Commit
•
62f11bd
1
Parent(s):
667b8cf
End of training
Browse files- README.md +3 -3
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/May23_18-36-22_DESKTOP-SKBE9FB/events.out.tfevents.1716512679.DESKTOP-SKBE9FB.14084.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +846 -0
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ 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:
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.8333333333333334
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 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: 0.5667
|
| 36 |
+
- Accuracy: 0.8333
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 39.48,
|
| 3 |
+
"eval_accuracy": 0.8333333333333334,
|
| 4 |
+
"eval_loss": 0.5666719079017639,
|
| 5 |
+
"eval_runtime": 1.1751,
|
| 6 |
+
"eval_samples_per_second": 51.059,
|
| 7 |
+
"eval_steps_per_second": 1.702,
|
| 8 |
+
"train_loss": 0.18620210740911333,
|
| 9 |
+
"train_runtime": 1570.0451,
|
| 10 |
+
"train_samples_per_second": 62.419,
|
| 11 |
+
"train_steps_per_second": 0.484
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 39.48,
|
| 3 |
+
"eval_accuracy": 0.8333333333333334,
|
| 4 |
+
"eval_loss": 0.5666719079017639,
|
| 5 |
+
"eval_runtime": 1.1751,
|
| 6 |
+
"eval_samples_per_second": 51.059,
|
| 7 |
+
"eval_steps_per_second": 1.702
|
| 8 |
+
}
|
runs/May23_18-36-22_DESKTOP-SKBE9FB/events.out.tfevents.1716512679.DESKTOP-SKBE9FB.14084.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6482edb473fd87748c86e614a871d662c755f8bbc6ba3c74da2d21c392784e5f
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 39.48,
|
| 3 |
+
"train_loss": 0.18620210740911333,
|
| 4 |
+
"train_runtime": 1570.0451,
|
| 5 |
+
"train_samples_per_second": 62.419,
|
| 6 |
+
"train_steps_per_second": 0.484
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,846 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8333333333333334,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-RU3-40\\checkpoint-134",
|
| 4 |
+
"epoch": 39.48051948051948,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 760,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.52,
|
| 13 |
+
"learning_rate": 1.4473684210526315e-05,
|
| 14 |
+
"loss": 1.3821,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.99,
|
| 19 |
+
"eval_accuracy": 0.48333333333333334,
|
| 20 |
+
"eval_loss": 1.3118996620178223,
|
| 21 |
+
"eval_runtime": 1.0501,
|
| 22 |
+
"eval_samples_per_second": 57.139,
|
| 23 |
+
"eval_steps_per_second": 1.905,
|
| 24 |
+
"step": 19
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.04,
|
| 28 |
+
"learning_rate": 2.894736842105263e-05,
|
| 29 |
+
"loss": 1.34,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.56,
|
| 34 |
+
"learning_rate": 4.342105263157895e-05,
|
| 35 |
+
"loss": 1.2698,
|
| 36 |
+
"step": 30
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 1.97,
|
| 40 |
+
"eval_accuracy": 0.6166666666666667,
|
| 41 |
+
"eval_loss": 1.085192084312439,
|
| 42 |
+
"eval_runtime": 1.0848,
|
| 43 |
+
"eval_samples_per_second": 55.308,
|
| 44 |
+
"eval_steps_per_second": 1.844,
|
| 45 |
+
"step": 38
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.08,
|
| 49 |
+
"learning_rate": 5.484764542936288e-05,
|
| 50 |
+
"loss": 1.1206,
|
| 51 |
+
"step": 40
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 2.6,
|
| 55 |
+
"learning_rate": 5.408587257617729e-05,
|
| 56 |
+
"loss": 0.9819,
|
| 57 |
