Commit
•
f559979
1
Parent(s):
4f76b64
End of training
Browse files- README.md +1 -1
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/May24_10-16-54_DESKTOP-SKBE9FB/events.out.tfevents.1716569140.DESKTOP-SKBE9FB.4252.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +870 -0
README.md
CHANGED
|
@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
- Accuracy: 0.8235
|
| 37 |
|
| 38 |
## Model description
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.5609
|
| 36 |
- Accuracy: 0.8235
|
| 37 |
|
| 38 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.8235294117647058,
|
| 4 |
+
"eval_loss": 0.5609031319618225,
|
| 5 |
+
"eval_runtime": 0.9276,
|
| 6 |
+
"eval_samples_per_second": 54.98,
|
| 7 |
+
"eval_steps_per_second": 2.156,
|
| 8 |
+
"train_loss": 0.1865533402003348,
|
| 9 |
+
"train_runtime": 1567.0918,
|
| 10 |
+
"train_samples_per_second": 64.833,
|
| 11 |
+
"train_steps_per_second": 0.51
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.8235294117647058,
|
| 4 |
+
"eval_loss": 0.5609031319618225,
|
| 5 |
+
"eval_runtime": 0.9276,
|
| 6 |
+
"eval_samples_per_second": 54.98,
|
| 7 |
+
"eval_steps_per_second": 2.156
|
| 8 |
+
}
|
runs/May24_10-16-54_DESKTOP-SKBE9FB/events.out.tfevents.1716569140.DESKTOP-SKBE9FB.4252.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6bd4986088b77c1bf2ccba869e522ca120ad36e18afae819999837fbadc2ee7e
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"train_loss": 0.1865533402003348,
|
| 4 |
+
"train_runtime": 1567.0918,
|
| 5 |
+
"train_samples_per_second": 64.833,
|
| 6 |
+
"train_steps_per_second": 0.51
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,870 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8235294117647058,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-U8-40c\\checkpoint-120",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.5,
|
| 13 |
+
"learning_rate": 1.25e-05,
|
| 14 |
+
"loss": 1.3827,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"learning_rate": 2.5e-05,
|
| 20 |
+
"loss": 1.3495,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 1.0,
|
| 25 |
+
"eval_accuracy": 0.47058823529411764,
|
| 26 |
+
"eval_loss": 1.3142259120941162,
|
| 27 |
+
"eval_runtime": 0.9121,
|
| 28 |
+
"eval_samples_per_second": 55.917,
|
| 29 |
+
"eval_steps_per_second": 2.193,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.5,
|
| 34 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 35 |
+
"loss": 1.2825,
|
| 36 |
+
"step": 30
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 2.0,
|
| 40 |
+
"learning_rate": 5e-05,
|
| 41 |
+
"loss": 1.1689,
|
| 42 |
+
"step": 40
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 2.0,
|
| 46 |
+
"eval_accuracy": 0.5686274509803921,
|
| 47 |
+
"eval_loss": 1.115267038345337,
|
| 48 |
+
"eval_runtime": 0.8177,
|
| 49 |
+
"eval_samples_per_second": 62.373,
|
| 50 |
+
"eval_steps_per_second": 2.446,
|
| 51 |
+
"step": 40
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 2.5,
|
| 55 |
+
"learning_rate": 4.9342105263157894e-05,
|
| 56 |
+
"loss": 1.0345,
|
| 57 |
+
"step": 50
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 3.0,
|
| 61 |
+
"learning_rate": 4.868421052631579e-05,
|
| 62 |
+
"loss": 0.8673,
|
| 63 |
+
"step": 60
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 3.0,
|
| 67 |
+
"eval_accuracy": 0.6666666666666666,
|
| 68 |
+
"eval_loss": 0.8497752547264099,
|
| 69 |
+
"eval_runtime": 0.8953,
|
| 70 |
+
"eval_samples_per_second": 56.964,
