Commit
•
eb67e68
1
Parent(s):
feb840d
End of training
Browse files- README.md +3 -3
- all_results.json +11 -6
- eval_results.json +8 -0
- train_results.json +6 -6
- trainer_state.json +1400 -53
README.md
CHANGED
|
@@ -20,7 +20,7 @@ model-index:
|
|
| 20 |
metrics:
|
| 21 |
- name: Accuracy
|
| 22 |
type: accuracy
|
| 23 |
-
value: 0.
|
| 24 |
---
|
| 25 |
|
| 26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,8 +30,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
This model was trained from scratch on the imagefolder dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
-
- Loss: 1.
|
| 34 |
-
- Accuracy: 0.
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
|
|
|
| 20 |
metrics:
|
| 21 |
- name: Accuracy
|
| 22 |
type: accuracy
|
| 23 |
+
value: 0.5917085427135679
|
| 24 |
---
|
| 25 |
|
| 26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 30 |
|
| 31 |
This model was trained from scratch on the imagefolder dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: 1.0104
|
| 34 |
+
- Accuracy: 0.5917
|
| 35 |
|
| 36 |
## Model description
|
| 37 |
|
all_results.json
CHANGED
|
@@ -1,8 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.5917085427135679,
|
| 4 |
+
"eval_loss": 1.0104140043258667,
|
| 5 |
+
"eval_runtime": 6.3748,
|
| 6 |
+
"eval_samples_per_second": 249.734,
|
| 7 |
+
"eval_steps_per_second": 2.039,
|
| 8 |
+
"total_flos": 1.7807825640923136e+19,
|
| 9 |
+
"train_loss": 0.5565686808313642,
|
| 10 |
+
"train_runtime": 3248.0342,
|
| 11 |
+
"train_samples_per_second": 220.564,
|
| 12 |
+
"train_steps_per_second": 0.431
|
| 13 |
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.5917085427135679,
|
| 4 |
+
"eval_loss": 1.0104140043258667,
|
| 5 |
+
"eval_runtime": 6.3748,
|
| 6 |
+
"eval_samples_per_second": 249.734,
|
| 7 |
+
"eval_steps_per_second": 2.039
|
| 8 |
+
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"total_flos": 1.7807825640923136e+19,
|
| 4 |
+
"train_loss": 0.5565686808313642,
|
| 5 |
+
"train_runtime": 3248.0342,
|
| 6 |
+
"train_samples_per_second": 220.564,
|
| 7 |
+
"train_steps_per_second": 0.431
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,112 +1,1459 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.36,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.71,
|
| 20 |
-
"grad_norm": 2.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 1.0,
|
| 27 |
-
"eval_accuracy": 0.
|
| 28 |
-
"eval_loss": 1.
|
| 29 |
-
"eval_runtime":
|
| 30 |
-
"eval_samples_per_second":
|
| 31 |
-
"eval_steps_per_second": 2.
|
| 32 |
"step": 28
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.07,
|
| 36 |
-
"grad_norm": 2.
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss": 1.
|
| 39 |
"step": 30
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 1.43,
|
| 43 |
-
"grad_norm": 2.
|
| 44 |
-
"learning_rate":
|
| 45 |
-
"loss": 1.
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 1.79,
|
| 50 |
-
"grad_norm":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss": 1.
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 2.0,
|
| 57 |
-
"eval_accuracy": 0.
|
| 58 |
-
"eval_loss": 1.
|
| 59 |
-
"eval_runtime": 5.
|
| 60 |
-
"eval_samples_per_second":
|
| 61 |
-
"eval_steps_per_second": 2.
|
| 62 |
"step": 56
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 2.14,
|
| 66 |
-
"grad_norm":
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss": 1.
|
| 69 |
"step": 60
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.5,
|
| 73 |
-
"grad_norm":
|
| 74 |
-
"learning_rate":
|
| 75 |
-
"loss": 1.
|
| 76 |
"step": 70
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.86,
|
| 80 |
-
"grad_norm":
|
| 81 |
-
"learning_rate": 2.
|
| 82 |
-
"loss": 1.