+
"step": 50
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 2.96,
|
| 61 |
+
"eval_accuracy": 0.7,
|
| 62 |
+
"eval_loss": 0.875718355178833,
|
| 63 |
+
"eval_runtime": 1.0604,
|
| 64 |
+
"eval_samples_per_second": 56.582,
|
| 65 |
+
"eval_steps_per_second": 1.886,
|
| 66 |
+
"step": 57
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 3.12,
|
| 70 |
+
"learning_rate": 5.332409972299169e-05,
|
| 71 |
+
"loss": 0.8387,
|
| 72 |
+
"step": 60
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 3.64,
|
| 76 |
+
"learning_rate": 5.256232686980609e-05,
|
| 77 |
+
"loss": 0.6671,
|
| 78 |
+
"step": 70
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.0,
|
| 82 |
+
"eval_accuracy": 0.7333333333333333,
|
| 83 |
+
"eval_loss": 0.768868088722229,
|
| 84 |
+
"eval_runtime": 1.1869,
|
| 85 |
+
"eval_samples_per_second": 50.551,
|
| 86 |
+
"eval_steps_per_second": 1.685,
|
| 87 |
+
"step": 77
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 4.16,
|
| 91 |
+
"learning_rate": 5.18005540166205e-05,
|
| 92 |
+
"loss": 0.5484,
|
| 93 |
+
"step": 80
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.68,
|
| 97 |
+
"learning_rate": 5.1038781163434903e-05,
|
| 98 |
+
"loss": 0.4248,
|
| 99 |
+
"step": 90
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.99,
|
| 103 |
+
"eval_accuracy": 0.7166666666666667,
|
| 104 |
+
"eval_loss": 0.7293522357940674,
|
| 105 |
+
"eval_runtime": 1.0714,
|
| 106 |
+
"eval_samples_per_second": 56.0,
|
| 107 |
+
"eval_steps_per_second": 1.867,
|
| 108 |
+
"step": 96
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 5.19,
|
| 112 |
+
"learning_rate": 5.027700831024931e-05,
|
| 113 |
+
"loss": 0.3571,
|
| 114 |
+
"step": 100
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.71,
|
| 118 |
+
"learning_rate": 4.9515235457063714e-05,
|
| 119 |
+
"loss": 0.3005,
|
| 120 |
+
"step": 110
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.97,
|
| 124 |
+
"eval_accuracy": 0.7833333333333333,
|
| 125 |
+
"eval_loss": 0.6517642140388489,
|
| 126 |
+
"eval_runtime": 1.0607,
|
| 127 |
+
"eval_samples_per_second": 56.569,
|
| 128 |
+
"eval_steps_per_second": 1.886,
|
| 129 |
+
"step": 115
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.23,
|
| 133 |
+
"learning_rate": 4.8753462603878116e-05,
|
| 134 |
+
"loss": 0.2452,
|
| 135 |
+
"step": 120
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 6.75,
|
| 139 |
+
"learning_rate": 4.7991689750692524e-05,
|
| 140 |
+
"loss": 0.2035,
|
| 141 |
+
"step": 130
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 6.96,
|
| 145 |
+
"eval_accuracy": 0.8333333333333334,
|
| 146 |
+
"eval_loss": 0.5666719079017639,
|
| 147 |
+
"eval_runtime": 1.0912,
|
| 148 |
+
"eval_samples_per_second": 54.986,
|
| 149 |
+
"eval_steps_per_second": 1.833,
|
| 150 |
+
"step": 134
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 7.27,
|
| 154 |
+
"learning_rate": 4.7229916897506926e-05,
|
| 155 |
+
"loss": 0.2138,
|
| 156 |
+
"step": 140
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 7.79,
|
| 160 |
+
"learning_rate": 4.6468144044321335e-05,
|
| 161 |
+
"loss": 0.2195,
|
| 162 |
+
"step": 150
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 8.0,
|
| 166 |
+
"eval_accuracy": 0.8333333333333334,
|
| 167 |
+
"eval_loss": 0.6646459102630615,
|
| 168 |
+
"eval_runtime": 1.0657,
|
| 169 |
+
"eval_samples_per_second": 56.3,
|
| 170 |
+
"eval_steps_per_second": 1.877,
|
| 171 |