|
| 71 |
+
"eval_steps_per_second": 2.234,
|
| 72 |
+
"step": 60
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 3.5,
|
| 76 |
+
"learning_rate": 4.802631578947368e-05,
|
| 77 |
+
"loss": 0.72,
|
| 78 |
+
"step": 70
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.0,
|
| 82 |
+
"learning_rate": 4.736842105263158e-05,
|
| 83 |
+
"loss": 0.5847,
|
| 84 |
+
"step": 80
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"eval_accuracy": 0.7843137254901961,
|
| 89 |
+
"eval_loss": 0.7220016121864319,
|
| 90 |
+
"eval_runtime": 0.8116,
|
| 91 |
+
"eval_samples_per_second": 62.841,
|
| 92 |
+
"eval_steps_per_second": 2.464,
|
| 93 |
+
"step": 80
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.5,
|
| 97 |
+
"learning_rate": 4.671052631578948e-05,
|
| 98 |
+
"loss": 0.4682,
|
| 99 |
+
"step": 90
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 5.0,
|
| 103 |
+
"learning_rate": 4.605263157894737e-05,
|
| 104 |
+
"loss": 0.4029,
|
| 105 |
+
"step": 100
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 5.0,
|
| 109 |
+
"eval_accuracy": 0.6274509803921569,
|
| 110 |
+
"eval_loss": 0.8654420971870422,
|
| 111 |
+
"eval_runtime": 0.817,
|
| 112 |
+
"eval_samples_per_second": 62.426,
|
| 113 |
+
"eval_steps_per_second": 2.448,
|
| 114 |
+
"step": 100
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.5,
|
| 118 |
+
"learning_rate": 4.539473684210527e-05,
|
| 119 |
+
"loss": 0.3146,
|
| 120 |
+
"step": 110
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 6.0,
|
| 124 |
+
"learning_rate": 4.473684210526316e-05,
|
| 125 |
+
"loss": 0.2562,
|
| 126 |
+
"step": 120
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 6.0,
|
| 130 |
+
"eval_accuracy": 0.8235294117647058,
|
| 131 |
+
"eval_loss": 0.5609031319618225,
|
| 132 |
+
"eval_runtime": 0.7989,
|
| 133 |
+
"eval_samples_per_second": 63.835,
|
| 134 |
+
"eval_steps_per_second": 2.503,
|
| 135 |
+
"step": 120
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 6.5,
|
| 139 |
+
"learning_rate": 4.407894736842105e-05,
|
| 140 |
+
"loss": 0.2305,
|
| 141 |
+
"step": 130
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 7.0,
|
| 145 |
+
"learning_rate": 4.342105263157895e-05,
|
| 146 |
+
"loss": 0.2352,
|
| 147 |
+
"step": 140
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 7.0,
|
| 151 |
+
"eval_accuracy": 0.7843137254901961,
|
| 152 |
+
"eval_loss": 0.727162778377533,
|
| 153 |
+
"eval_runtime": 0.8282,
|
| 154 |
+
"eval_samples_per_second": 61.579,
|
| 155 |
+
"eval_steps_per_second": 2.415,
|
| 156 |
+
"step": 140
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 7.5,
|
| 160 |
+
"learning_rate": 4.2763157894736847e-05,
|
| 161 |
+
"loss": 0.2511,
|
| 162 |
+
"step": 150
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 8.0,
|
| 166 |
+
"learning_rate": 4.210526315789474e-05,
|
| 167 |
+
"loss": 0.2131,
|
| 168 |
+
"step": 160
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 8.0,
|
| 172 |
+
"eval_accuracy": 0.7254901960784313,
|
| 173 |
+
"eval_loss": 0.758063793182373,
|
| 174 |
+
"eval_runtime": 0.8058,
|
| 175 |
+
"eval_samples_per_second": 63.289,
|
| 176 |
+
"eval_steps_per_second": 2.482,
|
| 177 |
+
"step": 160
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 8.5,
|
| 181 |
+
"learning_rate": 4.1447368421052636e-05,
|
| 182 |
+
"loss": 0.164,
|
| 183 |
+
"step": 170
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 9.0,
|
| 187 |
+
"learning_rate": 4.078947368421053e-05,
