"loss": 1.5428,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.923076923076923,
|
| 77 |
+
"eval_accuracy": 0.5220588235294118,
|
| 78 |
+
"eval_loss": 1.328136682510376,
|
| 79 |
+
"eval_runtime": 29.7396,
|
| 80 |
+
"eval_samples_per_second": 4.573,
|
| 81 |
+
"eval_steps_per_second": 0.168,
|
| 82 |
+
"step": 48
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.128205128205128,
|
| 86 |
+
"grad_norm": 3.6692676544189453,
|
| 87 |
+
"learning_rate": 4.938271604938271e-05,
|
| 88 |
+
"loss": 1.3931,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.948717948717949,
|
| 93 |
+
"eval_accuracy": 0.5588235294117647,
|
| 94 |
+
"eval_loss": 1.2043910026550293,
|
| 95 |
+
"eval_runtime": 29.8682,
|
| 96 |
+
"eval_samples_per_second": 4.553,
|
| 97 |
+
"eval_steps_per_second": 0.167,
|
| 98 |
+
"step": 58
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.153846153846154,
|
| 102 |
+
"grad_norm": 5.583779811859131,
|
| 103 |
+
"learning_rate": 4.814814814814815e-05,
|
| 104 |
+
"loss": 1.2669,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.9743589743589745,
|
| 109 |
+
"eval_accuracy": 0.6102941176470589,
|
| 110 |
+
"eval_loss": 1.0755624771118164,
|
| 111 |
+
"eval_runtime": 29.2586,
|
| 112 |
+
"eval_samples_per_second": 4.648,
|
| 113 |
+
"eval_steps_per_second": 0.171,
|
| 114 |
+
"step": 68
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.17948717948718,
|
| 118 |
+
"grad_norm": 4.539360046386719,
|
| 119 |
+
"learning_rate": 4.691358024691358e-05,
|
| 120 |
+
"loss": 1.1355,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"eval_accuracy": 0.6323529411764706,
|
| 126 |
+
"eval_loss": 0.9844695925712585,
|
| 127 |
+
"eval_runtime": 29.1373,
|
| 128 |
+
"eval_samples_per_second": 4.668,
|
| 129 |
+
"eval_steps_per_second": 0.172,
|
| 130 |
+
"step": 78
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 8.205128205128204,
|
| 134 |
+
"grad_norm": 4.369154453277588,
|
| 135 |
+
"learning_rate": 4.567901234567901e-05,
|
| 136 |
+
"loss": 1.0379,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 8.923076923076923,
|
| 141 |
+
"eval_accuracy": 0.6617647058823529,
|
| 142 |
+
"eval_loss": 0.9259934425354004,
|
| 143 |
+
"eval_runtime": 28.1315,
|
| 144 |
+
"eval_samples_per_second": 4.834,
|
| 145 |
+
"eval_steps_per_second": 0.178,
|
| 146 |
+
"step": 87
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 9.23076923076923,
|
| 150 |
+
"grad_norm": 9.379680633544922,
|
| 151 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 152 |
+
"loss": 0.9571,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 9.948717948717949,
|
| 157 |
+
"eval_accuracy": 0.6617647058823529,
|
| 158 |
+
"eval_loss": 0.8539159893989563,
|
| 159 |
+
"eval_runtime": 28.427,
|
| 160 |
+
"eval_samples_per_second": 4.784,
|
| 161 |
+
"eval_steps_per_second": 0.176,
|
| 162 |
+
"step": 97
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 10.256410256410255,
|
| 166 |
+
"grad_norm": 5.348514080047607,
|
| 167 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 168 |
+
"loss": 0.8376,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 10.974358974358974,
|
| 173 |
+
"eval_accuracy": 0.7279411764705882,
|
| 174 |
+
"eval_loss": 0.7998486757278442,
|
| 175 |