+
"step": 154
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 8.31,
|
| 175 |
+
"learning_rate": 4.570637119113573e-05,
|
| 176 |
+
"loss": 0.1806,
|
| 177 |
+
"step": 160
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 8.83,
|
| 181 |
+
"learning_rate": 4.494459833795014e-05,
|
| 182 |
+
"loss": 0.1654,
|
| 183 |
+
"step": 170
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 8.99,
|
| 187 |
+
"eval_accuracy": 0.8166666666666667,
|
| 188 |
+
"eval_loss": 0.6294049024581909,
|
| 189 |
+
"eval_runtime": 1.1008,
|
| 190 |
+
"eval_samples_per_second": 54.503,
|
| 191 |
+
"eval_steps_per_second": 1.817,
|
| 192 |
+
"step": 173
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 9.35,
|
| 196 |
+
"learning_rate": 4.418282548476455e-05,
|
| 197 |
+
"loss": 0.1641,
|
| 198 |
+
"step": 180
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 9.87,
|
| 202 |
+
"learning_rate": 4.342105263157895e-05,
|
| 203 |
+
"loss": 0.1581,
|
| 204 |
+
"step": 190
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.97,
|
| 208 |
+
"eval_accuracy": 0.7833333333333333,
|
| 209 |
+
"eval_loss": 0.7211242318153381,
|
| 210 |
+
"eval_runtime": 1.0634,
|
| 211 |
+
"eval_samples_per_second": 56.425,
|
| 212 |
+
"eval_steps_per_second": 1.881,
|
| 213 |
+
"step": 192
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 10.39,
|
| 217 |
+
"learning_rate": 4.265927977839336e-05,
|
| 218 |
+
"loss": 0.1147,
|
| 219 |
+
"step": 200
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 10.91,
|
| 223 |
+
"learning_rate": 4.189750692520776e-05,
|
| 224 |
+
"loss": 0.1338,
|
| 225 |
+
"step": 210
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 10.96,
|
| 229 |
+
"eval_accuracy": 0.7833333333333333,
|
| 230 |
+
"eval_loss": 0.8129342198371887,
|
| 231 |
+
"eval_runtime": 1.1857,
|
| 232 |
+
"eval_samples_per_second": 50.603,
|
| 233 |
+
"eval_steps_per_second": 1.687,
|
| 234 |
+
"step": 211
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 11.43,
|
| 238 |
+
"learning_rate": 4.113573407202216e-05,
|
| 239 |
+
"loss": 0.1313,
|
| 240 |
+
"step": 220
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 11.95,
|
| 244 |
+
"learning_rate": 4.037396121883656e-05,
|
| 245 |
+
"loss": 0.1188,
|
| 246 |
+
"step": 230
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 12.0,
|
| 250 |
+
"eval_accuracy": 0.8166666666666667,
|
| 251 |
+
"eval_loss": 0.7924687266349792,
|
| 252 |
+
"eval_runtime": 1.0887,
|
| 253 |
+
"eval_samples_per_second": 55.111,
|
| 254 |
+
"eval_steps_per_second": 1.837,
|
| 255 |
+
"step": 231
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 12.47,
|
| 259 |
+
"learning_rate": 3.961218836565097e-05,
|
| 260 |
+
"loss": 0.0969,
|
| 261 |
+
"step": 240
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 12.99,
|
| 265 |
+
"learning_rate": 3.885041551246538e-05,
|
| 266 |
+
"loss": 0.1179,
|
| 267 |
+
"step": 250
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 12.99,
|
| 271 |
+
"eval_accuracy": 0.7666666666666667,
|
| 272 |
+
"eval_loss": 0.958832323551178,
|
| 273 |
+
"eval_runtime": 1.165,
|
| 274 |
+
"eval_samples_per_second": 51.501,
|
| 275 |
+
"eval_steps_per_second": 1.717,
|
| 276 |
+
"step": 250
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 13.51,
|
| 280 |
+
"learning_rate": 3.808864265927978e-05,
|
| 281 |
+
"loss": 0.1017,
|
| 282 |
+
"step": 260
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 13.97,