|
| 188 |
+
"loss": 0.1616,
|
| 189 |
+
"step": 180
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 9.0,
|
| 193 |
+
"eval_accuracy": 0.8235294117647058,
|
| 194 |
+
"eval_loss": 0.5436545610427856,
|
| 195 |
+
"eval_runtime": 0.8354,
|
| 196 |
+
"eval_samples_per_second": 61.045,
|
| 197 |
+
"eval_steps_per_second": 2.394,
|
| 198 |
+
"step": 180
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 9.5,
|
| 202 |
+
"learning_rate": 4.0131578947368425e-05,
|
| 203 |
+
"loss": 0.138,
|
| 204 |
+
"step": 190
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 10.0,
|
| 208 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 209 |
+
"loss": 0.1266,
|
| 210 |
+
"step": 200
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 10.0,
|
| 214 |
+
"eval_accuracy": 0.803921568627451,
|
| 215 |
+
"eval_loss": 0.634531557559967,
|
| 216 |
+
"eval_runtime": 0.8867,
|
| 217 |
+
"eval_samples_per_second": 57.514,
|
| 218 |
+
"eval_steps_per_second": 2.255,
|
| 219 |
+
"step": 200
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.5,
|
| 223 |
+
"learning_rate": 3.8815789473684214e-05,
|
| 224 |
+
"loss": 0.116,
|
| 225 |
+
"step": 210
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 11.0,
|
| 229 |
+
"learning_rate": 3.815789473684211e-05,
|
| 230 |
+
"loss": 0.1557,
|
| 231 |
+
"step": 220
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 11.0,
|
| 235 |
+
"eval_accuracy": 0.7647058823529411,
|
| 236 |
+
"eval_loss": 0.8280105590820312,
|
| 237 |
+
"eval_runtime": 0.8226,
|
| 238 |
+
"eval_samples_per_second": 61.997,
|
| 239 |
+
"eval_steps_per_second": 2.431,
|
| 240 |
+
"step": 220
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 11.5,
|
| 244 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 245 |
+
"loss": 0.1116,
|
| 246 |
+
"step": 230
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 12.0,
|
| 250 |
+
"learning_rate": 3.6842105263157895e-05,
|
| 251 |
+
"loss": 0.0871,
|
| 252 |
+
"step": 240
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 12.0,
|
| 256 |
+
"eval_accuracy": 0.7058823529411765,
|
| 257 |
+
"eval_loss": 0.9015989899635315,
|
| 258 |
+
"eval_runtime": 0.7897,
|
| 259 |
+
"eval_samples_per_second": 64.581,
|
| 260 |
+
"eval_steps_per_second": 2.533,
|
| 261 |
+
"step": 240
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 12.5,
|
| 265 |
+
"learning_rate": 3.618421052631579e-05,
|
| 266 |
+
"loss": 0.134,
|
| 267 |
+
"step": 250
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 13.0,
|
| 271 |
+
"learning_rate": 3.5526315789473684e-05,
|
| 272 |
+
"loss": 0.0879,
|
| 273 |
+
"step": 260
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 13.0,
|
| 277 |
+
"eval_accuracy": 0.7647058823529411,
|
| 278 |
+
"eval_loss": 0.8098950982093811,
|
| 279 |
+
"eval_runtime": 0.815,
|
| 280 |
+
"eval_samples_per_second": 62.574,
|
| 281 |
+
"eval_steps_per_second": 2.454,
|
| 282 |
+
"step": 260
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 13.5,
|
| 286 |
+
"learning_rate": 3.4868421052631575e-05,
|
| 287 |
+
"loss": 0.1101,
|
| 288 |
+
"step": 270
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 14.0,
|
| 292 |
+
"learning_rate": 3.421052631578947e-05,
|
| 293 |
+
"loss": 0.0844,
|
| 294 |
+
"step": 280
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 14.0,
|
| 298 |
+
"eval_accuracy": 0.7254901960784313,
|
| 299 |
+
"eval_loss": 0.8790603280067444,
|
| 300 |
+
"eval_runtime": 0.825,
|
| 301 |
+