+
"eval_runtime": 27.68,
|
| 176 |
+
"eval_samples_per_second": 4.913,
|
| 177 |
+
"eval_steps_per_second": 0.181,
|
| 178 |
+
"step": 107
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 11.282051282051283,
|
| 182 |
+
"grad_norm": 5.533881664276123,
|
| 183 |
+
"learning_rate": 4.197530864197531e-05,
|
| 184 |
+
"loss": 0.7942,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"eval_accuracy": 0.75,
|
| 190 |
+
"eval_loss": 0.757344663143158,
|
| 191 |
+
"eval_runtime": 28.6943,
|
| 192 |
+
"eval_samples_per_second": 4.74,
|
| 193 |
+
"eval_steps_per_second": 0.174,
|
| 194 |
+
"step": 117
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 12.307692307692308,
|
| 198 |
+
"grad_norm": 8.20283031463623,
|
| 199 |
+
"learning_rate": 4.074074074074074e-05,
|
| 200 |
+
"loss": 0.7095,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 12.923076923076923,
|
| 205 |
+
"eval_accuracy": 0.7426470588235294,
|
| 206 |
+
"eval_loss": 0.7004891633987427,
|
| 207 |
+
"eval_runtime": 27.9492,
|
| 208 |
+
"eval_samples_per_second": 4.866,
|
| 209 |
+
"eval_steps_per_second": 0.179,
|
| 210 |
+
"step": 126
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.333333333333334,
|
| 214 |
+
"grad_norm": 7.526528358459473,
|
| 215 |
+
"learning_rate": 3.950617283950617e-05,
|
| 216 |
+
"loss": 0.7022,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 13.948717948717949,
|
| 221 |
+
"eval_accuracy": 0.7867647058823529,
|
| 222 |
+
"eval_loss": 0.6833786964416504,
|
| 223 |
+
"eval_runtime": 28.5203,
|
| 224 |
+
"eval_samples_per_second": 4.769,
|
| 225 |
+
"eval_steps_per_second": 0.175,
|
| 226 |
+
"step": 136
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 14.35897435897436,
|
| 230 |
+
"grad_norm": 8.826446533203125,
|
| 231 |
+
"learning_rate": 3.82716049382716e-05,
|
| 232 |
+
"loss": 0.6504,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.974358974358974,
|
| 237 |
+
"eval_accuracy": 0.7720588235294118,
|
| 238 |
+
"eval_loss": 0.6552230715751648,
|
| 239 |
+
"eval_runtime": 28.1583,
|
| 240 |
+
"eval_samples_per_second": 4.83,
|
| 241 |
+
"eval_steps_per_second": 0.178,
|
| 242 |
+
"step": 146
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 15.384615384615385,
|
| 246 |
+
"grad_norm": 11.680583953857422,
|
| 247 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 248 |
+
"loss": 0.589,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"eval_accuracy": 0.8014705882352942,
|
| 254 |
+
"eval_loss": 0.6191554665565491,
|
| 255 |
+
"eval_runtime": 28.3476,
|
| 256 |
+
"eval_samples_per_second": 4.798,
|
| 257 |
+
"eval_steps_per_second": 0.176,
|
| 258 |
+
"step": 156
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 16.41025641025641,
|
| 262 |
+
"grad_norm": 7.944546699523926,
|
| 263 |
+
"learning_rate": 3.580246913580247e-05,
|
| 264 |
+
"loss": 0.5679,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 16.923076923076923,
|
| 269 |
+
"eval_accuracy": 0.8088235294117647,
|
| 270 |
+
"eval_loss": 0.5738231539726257,
|
| 271 |
+
"eval_runtime": 28.4094,
|
| 272 |
+
"eval_samples_per_second": 4.787,
|
| 273 |
+
"eval_steps_per_second": 0.176,
|
| 274 |
+
"step": 165
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 17.435897435897434,