|
| 286 |
+
"eval_accuracy": 0.7166666666666667,
|
| 287 |
+
"eval_loss": 1.0874536037445068,
|
| 288 |
+
"eval_runtime": 1.0498,
|
| 289 |
+
"eval_samples_per_second": 57.151,
|
| 290 |
+
"eval_steps_per_second": 1.905,
|
| 291 |
+
"step": 269
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 14.03,
|
| 295 |
+
"learning_rate": 3.732686980609418e-05,
|
| 296 |
+
"loss": 0.1004,
|
| 297 |
+
"step": 270
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 14.55,
|
| 301 |
+
"learning_rate": 3.6565096952908585e-05,
|
| 302 |
+
"loss": 0.0845,
|
| 303 |
+
"step": 280
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 14.96,
|
| 307 |
+
"eval_accuracy": 0.7,
|
| 308 |
+
"eval_loss": 0.935455322265625,
|
| 309 |
+
"eval_runtime": 1.1044,
|
| 310 |
+
"eval_samples_per_second": 54.328,
|
| 311 |
+
"eval_steps_per_second": 1.811,
|
| 312 |
+
"step": 288
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 15.06,
|
| 316 |
+
"learning_rate": 3.5803324099722994e-05,
|
| 317 |
+
"loss": 0.0898,
|
| 318 |
+
"step": 290
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 15.58,
|
| 322 |
+
"learning_rate": 3.5041551246537395e-05,
|
| 323 |
+
"loss": 0.1109,
|
| 324 |
+
"step": 300
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 16.0,
|
| 328 |
+
"eval_accuracy": 0.8166666666666667,
|
| 329 |
+
"eval_loss": 0.9386860728263855,
|
| 330 |
+
"eval_runtime": 1.0479,
|
| 331 |
+
"eval_samples_per_second": 57.256,
|
| 332 |
+
"eval_steps_per_second": 1.909,
|
| 333 |
+
"step": 308
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 16.1,
|
| 337 |
+
"learning_rate": 3.4279778393351804e-05,
|
| 338 |
+
"loss": 0.0868,
|
| 339 |
+
"step": 310
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 16.62,
|
| 343 |
+
"learning_rate": 3.3518005540166206e-05,
|
| 344 |
+
"loss": 0.0711,
|
| 345 |
+
"step": 320
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 16.99,
|
| 349 |
+
"eval_accuracy": 0.7333333333333333,
|
| 350 |
+
"eval_loss": 1.121420979499817,
|
| 351 |
+
"eval_runtime": 1.0767,
|
| 352 |
+
"eval_samples_per_second": 55.726,
|
| 353 |
+
"eval_steps_per_second": 1.858,
|
| 354 |
+
"step": 327
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 17.14,
|
| 358 |
+
"learning_rate": 3.275623268698061e-05,
|
| 359 |
+
"loss": 0.079,
|
| 360 |
+
"step": 330
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 17.66,
|
| 364 |
+
"learning_rate": 3.1994459833795016e-05,
|
| 365 |
+
"loss": 0.0884,
|
| 366 |
+
"step": 340
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 17.97,
|
| 370 |
+
"eval_accuracy": 0.7666666666666667,
|
| 371 |
+
"eval_loss": 0.9687939286231995,
|
| 372 |
+
"eval_runtime": 1.0985,
|
| 373 |
+
"eval_samples_per_second": 54.619,
|
| 374 |
+
"eval_steps_per_second": 1.821,
|
| 375 |
+
"step": 346
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 18.18,
|
| 379 |
+
"learning_rate": 3.123268698060942e-05,
|
| 380 |
+
"loss": 0.0661,
|
| 381 |
+
"step": 350
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 18.7,
|
| 385 |
+
"learning_rate": 3.0470914127423827e-05,
|
| 386 |
+
"loss": 0.0668,
|
| 387 |
+
"step": 360
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 18.96,
|
| 391 |
+
"eval_accuracy": 0.8,
|
| 392 |
+
"eval_loss": 1.0306044816970825,
|
| 393 |
+
"eval_runtime": 1.1148,
|
| 394 |
+
"eval_samples_per_second": 53.823,
|
| 395 |
+
"eval_steps_per_second": 1.794,
|
| 396 |
+
"step": 365