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 3.0,
|
| 87 |
-
"eval_accuracy": 0.
|
| 88 |
-
"eval_loss": 1.
|
| 89 |
-
"eval_runtime": 5.
|
| 90 |
-
"eval_samples_per_second":
|
| 91 |
-
"eval_steps_per_second": 2.
|
| 92 |
"step": 84
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 3.
|
| 96 |
-
"
|
| 97 |
-
"
|
| 98 |
-
"
|
| 99 |
-
"
|
| 100 |
-
|
| 101 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
}
|
| 103 |
],
|
| 104 |
"logging_steps": 10,
|
| 105 |
-
"max_steps":
|
| 106 |
"num_input_tokens_seen": 0,
|
| 107 |
-
"num_train_epochs":
|
| 108 |
"save_steps": 500,
|
| 109 |
-
"total_flos": 1.
|
| 110 |
"train_batch_size": 128,
|
| 111 |
"trial_name": null,
|
| 112 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5917085427135679,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-224",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1400,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.36,
|
| 13 |
+
"grad_norm": 2.619962215423584,
|
| 14 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 15 |
+
"loss": 1.0415,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.71,
|
| 20 |
+
"grad_norm": 2.402837038040161,
|
| 21 |
+
"learning_rate": 7.142857142857143e-06,
|
| 22 |
+
"loss": 1.0339,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.5640703517587939,
|
| 28 |
+
"eval_loss": 1.0541082620620728,
|
| 29 |
+
"eval_runtime": 5.8961,
|
| 30 |
+
"eval_samples_per_second": 270.009,
|
| 31 |
+
"eval_steps_per_second": 2.205,
|
| 32 |
"step": 28
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.07,
|
| 36 |
+
"grad_norm": 2.223407745361328,
|
| 37 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 38 |
+
"loss": 1.025,
|
| 39 |
"step": 30
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 1.43,
|
| 43 |
+
"grad_norm": 2.5425541400909424,
|
| 44 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 45 |
+
"loss": 1.0072,
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 1.79,
|
| 50 |
+
"grad_norm": 3.219869375228882,
|
| 51 |
+
"learning_rate": 1.785714285714286e-05,
|
| 52 |
+
"loss": 1.0193,
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 2.0,
|
| 57 |
+
"eval_accuracy": 0.5621859296482412,
|
| 58 |
+
"eval_loss": 1.0463842153549194,
|
| 59 |
+
"eval_runtime": 5.8033,
|
| 60 |
+
"eval_samples_per_second": 274.326,
|
| 61 |
+
"eval_steps_per_second": 2.24,
|
| 62 |
"step": 56
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 2.14,
|
| 66 |
+
"grad_norm": 3.941281318664551,
|
| 67 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 68 |
+
"loss": 1.0133,
|
| 69 |
"step": 60
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.5,
|
| 73 |
+
"grad_norm": 4.169382095336914,
|
| 74 |
+
"learning_rate": 2.5e-05,
|
| 75 |
+
"loss": 1.0182,
|
| 76 |
"step": 70
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.86,
|
| 80 |
+
"grad_norm": 3.475947380065918,
|
| 81 |
+
"learning_rate": 2.857142857142857e-05,
|
| 82 |
+
"loss": 1.0348,
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 3.0,
|
| 87 |
+
"eval_accuracy": 0.5690954773869347,
|
| 88 |
+
"eval_loss": 1.0330955982208252,
|
| 89 |
+
"eval_runtime": 5.8594,
|
| 90 |
+
"eval_samples_per_second": 271.702,
|
| 91 |
+
"eval_steps_per_second": 2.219,
|
| 92 |
"step": 84
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 3.21,
|
| 96 |
+
"grad_norm": 3.1382639408111572,
|
| 97 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 98 |