|
| 278 |
+
"grad_norm": 13.33797550201416,
|
| 279 |
+
"learning_rate": 3.45679012345679e-05,
|
| 280 |
+
"loss": 0.5236,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 17.94871794871795,
|
| 285 |
+
"eval_accuracy": 0.8014705882352942,
|
| 286 |
+
"eval_loss": 0.5617075562477112,
|
| 287 |
+
"eval_runtime": 27.45,
|
| 288 |
+
"eval_samples_per_second": 4.954,
|
| 289 |
+
"eval_steps_per_second": 0.182,
|
| 290 |
+
"step": 175
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 18.46153846153846,
|
| 294 |
+
"grad_norm": 9.086050987243652,
|
| 295 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 296 |
+
"loss": 0.5244,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 18.974358974358974,
|
| 301 |
+
"eval_accuracy": 0.8235294117647058,
|
| 302 |
+
"eval_loss": 0.5072647333145142,
|
| 303 |
+
"eval_runtime": 27.8398,
|
| 304 |
+
"eval_samples_per_second": 4.885,
|
| 305 |
+
"eval_steps_per_second": 0.18,
|
| 306 |
+
"step": 185
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 19.487179487179485,
|
| 310 |
+
"grad_norm": 17.615938186645508,
|
| 311 |
+
"learning_rate": 3.209876543209876e-05,
|
| 312 |
+
"loss": 0.4781,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"eval_accuracy": 0.8161764705882353,
|
| 318 |
+
"eval_loss": 0.5111706256866455,
|
| 319 |
+
"eval_runtime": 28.0372,
|
| 320 |
+
"eval_samples_per_second": 4.851,
|
| 321 |
+
"eval_steps_per_second": 0.178,
|
| 322 |
+
"step": 195
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 20.51282051282051,
|
| 326 |
+
"grad_norm": 8.245569229125977,
|
| 327 |
+
"learning_rate": 3.08641975308642e-05,
|
| 328 |
+
"loss": 0.453,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.923076923076923,
|
| 333 |
+
"eval_accuracy": 0.8235294117647058,
|
| 334 |
+
"eval_loss": 0.46496695280075073,
|
| 335 |
+
"eval_runtime": 27.566,
|
| 336 |
+
"eval_samples_per_second": 4.934,
|
| 337 |
+
"eval_steps_per_second": 0.181,
|
| 338 |
+
"step": 204
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 21.53846153846154,
|
| 342 |
+
"grad_norm": 29.225616455078125,
|
| 343 |
+
"learning_rate": 2.962962962962963e-05,
|
| 344 |
+
"loss": 0.4544,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 21.94871794871795,
|
| 349 |
+
"eval_accuracy": 0.8455882352941176,
|
| 350 |
+
"eval_loss": 0.45913681387901306,
|
| 351 |
+
"eval_runtime": 28.694,
|
| 352 |
+
"eval_samples_per_second": 4.74,
|
| 353 |
+
"eval_steps_per_second": 0.174,
|
| 354 |
+
"step": 214
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 22.564102564102566,
|
| 358 |
+
"grad_norm": 11.300565719604492,
|
| 359 |
+
"learning_rate": 2.839506172839506e-05,
|
| 360 |
+
"loss": 0.419,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 22.974358974358974,
|
| 365 |
+
"eval_accuracy": 0.8308823529411765,
|
| 366 |
+
"eval_loss": 0.4402587115764618,
|
| 367 |
+
"eval_runtime": 27.5811,
|
| 368 |
+
"eval_samples_per_second": 4.931,
|
| 369 |
+
"eval_steps_per_second": 0.181,
|
| 370 |
+
"step": 224
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 23.58974358974359,
|
| 374 |
+
"grad_norm": 24.8338680267334,
|
| 375 |
+
"learning_rate": 2.7160493827160493e-05,
|
| 376 |
+
"loss": 0.4146,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"eval_accuracy": 0.8382352941176471,