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 19.22,
|
| 400 |
+
"learning_rate": 2.9709141274238225e-05,
|
| 401 |
+
"loss": 0.0655,
|
| 402 |
+
"step": 370
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 19.74,
|
| 406 |
+
"learning_rate": 2.894736842105263e-05,
|
| 407 |
+
"loss": 0.0716,
|
| 408 |
+
"step": 380
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 20.0,
|
| 412 |
+
"eval_accuracy": 0.7166666666666667,
|
| 413 |
+
"eval_loss": 1.2652896642684937,
|
| 414 |
+
"eval_runtime": 1.097,
|
| 415 |
+
"eval_samples_per_second": 54.696,
|
| 416 |
+
"eval_steps_per_second": 1.823,
|
| 417 |
+
"step": 385
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 20.26,
|
| 421 |
+
"learning_rate": 2.8185595567867035e-05,
|
| 422 |
+
"loss": 0.0725,
|
| 423 |
+
"step": 390
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 20.78,
|
| 427 |
+
"learning_rate": 2.742382271468144e-05,
|
| 428 |
+
"loss": 0.0643,
|
| 429 |
+
"step": 400
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 20.99,
|
| 433 |
+
"eval_accuracy": 0.7833333333333333,
|
| 434 |
+
"eval_loss": 0.9893661737442017,
|
| 435 |
+
"eval_runtime": 1.1231,
|
| 436 |
+
"eval_samples_per_second": 53.424,
|
| 437 |
+
"eval_steps_per_second": 1.781,
|
| 438 |
+
"step": 404
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 21.3,
|
| 442 |
+
"learning_rate": 2.6662049861495846e-05,
|
| 443 |
+
"loss": 0.0509,
|
| 444 |
+
"step": 410
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 21.82,
|
| 448 |
+
"learning_rate": 2.590027700831025e-05,
|
| 449 |
+
"loss": 0.0517,
|
| 450 |
+
"step": 420
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 21.97,
|
| 454 |
+
"eval_accuracy": 0.7666666666666667,
|
| 455 |
+
"eval_loss": 1.0438777208328247,
|
| 456 |
+
"eval_runtime": 1.1007,
|
| 457 |
+
"eval_samples_per_second": 54.511,
|
| 458 |
+
"eval_steps_per_second": 1.817,
|
| 459 |
+
"step": 423
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 22.34,
|
| 463 |
+
"learning_rate": 2.5138504155124656e-05,
|
| 464 |
+
"loss": 0.0637,
|
| 465 |
+
"step": 430
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 22.86,
|
| 469 |
+
"learning_rate": 2.4376731301939058e-05,
|
| 470 |
+
"loss": 0.0597,
|
| 471 |
+
"step": 440
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 22.96,
|
| 475 |
+
"eval_accuracy": 0.7666666666666667,
|
| 476 |
+
"eval_loss": 1.1469690799713135,
|
| 477 |
+
"eval_runtime": 1.0658,
|
| 478 |
+
"eval_samples_per_second": 56.294,
|
| 479 |
+
"eval_steps_per_second": 1.876,
|
| 480 |
+
"step": 442
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 23.38,
|
| 484 |
+
"learning_rate": 2.3614958448753463e-05,
|
| 485 |
+
"loss": 0.0377,
|
| 486 |
+
"step": 450
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 23.9,
|
| 490 |
+
"learning_rate": 2.2853185595567865e-05,
|
| 491 |
+
"loss": 0.0533,
|
| 492 |
+
"step": 460
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 24.0,
|
| 496 |
+
"eval_accuracy": 0.7833333333333333,
|
| 497 |
+
"eval_loss": 1.0848194360733032,
|
| 498 |
+
"eval_runtime": 1.148,
|
| 499 |
+
"eval_samples_per_second": 52.264,
|
| 500 |
+
"eval_steps_per_second": 1.742,
|
| 501 |
+
"step": 462
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 24.42,
|
| 505 |
+
"learning_rate": 2.2091412742382273e-05,
|
| 506 |
+
"loss": 0.0599,
|
| 507 |
+
"step": 470
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 24.94,
|