+
"loss": 1.0266,
|
| 99 |
+
"step": 90
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 3.57,
|
| 103 |
+
"grad_norm": 2.344207525253296,
|
| 104 |
+
"learning_rate": 3.571428571428572e-05,
|
| 105 |
+
"loss": 1.0075,
|
| 106 |
+
"step": 100
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 3.93,
|
| 110 |
+
"grad_norm": 2.2122068405151367,
|
| 111 |
+
"learning_rate": 3.928571428571429e-05,
|
| 112 |
+
"loss": 1.0072,
|
| 113 |
+
"step": 110
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.0,
|
| 117 |
+
"eval_accuracy": 0.5847989949748744,
|
| 118 |
+
"eval_loss": 1.0253527164459229,
|
| 119 |
+
"eval_runtime": 5.8726,
|
| 120 |
+
"eval_samples_per_second": 271.091,
|
| 121 |
+
"eval_steps_per_second": 2.214,
|
| 122 |
+
"step": 112
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.29,
|
| 126 |
+
"grad_norm": 2.9957728385925293,
|
| 127 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 128 |
+
"loss": 1.0024,
|
| 129 |
+
"step": 120
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 4.64,
|
| 133 |
+
"grad_norm": 2.9330532550811768,
|
| 134 |
+
"learning_rate": 4.642857142857143e-05,
|
| 135 |
+
"loss": 0.991,
|
| 136 |
+
"step": 130
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 5.0,
|
| 140 |
+
"grad_norm": 2.355830430984497,
|
| 141 |
+
"learning_rate": 5e-05,
|
| 142 |
+
"loss": 0.9892,
|
| 143 |
+
"step": 140
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 5.0,
|
| 147 |
+
"eval_accuracy": 0.5753768844221105,
|
| 148 |
+
"eval_loss": 1.012104868888855,
|
| 149 |
+
"eval_runtime": 5.622,
|
| 150 |
+
"eval_samples_per_second": 283.175,
|
| 151 |
+
"eval_steps_per_second": 2.312,
|
| 152 |
+
"step": 140
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.36,
|
| 156 |
+
"grad_norm": 3.8320775032043457,
|
| 157 |
+
"learning_rate": 4.960317460317461e-05,
|
| 158 |
+
"loss": 0.9517,
|
| 159 |
+
"step": 150
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 5.71,
|
| 163 |
+
"grad_norm": 3.0432510375976562,
|
| 164 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 165 |
+
"loss": 0.9379,
|
| 166 |
+
"step": 160
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 6.0,
|
| 170 |
+
"eval_accuracy": 0.5810301507537688,
|
| 171 |
+
"eval_loss": 1.017525553703308,
|
| 172 |
+
"eval_runtime": 6.0814,
|
| 173 |
+
"eval_samples_per_second": 261.781,
|
| 174 |
+
"eval_steps_per_second": 2.138,
|
| 175 |
+
"step": 168
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 6.07,
|
| 179 |
+
"grad_norm": 3.3289706707000732,
|
| 180 |
+
"learning_rate": 4.880952380952381e-05,
|
| 181 |
+
"loss": 0.9475,
|
| 182 |
+
"step": 170
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.43,
|
| 186 |
+
"grad_norm": 3.335624933242798,
|
| 187 |
+
"learning_rate": 4.841269841269841e-05,
|
| 188 |
+
"loss": 0.9103,
|
| 189 |
+
"step": 180
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 6.79,
|
| 193 |
+
"grad_norm": 3.2209532260894775,
|
| 194 |
+
"learning_rate": 4.801587301587302e-05,
|
| 195 |
+
"loss": 0.9123,
|
| 196 |
+
"step": 190
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 7.0,
|
| 200 |
+
"eval_accuracy": 0.5866834170854272,
|
| 201 |
+
"eval_loss": 1.0119962692260742,
|
| 202 |
+
"eval_runtime": 5.8208,
|
| 203 |
+
"eval_samples_per_second": 273.502,
|
| 204 |
+
"eval_steps_per_second": 2.233,
|
| 205 |
+
"step": 196
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.14,