|
| 382 |
+
"eval_loss": 0.4292435348033905,
|
| 383 |
+
"eval_runtime": 28.0224,
|
| 384 |
+
"eval_samples_per_second": 4.853,
|
| 385 |
+
"eval_steps_per_second": 0.178,
|
| 386 |
+
"step": 234
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 24.615384615384617,
|
| 390 |
+
"grad_norm": 33.230125427246094,
|
| 391 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 392 |
+
"loss": 0.398,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 24.923076923076923,
|
| 397 |
+
"eval_accuracy": 0.8382352941176471,
|
| 398 |
+
"eval_loss": 0.4315250515937805,
|
| 399 |
+
"eval_runtime": 27.9117,
|
| 400 |
+
"eval_samples_per_second": 4.873,
|
| 401 |
+
"eval_steps_per_second": 0.179,
|
| 402 |
+
"step": 243
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 25.641025641025642,
|
| 406 |
+
"grad_norm": 11.8997802734375,
|
| 407 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 408 |
+
"loss": 0.3918,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 25.94871794871795,
|
| 413 |
+
"eval_accuracy": 0.8676470588235294,
|
| 414 |
+
"eval_loss": 0.39800333976745605,
|
| 415 |
+
"eval_runtime": 27.1303,
|
| 416 |
+
"eval_samples_per_second": 5.013,
|
| 417 |
+
"eval_steps_per_second": 0.184,
|
| 418 |
+
"step": 253
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 26.666666666666668,
|
| 422 |
+
"grad_norm": 16.35055923461914,
|
| 423 |
+
"learning_rate": 2.345679012345679e-05,
|
| 424 |
+
"loss": 0.361,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 26.974358974358974,
|
| 429 |
+
"eval_accuracy": 0.8602941176470589,
|
| 430 |
+
"eval_loss": 0.37583470344543457,
|
| 431 |
+
"eval_runtime": 27.9321,
|
| 432 |
+
"eval_samples_per_second": 4.869,
|
| 433 |
+
"eval_steps_per_second": 0.179,
|
| 434 |
+
"step": 263
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 27.692307692307693,
|
| 438 |
+
"grad_norm": 7.5509724617004395,
|
| 439 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 440 |
+
"loss": 0.3355,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"eval_accuracy": 0.8602941176470589,
|
| 446 |
+
"eval_loss": 0.36569300293922424,
|
| 447 |
+
"eval_runtime": 27.4222,
|
| 448 |
+
"eval_samples_per_second": 4.959,
|
| 449 |
+
"eval_steps_per_second": 0.182,
|
| 450 |
+
"step": 273
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 28.71794871794872,
|
| 454 |
+
"grad_norm": 16.6190128326416,
|
| 455 |
+
"learning_rate": 2.0987654320987655e-05,
|
| 456 |
+
"loss": 0.3483,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 28.923076923076923,
|
| 461 |
+
"eval_accuracy": 0.875,
|
| 462 |
+
"eval_loss": 0.366921603679657,
|
| 463 |
+
"eval_runtime": 27.8307,
|
| 464 |
+
"eval_samples_per_second": 4.887,
|
| 465 |
+
"eval_steps_per_second": 0.18,
|
| 466 |
+
"step": 282
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 29.743589743589745,
|
| 470 |
+
"grad_norm": 14.241522789001465,
|
| 471 |
+
"learning_rate": 1.9753086419753087e-05,
|
| 472 |
+
"loss": 0.3171,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 29.94871794871795,
|
| 477 |
+
"eval_accuracy": 0.8602941176470589,
|
| 478 |
+
"eval_loss": 0.3492419719696045,
|
| 479 |
+
"eval_runtime": 27.0548,
|
| 480 |
+