| 511 |
+
"learning_rate": 2.132963988919668e-05,
|
| 512 |
+
"loss": 0.0529,
|
| 513 |
+
"step": 480
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 24.99,
|
| 517 |
+
"eval_accuracy": 0.75,
|
| 518 |
+
"eval_loss": 1.1481404304504395,
|
| 519 |
+
"eval_runtime": 1.0909,
|
| 520 |
+
"eval_samples_per_second": 55.002,
|
| 521 |
+
"eval_steps_per_second": 1.833,
|
| 522 |
+
"step": 481
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 25.45,
|
| 526 |
+
"learning_rate": 2.056786703601108e-05,
|
| 527 |
+
"loss": 0.0373,
|
| 528 |
+
"step": 490
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 25.97,
|
| 532 |
+
"learning_rate": 1.9806094182825486e-05,
|
| 533 |
+
"loss": 0.0524,
|
| 534 |
+
"step": 500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 25.97,
|
| 538 |
+
"eval_accuracy": 0.7333333333333333,
|
| 539 |
+
"eval_loss": 1.1321938037872314,
|
| 540 |
+
"eval_runtime": 1.1085,
|
| 541 |
+
"eval_samples_per_second": 54.127,
|
| 542 |
+
"eval_steps_per_second": 1.804,
|
| 543 |
+
"step": 500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 26.49,
|
| 547 |
+
"learning_rate": 1.904432132963989e-05,
|
| 548 |
+
"loss": 0.0525,
|
| 549 |
+
"step": 510
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 26.96,
|
| 553 |
+
"eval_accuracy": 0.7333333333333333,
|
| 554 |
+
"eval_loss": 1.1867502927780151,
|
| 555 |
+
"eval_runtime": 1.0522,
|
| 556 |
+
"eval_samples_per_second": 57.023,
|
| 557 |
+
"eval_steps_per_second": 1.901,
|
| 558 |
+
"step": 519
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 27.01,
|
| 562 |
+
"learning_rate": 1.8282548476454293e-05,
|
| 563 |
+
"loss": 0.0574,
|
| 564 |
+
"step": 520
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 27.53,
|
| 568 |
+
"learning_rate": 1.7520775623268698e-05,
|
| 569 |
+
"loss": 0.0517,
|
| 570 |
+
"step": 530
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 28.0,
|
| 574 |
+
"eval_accuracy": 0.7166666666666667,
|
| 575 |
+
"eval_loss": 1.1561001539230347,
|
| 576 |
+
"eval_runtime": 1.1047,
|
| 577 |
+
"eval_samples_per_second": 54.311,
|
| 578 |
+
"eval_steps_per_second": 1.81,
|
| 579 |
+
"step": 539
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 28.05,
|
| 583 |
+
"learning_rate": 1.6759002770083103e-05,
|
| 584 |
+
"loss": 0.0505,
|
| 585 |
+
"step": 540
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 28.57,
|
| 589 |
+
"learning_rate": 1.5997229916897508e-05,
|
| 590 |
+
"loss": 0.0309,
|
| 591 |
+
"step": 550
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 28.99,
|
| 595 |
+
"eval_accuracy": 0.7833333333333333,
|
| 596 |
+
"eval_loss": 1.0561987161636353,
|
| 597 |
+
"eval_runtime": 1.0565,
|
| 598 |
+
"eval_samples_per_second": 56.79,
|
| 599 |
+
"eval_steps_per_second": 1.893,
|
| 600 |
+
"step": 558
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 29.09,
|
| 604 |
+
"learning_rate": 1.5235457063711913e-05,
|
| 605 |
+
"loss": 0.0429,
|
| 606 |
+
"step": 560
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 29.61,
|
| 610 |
+
"learning_rate": 1.4473684210526315e-05,
|
| 611 |
+
"loss": 0.0403,
|
| 612 |
+
"step": 570
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 29.97,
|
| 616 |
+
"eval_accuracy": 0.7333333333333333,
|
| 617 |
+
"eval_loss": 1.2900880575180054,
|
| 618 |
+
"eval_runtime": 1.099,
|
| 619 |
+
"eval_samples_per_second": 54.594,
|
| 620 |
+
"eval_steps_per_second": 1.82,
|
| 621 |
+
"step": 577