|
| 209 |
+
"grad_norm": 3.194115400314331,
|
| 210 |
+
"learning_rate": 4.761904761904762e-05,
|
| 211 |
+
"loss": 0.896,
|
| 212 |
+
"step": 200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.5,
|
| 216 |
+
"grad_norm": 2.89758563041687,
|
| 217 |
+
"learning_rate": 4.722222222222222e-05,
|
| 218 |
+
"loss": 0.9013,
|
| 219 |
+
"step": 210
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 7.86,
|
| 223 |
+
"grad_norm": 3.0895237922668457,
|
| 224 |
+
"learning_rate": 4.682539682539683e-05,
|
| 225 |
+
"loss": 0.8865,
|
| 226 |
+
"step": 220
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_accuracy": 0.5917085427135679,
|
| 231 |
+
"eval_loss": 1.0104140043258667,
|
| 232 |
+
"eval_runtime": 6.0762,
|
| 233 |
+
"eval_samples_per_second": 262.008,
|
| 234 |
+
"eval_steps_per_second": 2.14,
|
| 235 |
+
"step": 224
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 8.21,
|
| 239 |
+
"grad_norm": 2.715590715408325,
|
| 240 |
+
"learning_rate": 4.642857142857143e-05,
|
| 241 |
+
"loss": 0.8758,
|
| 242 |
+
"step": 230
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 8.57,
|
| 246 |
+
"grad_norm": 3.639813184738159,
|
| 247 |
+
"learning_rate": 4.603174603174603e-05,
|
| 248 |
+
"loss": 0.8516,
|
| 249 |
+
"step": 240
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 8.93,
|
| 253 |
+
"grad_norm": 3.1830241680145264,
|
| 254 |
+
"learning_rate": 4.563492063492064e-05,
|
| 255 |
+
"loss": 0.8668,
|
| 256 |
+
"step": 250
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.0,
|
| 260 |
+
"eval_accuracy": 0.5873115577889447,
|
| 261 |
+
"eval_loss": 1.0236343145370483,
|
| 262 |
+
"eval_runtime": 6.0393,
|
| 263 |
+
"eval_samples_per_second": 263.608,
|
| 264 |
+
"eval_steps_per_second": 2.153,
|
| 265 |
+
"step": 252
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 9.29,
|
| 269 |
+
"grad_norm": 3.1714346408843994,
|
| 270 |
+
"learning_rate": 4.523809523809524e-05,
|
| 271 |
+
"loss": 0.8352,
|
| 272 |
+
"step": 260
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 9.64,
|
| 276 |
+
"grad_norm": 3.9194183349609375,
|
| 277 |
+
"learning_rate": 4.4841269841269846e-05,
|
| 278 |
+
"loss": 0.8207,
|
| 279 |
+
"step": 270
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 10.0,
|
| 283 |
+
"grad_norm": 4.183494567871094,
|
| 284 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 285 |
+
"loss": 0.8189,
|
| 286 |
+
"step": 280
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 10.0,
|
| 290 |
+
"eval_accuracy": 0.5829145728643216,
|
| 291 |
+
"eval_loss": 1.0359796285629272,
|
| 292 |
+
"eval_runtime": 6.1118,
|
| 293 |
+
"eval_samples_per_second": 260.481,
|
| 294 |
+
"eval_steps_per_second": 2.127,
|
| 295 |
+
"step": 280
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 10.36,
|
| 299 |
+
"grad_norm": 3.281873941421509,
|
| 300 |
+
"learning_rate": 4.404761904761905e-05,
|
| 301 |
+
"loss": 0.8051,
|
| 302 |
+
"step": 290
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 10.71,
|
| 306 |
+
"grad_norm": 2.9959723949432373,
|
| 307 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 308 |
+
"loss": 0.7933,
|
| 309 |
+
"step": 300
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 11.0,
|
| 313 |
+
"eval_accuracy": 0.5835427135678392,
|
| 314 |
+
"eval_loss": 1.039525032043457,
|
| 315 |
+
"eval_runtime": 5.9823,
|
| 316 |
+
"eval_samples_per_second": 266.118,