"eval_samples_per_second": 5.027,
|
| 481 |
+
"eval_steps_per_second": 0.185,
|
| 482 |
+
"step": 292
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 30.76923076923077,
|
| 486 |
+
"grad_norm": 13.727191925048828,
|
| 487 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 488 |
+
"loss": 0.3249,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 30.974358974358974,
|
| 493 |
+
"eval_accuracy": 0.875,
|
| 494 |
+
"eval_loss": 0.3399755358695984,
|
| 495 |
+
"eval_runtime": 27.5478,
|
| 496 |
+
"eval_samples_per_second": 4.937,
|
| 497 |
+
"eval_steps_per_second": 0.182,
|
| 498 |
+
"step": 302
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 31.794871794871796,
|
| 502 |
+
"grad_norm": 16.1718692779541,
|
| 503 |
+
"learning_rate": 1.728395061728395e-05,
|
| 504 |
+
"loss": 0.3087,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 32.0,
|
| 509 |
+
"eval_accuracy": 0.875,
|
| 510 |
+
"eval_loss": 0.3251068592071533,
|
| 511 |
+
"eval_runtime": 28.8731,
|
| 512 |
+
"eval_samples_per_second": 4.71,
|
| 513 |
+
"eval_steps_per_second": 0.173,
|
| 514 |
+
"step": 312
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 32.82051282051282,
|
| 518 |
+
"grad_norm": 14.378969192504883,
|
| 519 |
+
"learning_rate": 1.604938271604938e-05,
|
| 520 |
+
"loss": 0.3029,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 32.92307692307692,
|
| 525 |
+
"eval_accuracy": 0.8823529411764706,
|
| 526 |
+
"eval_loss": 0.31666299700737,
|
| 527 |
+
"eval_runtime": 27.475,
|
| 528 |
+
"eval_samples_per_second": 4.95,
|
| 529 |
+
"eval_steps_per_second": 0.182,
|
| 530 |
+
"step": 321
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 33.84615384615385,
|
| 534 |
+
"grad_norm": 12.18956470489502,
|
| 535 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 536 |
+
"loss": 0.3018,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 33.94871794871795,
|
| 541 |
+
"eval_accuracy": 0.875,
|
| 542 |
+
"eval_loss": 0.3191751539707184,
|
| 543 |
+
"eval_runtime": 27.7647,
|
| 544 |
+
"eval_samples_per_second": 4.898,
|
| 545 |
+
"eval_steps_per_second": 0.18,
|
| 546 |
+
"step": 331
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 34.87179487179487,
|
| 550 |
+
"grad_norm": 12.676053047180176,
|
| 551 |
+
"learning_rate": 1.3580246913580247e-05,
|
| 552 |
+
"loss": 0.2823,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 34.97435897435897,
|
| 557 |
+
"eval_accuracy": 0.875,
|
| 558 |
+
"eval_loss": 0.3066389560699463,
|
| 559 |
+
"eval_runtime": 27.6676,
|
| 560 |
+
"eval_samples_per_second": 4.916,
|
| 561 |
+
"eval_steps_per_second": 0.181,
|
| 562 |
+
"step": 341
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 35.8974358974359,
|
| 566 |
+
"grad_norm": 10.490346908569336,
|
| 567 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 568 |
+
"loss": 0.2744,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 36.0,
|
| 573 |
+
"eval_accuracy": 0.875,
|
| 574 |
+
"eval_loss": 0.300258070230484,
|
| 575 |
+
"eval_runtime": 29.0445,
|
| 576 |
+
"eval_samples_per_second": 4.682,
|
| 577 |
+
"eval_steps_per_second": 0.172,
|
| 578 |
+
"step": 351
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 36.92307692307692,
|
| 582 |
+
"grad_norm": 17.77071189880371,
|