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 30.13,
|
| 625 |
+
"learning_rate": 1.371191135734072e-05,
|
| 626 |
+
"loss": 0.0442,
|
| 627 |
+
"step": 580
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 30.65,
|
| 631 |
+
"learning_rate": 1.2950138504155125e-05,
|
| 632 |
+
"loss": 0.0392,
|
| 633 |
+
"step": 590
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 30.96,
|
| 637 |
+
"eval_accuracy": 0.7666666666666667,
|
| 638 |
+
"eval_loss": 1.1295415163040161,
|
| 639 |
+
"eval_runtime": 1.0734,
|
| 640 |
+
"eval_samples_per_second": 55.898,
|
| 641 |
+
"eval_steps_per_second": 1.863,
|
| 642 |
+
"step": 596
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 31.17,
|
| 646 |
+
"learning_rate": 1.2188365650969529e-05,
|
| 647 |
+
"loss": 0.0274,
|
| 648 |
+
"step": 600
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 31.69,
|
| 652 |
+
"learning_rate": 1.1426592797783932e-05,
|
| 653 |
+
"loss": 0.0404,
|
| 654 |
+
"step": 610
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 32.0,
|
| 658 |
+
"eval_accuracy": 0.7666666666666667,
|
| 659 |
+
"eval_loss": 1.119759440422058,
|
| 660 |
+
"eval_runtime": 1.2454,
|
| 661 |
+
"eval_samples_per_second": 48.178,
|
| 662 |
+
"eval_steps_per_second": 1.606,
|
| 663 |
+
"step": 616
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 32.21,
|
| 667 |
+
"learning_rate": 1.066481994459834e-05,
|
| 668 |
+
"loss": 0.0212,
|
| 669 |
+
"step": 620
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 32.73,
|
| 673 |
+
"learning_rate": 9.903047091412743e-06,
|
| 674 |
+
"loss": 0.0381,
|
| 675 |
+
"step": 630
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 32.99,
|
| 679 |
+
"eval_accuracy": 0.7166666666666667,
|
| 680 |
+
"eval_loss": 1.298552393913269,
|
| 681 |
+
"eval_runtime": 1.1082,
|
| 682 |
+
"eval_samples_per_second": 54.143,
|
| 683 |
+
"eval_steps_per_second": 1.805,
|
| 684 |
+
"step": 635
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 33.25,
|
| 688 |
+
"learning_rate": 9.141274238227146e-06,
|
| 689 |
+
"loss": 0.0259,
|
| 690 |
+
"step": 640
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 33.77,
|
| 694 |
+
"learning_rate": 8.379501385041551e-06,
|
| 695 |
+
"loss": 0.0262,
|
| 696 |
+
"step": 650
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 33.97,
|
| 700 |
+
"eval_accuracy": 0.75,
|
| 701 |
+
"eval_loss": 1.1655079126358032,
|
| 702 |
+
"eval_runtime": 1.1294,
|
| 703 |
+
"eval_samples_per_second": 53.126,
|
| 704 |
+
"eval_steps_per_second": 1.771,
|
| 705 |
+
"step": 654
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 34.29,
|
| 709 |
+
"learning_rate": 7.617728531855957e-06,
|
| 710 |
+
"loss": 0.0301,
|
| 711 |
+
"step": 660
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 34.81,
|
| 715 |
+
"learning_rate": 6.85595567867036e-06,
|
| 716 |
+
"loss": 0.0354,
|
| 717 |
+
"step": 670
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 34.96,
|
| 721 |
+
"eval_accuracy": 0.7833333333333333,
|
| 722 |
+
"eval_loss": 1.1222782135009766,
|
| 723 |
+
"eval_runtime": 1.0831,
|
| 724 |
+
"eval_samples_per_second": 55.396,
|
| 725 |
+
"eval_steps_per_second": 1.847,
|
| 726 |
+
"step": 673
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 35.32,
|
| 730 |
+
"learning_rate": 6.0941828254847645e-06,
|
| 731 |
+
"loss": 0.0291,
|
| 732 |
+
"step": 680
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 35.84,
|
| 736 |