|
| 317 |
+
"eval_steps_per_second": 2.173,
|
| 318 |
+
"step": 308
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 11.07,
|
| 322 |
+
"grad_norm": 3.470358371734619,
|
| 323 |
+
"learning_rate": 4.3253968253968256e-05,
|
| 324 |
+
"loss": 0.797,
|
| 325 |
+
"step": 310
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 11.43,
|
| 329 |
+
"grad_norm": 3.285288095474243,
|
| 330 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 331 |
+
"loss": 0.7723,
|
| 332 |
+
"step": 320
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 11.79,
|
| 336 |
+
"grad_norm": 4.59849739074707,
|
| 337 |
+
"learning_rate": 4.2460317460317464e-05,
|
| 338 |
+
"loss": 0.7765,
|
| 339 |
+
"step": 330
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 12.0,
|
| 343 |
+
"eval_accuracy": 0.5728643216080402,
|
| 344 |
+
"eval_loss": 1.0594085454940796,
|
| 345 |
+
"eval_runtime": 5.9856,
|
| 346 |
+
"eval_samples_per_second": 265.972,
|
| 347 |
+
"eval_steps_per_second": 2.172,
|
| 348 |
+
"step": 336
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 12.14,
|
| 352 |
+
"grad_norm": 4.7830305099487305,
|
| 353 |
+
"learning_rate": 4.2063492063492065e-05,
|
| 354 |
+
"loss": 0.7567,
|
| 355 |
+
"step": 340
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 12.5,
|
| 359 |
+
"grad_norm": 3.7676315307617188,
|
| 360 |
+
"learning_rate": 4.166666666666667e-05,
|
| 361 |
+
"loss": 0.738,
|
| 362 |
+
"step": 350
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 12.86,
|
| 366 |
+
"grad_norm": 4.092833518981934,
|
| 367 |
+
"learning_rate": 4.126984126984127e-05,
|
| 368 |
+
"loss": 0.7538,
|
| 369 |
+
"step": 360
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 13.0,
|
| 373 |
+
"eval_accuracy": 0.5879396984924623,
|
| 374 |
+
"eval_loss": 1.0552036762237549,
|
| 375 |
+
"eval_runtime": 5.8071,
|
| 376 |
+
"eval_samples_per_second": 274.146,
|
| 377 |
+
"eval_steps_per_second": 2.239,
|
| 378 |
+
"step": 364
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 13.21,
|
| 382 |
+
"grad_norm": 4.582472801208496,
|
| 383 |
+
"learning_rate": 4.0873015873015874e-05,
|
| 384 |
+
"loss": 0.7175,
|
| 385 |
+
"step": 370
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 13.57,
|
| 389 |
+
"grad_norm": 4.479797840118408,
|
| 390 |
+
"learning_rate": 4.047619047619048e-05,
|
| 391 |
+
"loss": 0.7101,
|
| 392 |
+
"step": 380
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 13.93,
|
| 396 |
+
"grad_norm": 4.223387241363525,
|
| 397 |
+
"learning_rate": 4.007936507936508e-05,
|
| 398 |
+
"loss": 0.7146,
|
| 399 |
+
"step": 390
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 14.0,
|
| 403 |
+
"eval_accuracy": 0.5829145728643216,
|
| 404 |
+
"eval_loss": 1.0619994401931763,
|
| 405 |
+
"eval_runtime": 5.8703,
|
| 406 |
+
"eval_samples_per_second": 271.194,
|
| 407 |
+
"eval_steps_per_second": 2.215,
|
| 408 |
+
"step": 392
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.29,
|
| 412 |
+
"grad_norm": 4.616257190704346,
|
| 413 |
+
"learning_rate": 3.968253968253968e-05,
|
| 414 |
+
"loss": 0.7031,
|
| 415 |
+
"step": 400
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 14.64,
|
| 419 |
+
"grad_norm": 3.6395843029022217,
|
| 420 |
+
"learning_rate": 3.928571428571429e-05,
|
| 421 |
+
"loss": 0.6901,
|
| 422 |
+
"step": 410
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 15.0,
|
| 426 |
+