"eval_samples_per_second": 61.819,
|
| 302 |
+
"eval_steps_per_second": 2.424,
|
| 303 |
+
"step": 280
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 14.5,
|
| 307 |
+
"learning_rate": 3.355263157894737e-05,
|
| 308 |
+
"loss": 0.0758,
|
| 309 |
+
"step": 290
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 15.0,
|
| 313 |
+
"learning_rate": 3.289473684210527e-05,
|
| 314 |
+
"loss": 0.0865,
|
| 315 |
+
"step": 300
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 15.0,
|
| 319 |
+
"eval_accuracy": 0.7843137254901961,
|
| 320 |
+
"eval_loss": 0.9712991118431091,
|
| 321 |
+
"eval_runtime": 0.7949,
|
| 322 |
+
"eval_samples_per_second": 64.158,
|
| 323 |
+
"eval_steps_per_second": 2.516,
|
| 324 |
+
"step": 300
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 15.5,
|
| 328 |
+
"learning_rate": 3.223684210526316e-05,
|
| 329 |
+
"loss": 0.0572,
|
| 330 |
+
"step": 310
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 16.0,
|
| 334 |
+
"learning_rate": 3.157894736842105e-05,
|
| 335 |
+
"loss": 0.1005,
|
| 336 |
+
"step": 320
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 16.0,
|
| 340 |
+
"eval_accuracy": 0.7843137254901961,
|
| 341 |
+
"eval_loss": 0.9965818524360657,
|
| 342 |
+
"eval_runtime": 0.8138,
|
| 343 |
+
"eval_samples_per_second": 62.668,
|
| 344 |
+
"eval_steps_per_second": 2.458,
|
| 345 |
+
"step": 320
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 16.5,
|
| 349 |
+
"learning_rate": 3.092105263157895e-05,
|
| 350 |
+
"loss": 0.0876,
|
| 351 |
+
"step": 330
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 17.0,
|
| 355 |
+
"learning_rate": 3.0263157894736844e-05,
|
| 356 |
+
"loss": 0.0718,
|
| 357 |
+
"step": 340
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 17.0,
|
| 361 |
+
"eval_accuracy": 0.7647058823529411,
|
| 362 |
+
"eval_loss": 1.046757459640503,
|
| 363 |
+
"eval_runtime": 0.7997,
|
| 364 |
+
"eval_samples_per_second": 63.777,
|
| 365 |
+
"eval_steps_per_second": 2.501,
|
| 366 |
+
"step": 340
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 17.5,
|
| 370 |
+
"learning_rate": 2.9605263157894735e-05,
|
| 371 |
+
"loss": 0.0955,
|
| 372 |
+
"step": 350
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 18.0,
|
| 376 |
+
"learning_rate": 2.8947368421052634e-05,
|
| 377 |
+
"loss": 0.0591,
|
| 378 |
+
"step": 360
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 18.0,
|
| 382 |
+
"eval_accuracy": 0.7843137254901961,
|
| 383 |
+
"eval_loss": 0.9471318125724792,
|
| 384 |
+
"eval_runtime": 0.8393,
|
| 385 |
+
"eval_samples_per_second": 60.762,
|
| 386 |
+
"eval_steps_per_second": 2.383,
|
| 387 |
+
"step": 360
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 18.5,
|
| 391 |
+
"learning_rate": 2.8289473684210528e-05,
|
| 392 |
+
"loss": 0.0842,
|
| 393 |
+
"step": 370
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 19.0,
|
| 397 |
+
"learning_rate": 2.7631578947368426e-05,
|
| 398 |
+
"loss": 0.0641,
|
| 399 |
+
"step": 380
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 19.0,
|
| 403 |
+
"eval_accuracy": 0.7450980392156863,
|
| 404 |
+
"eval_loss": 0.9904577136039734,
|
| 405 |
+
"eval_runtime": 0.8249,
|
| 406 |
+
"eval_samples_per_second": 61.825,
|
| 407 |
+
"eval_steps_per_second": 2.425,
|
| 408 |
+
"step": 380
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 19.5,
|
| 412 |
+
"learning_rate": 2.6973684210526317e-05,
|
| 413 |
+
"loss": 0.0738,
|
| 414 |
+
"step": 390