+
"learning_rate": 5.33240997229917e-06,
|
| 737 |
+
"loss": 0.0224,
|
| 738 |
+
"step": 690
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 36.0,
|
| 742 |
+
"eval_accuracy": 0.7833333333333333,
|
| 743 |
+
"eval_loss": 1.1678553819656372,
|
| 744 |
+
"eval_runtime": 1.0793,
|
| 745 |
+
"eval_samples_per_second": 55.591,
|
| 746 |
+
"eval_steps_per_second": 1.853,
|
| 747 |
+
"step": 693
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 36.36,
|
| 751 |
+
"learning_rate": 4.570637119113573e-06,
|
| 752 |
+
"loss": 0.0349,
|
| 753 |
+
"step": 700
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 36.88,
|
| 757 |
+
"learning_rate": 3.8088642659279783e-06,
|
| 758 |
+
"loss": 0.0244,
|
| 759 |
+
"step": 710
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 36.99,
|
| 763 |
+
"eval_accuracy": 0.8166666666666667,
|
| 764 |
+
"eval_loss": 1.0998696088790894,
|
| 765 |
+
"eval_runtime": 1.077,
|
| 766 |
+
"eval_samples_per_second": 55.712,
|
| 767 |
+
"eval_steps_per_second": 1.857,
|
| 768 |
+
"step": 712
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 37.4,
|
| 772 |
+
"learning_rate": 3.0470914127423822e-06,
|
| 773 |
+
"loss": 0.0412,
|
| 774 |
+
"step": 720
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 37.92,
|
| 778 |
+
"learning_rate": 2.2853185595567866e-06,
|
| 779 |
+
"loss": 0.0368,
|
| 780 |
+
"step": 730
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 37.97,
|
| 784 |
+
"eval_accuracy": 0.7833333333333333,
|
| 785 |
+
"eval_loss": 1.121293306350708,
|
| 786 |
+
"eval_runtime": 1.0596,
|
| 787 |
+
"eval_samples_per_second": 56.628,
|
| 788 |
+
"eval_steps_per_second": 1.888,
|
| 789 |
+
"step": 731
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 38.44,
|
| 793 |
+
"learning_rate": 1.5235457063711911e-06,
|
| 794 |
+
"loss": 0.0273,
|
| 795 |
+
"step": 740
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 38.96,
|
| 799 |
+
"learning_rate": 7.617728531855956e-07,
|
| 800 |
+
"loss": 0.0199,
|
| 801 |
+
"step": 750
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 38.96,
|
| 805 |
+
"eval_accuracy": 0.8,
|
| 806 |
+
"eval_loss": 1.100300669670105,
|
| 807 |
+
"eval_runtime": 1.0701,
|
| 808 |
+
"eval_samples_per_second": 56.068,
|
| 809 |
+
"eval_steps_per_second": 1.869,
|
| 810 |
+
"step": 750
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 39.48,
|
| 814 |
+
"learning_rate": 0.0,
|
| 815 |
+
"loss": 0.028,
|
| 816 |
+
"step": 760
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 39.48,
|
| 820 |
+
"eval_accuracy": 0.8,
|
| 821 |
+
"eval_loss": 1.0988926887512207,
|
| 822 |
+
"eval_runtime": 1.0717,
|
| 823 |
+
"eval_samples_per_second": 55.986,
|
| 824 |
+
"eval_steps_per_second": 1.866,
|
| 825 |
+
"step": 760
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 39.48,
|
| 829 |
+
"step": 760,
|
| 830 |
+
"total_flos": 7.496244493905936e+18,
|
| 831 |
+
"train_loss": 0.18620210740911333,
|
| 832 |
+
"train_runtime": 1570.0451,
|
| 833 |
+
"train_samples_per_second": 62.419,
|
| 834 |
+
"train_steps_per_second": 0.484
|
| 835 |
+
}
|
| 836 |
+
],
|
| 837 |
+
"logging_steps": 10,
|
| 838 |
+
"max_steps": 760,
|
| 839 |
+
"num_input_tokens_seen": 0,
|
| 840 |
+
"num_train_epochs": 40,
|
| 841 |
+
"save_steps": 500,
|
| 842 |
+
"total_flos": 7.496244493905936e+18,
|
| 843 |
+
"train_batch_size": 32,
|
| 844 |
+
"trial_name": null,
|
| 845 |
+
"trial_params": null
|
| 846 |
+
}
|