| 583 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 584 |
+
"loss": 0.258,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 36.92307692307692,
|
| 589 |
+
"eval_accuracy": 0.875,
|
| 590 |
+
"eval_loss": 0.2963866591453552,
|
| 591 |
+
"eval_runtime": 27.8633,
|
| 592 |
+
"eval_samples_per_second": 4.881,
|
| 593 |
+
"eval_steps_per_second": 0.179,
|
| 594 |
+
"step": 360
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 37.94871794871795,
|
| 598 |
+
"grad_norm": 14.726465225219727,
|
| 599 |
+
"learning_rate": 9.876543209876543e-06,
|
| 600 |
+
"loss": 0.2714,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 37.94871794871795,
|
| 605 |
+
"eval_accuracy": 0.875,
|
| 606 |
+
"eval_loss": 0.3039358854293823,
|
| 607 |
+
"eval_runtime": 27.3906,
|
| 608 |
+
"eval_samples_per_second": 4.965,
|
| 609 |
+
"eval_steps_per_second": 0.183,
|
| 610 |
+
"step": 370
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 38.97435897435897,
|
| 614 |
+
"grad_norm": 5.834155082702637,
|
| 615 |
+
"learning_rate": 8.641975308641975e-06,
|
| 616 |
+
"loss": 0.2486,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 38.97435897435897,
|
| 621 |
+
"eval_accuracy": 0.875,
|
| 622 |
+
"eval_loss": 0.29370391368865967,
|
| 623 |
+
"eval_runtime": 27.4875,
|
| 624 |
+
"eval_samples_per_second": 4.948,
|
| 625 |
+
"eval_steps_per_second": 0.182,
|
| 626 |
+
"step": 380
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 40.0,
|
| 630 |
+
"grad_norm": 48.52106857299805,
|
| 631 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 632 |
+
"loss": 0.2511,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 40.0,
|
| 637 |
+
"eval_accuracy": 0.8823529411764706,
|
| 638 |
+
"eval_loss": 0.27393513917922974,
|
| 639 |
+
"eval_runtime": 27.5378,
|
| 640 |
+
"eval_samples_per_second": 4.939,
|
| 641 |
+
"eval_steps_per_second": 0.182,
|
| 642 |
+
"step": 390
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 40.92307692307692,
|
| 646 |
+
"eval_accuracy": 0.8897058823529411,
|
| 647 |
+
"eval_loss": 0.28356170654296875,
|
| 648 |
+
"eval_runtime": 27.6475,
|
| 649 |
+
"eval_samples_per_second": 4.919,
|
| 650 |
+
"eval_steps_per_second": 0.181,
|
| 651 |
+
"step": 399
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 41.02564102564103,
|
| 655 |
+
"grad_norm": 12.991565704345703,
|
| 656 |
+
"learning_rate": 6.172839506172839e-06,
|
| 657 |
+
"loss": 0.2659,
|
| 658 |
+
"step": 400
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 41.94871794871795,
|
| 662 |
+
"eval_accuracy": 0.875,
|
| 663 |
+
"eval_loss": 0.2804461717605591,
|
| 664 |
+
"eval_runtime": 28.1041,
|
| 665 |
+
"eval_samples_per_second": 4.839,
|
| 666 |
+
"eval_steps_per_second": 0.178,
|
| 667 |
+
"step": 409
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 42.05128205128205,
|
| 671 |
+
"grad_norm": 10.837814331054688,
|
| 672 |
+
"learning_rate": 4.938271604938272e-06,
|
| 673 |
+
"loss": 0.2379,
|
| 674 |
+
"step": 410
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 42.97435897435897,
|
| 678 |
+
"eval_accuracy": 0.8823529411764706,
|
| 679 |
+
"eval_loss": 0.2746768891811371,
|
| 680 |
+
"eval_runtime": 27.287,
|
| 681 |
+
"eval_samples_per_second": 4.984,
|
| 682 |
+