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 20.0,
|
| 418 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 419 |
+
"loss": 0.0542,
|
| 420 |
+
"step": 400
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 20.0,
|
| 424 |
+
"eval_accuracy": 0.7450980392156863,
|
| 425 |
+
"eval_loss": 1.029980182647705,
|
| 426 |
+
"eval_runtime": 0.8442,
|
| 427 |
+
"eval_samples_per_second": 60.41,
|
| 428 |
+
"eval_steps_per_second": 2.369,
|
| 429 |
+
"step": 400
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 20.5,
|
| 433 |
+
"learning_rate": 2.565789473684211e-05,
|
| 434 |
+
"loss": 0.0708,
|
| 435 |
+
"step": 410
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 21.0,
|
| 439 |
+
"learning_rate": 2.5e-05,
|
| 440 |
+
"loss": 0.0813,
|
| 441 |
+
"step": 420
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 21.0,
|
| 445 |
+
"eval_accuracy": 0.7647058823529411,
|
| 446 |
+
"eval_loss": 1.0329537391662598,
|
| 447 |
+
"eval_runtime": 0.819,
|
| 448 |
+
"eval_samples_per_second": 62.27,
|
| 449 |
+
"eval_steps_per_second": 2.442,
|
| 450 |
+
"step": 420
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 21.5,
|
| 454 |
+
"learning_rate": 2.4342105263157896e-05,
|
| 455 |
+
"loss": 0.0638,
|
| 456 |
+
"step": 430
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 22.0,
|
| 460 |
+
"learning_rate": 2.368421052631579e-05,
|
| 461 |
+
"loss": 0.059,
|
| 462 |
+
"step": 440
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 22.0,
|
| 466 |
+
"eval_accuracy": 0.7647058823529411,
|
| 467 |
+
"eval_loss": 0.999519944190979,
|
| 468 |
+
"eval_runtime": 0.8027,
|
| 469 |
+
"eval_samples_per_second": 63.535,
|
| 470 |
+
"eval_steps_per_second": 2.492,
|
| 471 |
+
"step": 440
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 22.5,
|
| 475 |
+
"learning_rate": 2.3026315789473685e-05,
|
| 476 |
+
"loss": 0.0518,
|
| 477 |
+
"step": 450
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 23.0,
|
| 481 |
+
"learning_rate": 2.236842105263158e-05,
|
| 482 |
+
"loss": 0.0679,
|
| 483 |
+
"step": 460
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 23.0,
|
| 487 |
+
"eval_accuracy": 0.7450980392156863,
|
| 488 |
+
"eval_loss": 0.9327283501625061,
|
| 489 |
+
"eval_runtime": 0.8141,
|
| 490 |
+
"eval_samples_per_second": 62.642,
|
| 491 |
+
"eval_steps_per_second": 2.457,
|
| 492 |
+
"step": 460
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 23.5,
|
| 496 |
+
"learning_rate": 2.1710526315789474e-05,
|
| 497 |
+
"loss": 0.087,
|
| 498 |
+
"step": 470
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 24.0,
|
| 502 |
+
"learning_rate": 2.105263157894737e-05,
|
| 503 |
+
"loss": 0.0611,
|
| 504 |
+
"step": 480
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 24.0,
|
| 508 |
+
"eval_accuracy": 0.7647058823529411,
|
| 509 |
+
"eval_loss": 1.0073403120040894,
|
| 510 |
+
"eval_runtime": 0.7891,
|
| 511 |
+
"eval_samples_per_second": 64.627,
|
| 512 |
+
"eval_steps_per_second": 2.534,
|
| 513 |
+
"step": 480
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 24.5,
|
| 517 |
+
"learning_rate": 2.0394736842105264e-05,
|
| 518 |
+
"loss": 0.0475,
|
| 519 |
+
"step": 490
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 25.0,
|
| 523 |
+
"learning_rate": 1.9736842105263158e-05,
|
| 524 |
+
"loss": 0.0694,
|
| 525 |
+
"step": 500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 25.0,
|
| 529 |
+
"eval_accuracy": 0.7647058823529411,
|
| 530 |
+
"eval_loss": 0.9348233342170715,
|
| 531 |
+
"eval_runtime": 0.7727,
|
| 532 |
+
"eval_samples_per_second": 66.0,
|
| 533 |
+
"eval_steps_per_second": 2.588,
|
| 534 |
+
"step": 500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 25.5,
|
| 538 |
+
"learning_rate": 1.9078947368421056e-05,
|
| 539 |
+
"loss": 0.0546,
|
| 540 |
+
"step": 510
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 26.0,
|
| 544 |
+
"learning_rate": 1.8421052631578947e-05,
|
| 545 |
+
"loss": 0.0454,
|
| 546 |
+
"step": 520
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 26.0,
|
| 550 |
+
"eval_accuracy": 0.7843137254901961,
|
| 551 |
+
"eval_loss": 0.8550857901573181,
|
| 552 |
+
"eval_runtime": 0.7842,
|
| 553 |
+
"eval_samples_per_second": 65.031,
|
| 554 |
+
"eval_steps_per_second": 2.55,
|
| 555 |
+
"step": 520
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 26.5,
|
| 559 |
+
"learning_rate": 1.7763157894736842e-05,
|
| 560 |
+
"loss": 0.0503,
|
| 561 |
+
"step": 530
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 27.0,
|
| 565 |
+
"learning_rate": 1.7105263157894737e-05,
|
| 566 |
+
"loss": 0.0536,
|
| 567 |
+
"step": 540
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 27.0,
|
| 571 |
+
"eval_accuracy": 0.7647058823529411,
|
| 572 |
+
"eval_loss": 0.9782336354255676,
|
| 573 |
+
"eval_runtime": 0.8246,
|
| 574 |
+
"eval_samples_per_second": 61.852,
|
| 575 |
+
"eval_steps_per_second": 2.426,
|
| 576 |
+
"step": 540
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 27.5,
|
| 580 |
+
"learning_rate": 1.6447368421052635e-05,
|
| 581 |
+
"loss": 0.0331,
|
| 582 |
+
"step": 550
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 28.0,
|
| 586 |
+
"learning_rate": 1.5789473684210526e-05,
|
| 587 |
+
"loss": 0.0429,
|
| 588 |
+
"step": 560
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 28.0,
|
| 592 |
+
"eval_accuracy": 0.7843137254901961,
|
| 593 |
+
"eval_loss": 0.9203042984008789,
|
| 594 |
+
"eval_runtime": 0.8373,
|
| 595 |
+
"eval_samples_per_second": 60.907,
|
| 596 |
+
"eval_steps_per_second": 2.389,
|
| 597 |
+
"step": 560
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 28.5,
|
| 601 |
+
"learning_rate": 1.5131578947368422e-05,
|
| 602 |
+
"loss": 0.0323,
|
| 603 |
+
"step": 570
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 29.0,
|
| 607 |
+
"learning_rate": 1.4473684210526317e-05,
|
| 608 |
+
"loss": 0.0386,
|
| 609 |
+
"step": 580
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 29.0,
|
| 613 |
+
"eval_accuracy": 0.803921568627451,
|
| 614 |
+
"eval_loss": 0.8732457160949707,
|
| 615 |
+
"eval_runtime": 0.8127,
|
| 616 |
+
"eval_samples_per_second": 62.756,
|
| 617 |
+
"eval_steps_per_second": 2.461,
|
| 618 |
+
"step": 580
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 29.5,
|
| 622 |
+
"learning_rate": 1.3815789473684213e-05,
|
| 623 |
+
"loss": 0.0475,
|
| 624 |
+
"step": 590
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 30.0,
|
| 628 |
+
"learning_rate": 1.3157894736842106e-05,
|
| 629 |
+
"loss": 0.0433,
|
| 630 |
+
"step": 600
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 30.0,
|
| 634 |
+
"eval_accuracy": 0.7647058823529411,
|
| 635 |
+
"eval_loss": 0.9375914931297302,
|
| 636 |
+
"eval_runtime": 0.7855,
|
| 637 |
+
"eval_samples_per_second": 64.925,
|
| 638 |
+
"eval_steps_per_second": 2.546,
|
| 639 |
+
"step": 600
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 30.5,
|
| 643 |
+
"learning_rate": 1.25e-05,