"eval_steps_per_second": 0.183,
|
| 683 |
+
"step": 419
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 43.07692307692308,
|
| 687 |
+
"grad_norm": 21.34435272216797,
|
| 688 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 689 |
+
"loss": 0.2279,
|
| 690 |
+
"step": 420
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 44.0,
|
| 694 |
+
"eval_accuracy": 0.8897058823529411,
|
| 695 |
+
"eval_loss": 0.27258822321891785,
|
| 696 |
+
"eval_runtime": 27.4605,
|
| 697 |
+
"eval_samples_per_second": 4.953,
|
| 698 |
+
"eval_steps_per_second": 0.182,
|
| 699 |
+
"step": 429
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 44.1025641025641,
|
| 703 |
+
"grad_norm": 15.666738510131836,
|
| 704 |
+
"learning_rate": 2.469135802469136e-06,
|
| 705 |
+
"loss": 0.2153,
|
| 706 |
+
"step": 430
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 44.92307692307692,
|
| 710 |
+
"eval_accuracy": 0.8897058823529411,
|
| 711 |
+
"eval_loss": 0.27316194772720337,
|
| 712 |
+
"eval_runtime": 28.0117,
|
| 713 |
+
"eval_samples_per_second": 4.855,
|
| 714 |
+
"eval_steps_per_second": 0.178,
|
| 715 |
+
"step": 438
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 45.12820512820513,
|
| 719 |
+
"grad_norm": 9.311797142028809,
|
| 720 |
+
"learning_rate": 1.234567901234568e-06,
|
| 721 |
+
"loss": 0.2461,
|
| 722 |
+
"step": 440
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 45.94871794871795,
|
| 726 |
+
"eval_accuracy": 0.8897058823529411,
|
| 727 |
+
"eval_loss": 0.27384060621261597,
|
| 728 |
+
"eval_runtime": 28.0924,
|
| 729 |
+
"eval_samples_per_second": 4.841,
|
| 730 |
+
"eval_steps_per_second": 0.178,
|
| 731 |
+
"step": 448
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 46.15384615384615,
|
| 735 |
+
"grad_norm": 23.28314971923828,
|
| 736 |
+
"learning_rate": 0.0,
|
| 737 |
+
"loss": 0.2482,
|
| 738 |
+
"step": 450
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 46.15384615384615,
|
| 742 |
+
"eval_accuracy": 0.8897058823529411,
|
| 743 |
+
"eval_loss": 0.2737831771373749,
|
| 744 |
+
"eval_runtime": 28.0709,
|
| 745 |
+
"eval_samples_per_second": 4.845,
|
| 746 |
+
"eval_steps_per_second": 0.178,
|
| 747 |
+
"step": 450
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 46.15384615384615,
|
| 751 |
+
"step": 450,
|
| 752 |
+
"total_flos": 1.4197680152537334e+18,
|
| 753 |
+
"train_loss": 0.6011993959214952,
|
| 754 |
+
"train_runtime": 12614.4578,
|
| 755 |
+
"train_samples_per_second": 4.852,
|
| 756 |
+
"train_steps_per_second": 0.036
|
| 757 |
+
}
|
| 758 |
+
],
|
| 759 |
+
"logging_steps": 10,
|
| 760 |
+
"max_steps": 450,
|
| 761 |
+
"num_input_tokens_seen": 0,
|
| 762 |
+
"num_train_epochs": 50,
|
| 763 |
+
"save_steps": 500,
|
| 764 |
+
"stateful_callbacks": {
|
| 765 |
+
"TrainerControl": {
|
| 766 |
+
"args": {
|
| 767 |
+
"should_epoch_stop": false,
|
| 768 |
+
"should_evaluate": false,
|
| 769 |
+
"should_log": false,
|
| 770 |
+
"should_save": true,
|
| 771 |
+
"should_training_stop": true
|
| 772 |
+
},
|
| 773 |
+
"attributes": {}
|
| 774 |
+
}
|
| 775 |
+
},
|
| 776 |
+
"total_flos": 1.4197680152537334e+18,
|
| 777 |
+
"train_batch_size": 32,
|
| 778 |
+
"trial_name": null,
|
| 779 |
+
"trial_params": null
|
| 780 |
+
}
|