|
| 644 |
+
"loss": 0.054,
|
| 645 |
+
"step": 610
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 31.0,
|
| 649 |
+
"learning_rate": 1.1842105263157895e-05,
|
| 650 |
+
"loss": 0.0353,
|
| 651 |
+
"step": 620
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 31.0,
|
| 655 |
+
"eval_accuracy": 0.7843137254901961,
|
| 656 |
+
"eval_loss": 0.8531509041786194,
|
| 657 |
+
"eval_runtime": 0.7884,
|
| 658 |
+
"eval_samples_per_second": 64.687,
|
| 659 |
+
"eval_steps_per_second": 2.537,
|
| 660 |
+
"step": 620
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 31.5,
|
| 664 |
+
"learning_rate": 1.118421052631579e-05,
|
| 665 |
+
"loss": 0.0412,
|
| 666 |
+
"step": 630
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 32.0,
|
| 670 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 671 |
+
"loss": 0.0332,
|
| 672 |
+
"step": 640
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 32.0,
|
| 676 |
+
"eval_accuracy": 0.803921568627451,
|
| 677 |
+
"eval_loss": 0.9123408794403076,
|
| 678 |
+
"eval_runtime": 0.8508,
|
| 679 |
+
"eval_samples_per_second": 59.946,
|
| 680 |
+
"eval_steps_per_second": 2.351,
|
| 681 |
+
"step": 640
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 32.5,
|
| 685 |
+
"learning_rate": 9.868421052631579e-06,
|
| 686 |
+
"loss": 0.0255,
|
| 687 |
+
"step": 650
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 33.0,
|
| 691 |
+
"learning_rate": 9.210526315789474e-06,
|
| 692 |
+
"loss": 0.0405,
|
| 693 |
+
"step": 660
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 33.0,
|
| 697 |
+
"eval_accuracy": 0.803921568627451,
|
| 698 |
+
"eval_loss": 0.9603415727615356,
|
| 699 |
+
"eval_runtime": 0.772,
|
| 700 |
+
"eval_samples_per_second": 66.066,
|
| 701 |
+
"eval_steps_per_second": 2.591,
|
| 702 |
+
"step": 660
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 33.5,
|
| 706 |
+
"learning_rate": 8.552631578947368e-06,
|
| 707 |
+
"loss": 0.0445,
|
| 708 |
+
"step": 670
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 34.0,
|
| 712 |
+
"learning_rate": 7.894736842105263e-06,
|
| 713 |
+
"loss": 0.0423,
|
| 714 |
+
"step": 680
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 34.0,
|
| 718 |
+
"eval_accuracy": 0.803921568627451,
|
| 719 |
+
"eval_loss": 0.9424175024032593,
|
| 720 |
+
"eval_runtime": 0.8978,
|
| 721 |
+
"eval_samples_per_second": 56.806,
|
| 722 |
+
"eval_steps_per_second": 2.228,
|
| 723 |
+
"step": 680
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 34.5,
|
| 727 |
+
"learning_rate": 7.236842105263158e-06,
|
| 728 |
+
"loss": 0.0317,
|
| 729 |
+
"step": 690
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 35.0,
|
| 733 |
+
"learning_rate": 6.578947368421053e-06,
|
| 734 |
+
"loss": 0.0383,
|
| 735 |
+
"step": 700
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 35.0,
|
| 739 |
+
"eval_accuracy": 0.8235294117647058,
|
| 740 |
+
"eval_loss": 0.9687171578407288,
|
| 741 |
+
"eval_runtime": 0.7905,
|
| 742 |
+
"eval_samples_per_second": 64.514,
|
| 743 |
+
"eval_steps_per_second": 2.53,
|
| 744 |
+
"step": 700
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 35.5,
|
| 748 |
+
"learning_rate": 5.921052631578948e-06,
|
| 749 |
+
"loss": 0.0435,
|
| 750 |
+
"step": 710
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 36.0,
|
| 754 |
+
"learning_rate": 5.263157894736842e-06,
|
| 755 |
+
"loss": 0.0245,
|
| 756 |
+
"step": 720
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 36.0,
|
| 760 |
+
"eval_accuracy": 0.8235294117647058,
|
| 761 |
+
"eval_loss": 0.9509010314941406,
|
| 762 |
+
"eval_runtime": 0.8121,
|
| 763 |
+
"eval_samples_per_second": 62.798,
|
| 764 |
+
"eval_steps_per_second": 2.463,
|
| 765 |
+
"step": 720
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 36.5,
|
| 769 |
+
"learning_rate": 4.605263157894737e-06,
|
| 770 |
+
"loss": 0.0236,
|
| 771 |
+
"step": 730
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 37.0,
|
| 775 |
+
"learning_rate": 3.9473684210526315e-06,
|
| 776 |
+
"loss": 0.0309,
|
| 777 |
+
"step": 740
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 37.0,
|
| 781 |
+
"eval_accuracy": 0.8235294117647058,
|
| 782 |
+
"eval_loss": 0.8950434923171997,
|
| 783 |
+
"eval_runtime": 0.8148,
|
| 784 |
+
"eval_samples_per_second": 62.589,
|
| 785 |
+
"eval_steps_per_second": 2.454,
|
| 786 |
+
"step": 740
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 37.5,
|
| 790 |
+
"learning_rate": 3.2894736842105265e-06,
|
| 791 |
+
"loss": 0.0306,
|
| 792 |
+
"step": 750
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 38.0,
|
| 796 |
+
"learning_rate": 2.631578947368421e-06,
|
| 797 |
+
"loss": 0.026,
|
| 798 |
+
"step": 760
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 38.0,
|
| 802 |
+
"eval_accuracy": 0.803921568627451,
|
| 803 |
+
"eval_loss": 0.9081793427467346,
|
| 804 |
+
"eval_runtime": 0.8266,
|
| 805 |
+
"eval_samples_per_second": 61.7,
|
| 806 |
+
"eval_steps_per_second": 2.42,
|
| 807 |
+
"step": 760
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 38.5,
|
| 811 |
+
"learning_rate": 1.9736842105263157e-06,
|
| 812 |
+
"loss": 0.0335,
|
| 813 |
+
"step": 770
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 39.0,
|
| 817 |
+
"learning_rate": 1.3157894736842106e-06,
|
| 818 |
+
"loss": 0.0192,
|
| 819 |
+
"step": 780
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 39.0,
|
| 823 |
+
"eval_accuracy": 0.8235294117647058,
|
| 824 |
+
"eval_loss": 0.8859331607818604,
|
| 825 |
+
"eval_runtime": 0.8155,
|
| 826 |
+
"eval_samples_per_second": 62.535,
|
| 827 |
+
"eval_steps_per_second": 2.452,
|
| 828 |
+
"step": 780
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 39.5,
|
| 832 |
+
"learning_rate": 6.578947368421053e-07,
|
| 833 |
+
"loss": 0.0235,
|
| 834 |
+
"step": 790
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 40.0,
|
| 838 |
+
"learning_rate": 0.0,
|
| 839 |
+
"loss": 0.0322,
|
| 840 |
+
"step": 800
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 40.0,
|
| 844 |
+
"eval_accuracy": 0.8235294117647058,
|
| 845 |
+
"eval_loss": 0.8968344926834106,
|
| 846 |
+
"eval_runtime": 0.8319,
|
| 847 |
+
"eval_samples_per_second": 61.307,
|
| 848 |
+
"eval_steps_per_second": 2.404,
|
| 849 |
+
"step": 800
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 40.0,
|
| 853 |
+
"step": 800,
|
| 854 |
+
"total_flos": 7.873327274596762e+18,
|
| 855 |
+
"train_loss": 0.1865533402003348,
|
| 856 |
+
"train_runtime": 1567.0918,
|
| 857 |
+
"train_samples_per_second": 64.833,
|
| 858 |
+
"train_steps_per_second": 0.51
|
| 859 |
+
}
|
| 860 |
+
],
|
| 861 |
+
"logging_steps": 10,
|
| 862 |
+
"max_steps": 800,
|
| 863 |
+
"num_input_tokens_seen": 0,
|
| 864 |
+
"num_train_epochs": 40,
|
| 865 |
+
"save_steps": 500,
|
| 866 |
+
"total_flos": 7.873327274596762e+18,
|
| 867 |
+
"train_batch_size": 32,
|
| 868 |
+
"trial_name": null,
|
| 869 |
+
"trial_params": null
|
| 870 |
+
}
|