"grad_norm": 3.426710367202759,
|
| 427 |
+
"learning_rate": 3.888888888888889e-05,
|
| 428 |
+
"loss": 0.6885,
|
| 429 |
+
"step": 420
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 15.0,
|
| 433 |
+
"eval_accuracy": 0.5841708542713567,
|
| 434 |
+
"eval_loss": 1.0782768726348877,
|
| 435 |
+
"eval_runtime": 6.0171,
|
| 436 |
+
"eval_samples_per_second": 264.578,
|
| 437 |
+
"eval_steps_per_second": 2.16,
|
| 438 |
+
"step": 420
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 15.36,
|
| 442 |
+
"grad_norm": 3.5050458908081055,
|
| 443 |
+
"learning_rate": 3.84920634920635e-05,
|
| 444 |
+
"loss": 0.6697,
|
| 445 |
+
"step": 430
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 15.71,
|
| 449 |
+
"grad_norm": 4.314873218536377,
|
| 450 |
+
"learning_rate": 3.809523809523809e-05,
|
| 451 |
+
"loss": 0.6556,
|
| 452 |
+
"step": 440
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 16.0,
|
| 456 |
+
"eval_accuracy": 0.5816582914572864,
|
| 457 |
+
"eval_loss": 1.1010228395462036,
|
| 458 |
+
"eval_runtime": 6.002,
|
| 459 |
+
"eval_samples_per_second": 265.244,
|
| 460 |
+
"eval_steps_per_second": 2.166,
|
| 461 |
+
"step": 448
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 16.07,
|
| 465 |
+
"grad_norm": 3.6204850673675537,
|
| 466 |
+
"learning_rate": 3.76984126984127e-05,
|
| 467 |
+
"loss": 0.6621,
|
| 468 |
+
"step": 450
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 16.43,
|
| 472 |
+
"grad_norm": 5.87465238571167,
|
| 473 |
+
"learning_rate": 3.730158730158731e-05,
|
| 474 |
+
"loss": 0.6388,
|
| 475 |
+
"step": 460
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 16.79,
|
| 479 |
+
"grad_norm": 3.773968458175659,
|
| 480 |
+
"learning_rate": 3.690476190476191e-05,
|
| 481 |
+
"loss": 0.6453,
|
| 482 |
+
"step": 470
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 17.0,
|
| 486 |
+
"eval_accuracy": 0.5734924623115578,
|
| 487 |
+
"eval_loss": 1.1130640506744385,
|
| 488 |
+
"eval_runtime": 5.9355,
|
| 489 |
+
"eval_samples_per_second": 268.215,
|
| 490 |
+
"eval_steps_per_second": 2.19,
|
| 491 |
+
"step": 476
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 17.14,
|
| 495 |
+
"grad_norm": 3.7962214946746826,
|
| 496 |
+
"learning_rate": 3.650793650793651e-05,
|
| 497 |
+
"loss": 0.6504,
|
| 498 |
+
"step": 480
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 17.5,
|
| 502 |
+
"grad_norm": 4.2159013748168945,
|
| 503 |
+
"learning_rate": 3.611111111111111e-05,
|
| 504 |
+
"loss": 0.6058,
|
| 505 |
+
"step": 490
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 17.86,
|
| 509 |
+
"grad_norm": 4.258510112762451,
|
| 510 |
+
"learning_rate": 3.571428571428572e-05,
|
| 511 |
+
"loss": 0.6175,
|
| 512 |
+
"step": 500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 18.0,
|
| 516 |
+
"eval_accuracy": 0.5891959798994975,
|
| 517 |
+
"eval_loss": 1.1074261665344238,
|
| 518 |
+
"eval_runtime": 5.4208,
|
| 519 |
+
"eval_samples_per_second": 293.686,
|
| 520 |
+
"eval_steps_per_second": 2.398,
|
| 521 |
+
"step": 504
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 18.21,
|
| 525 |
+
"grad_norm": 4.8816914558410645,
|
| 526 |
+
"learning_rate": 3.5317460317460324e-05,
|
| 527 |
+
"loss": 0.5942,
|
| 528 |
+
"step": 510
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 18.57,
|
| 532 |
+
"grad_norm": 4.609433650970459,
|
| 533 |
+
"learning_rate": 3.492063492063492e-05,
|
| 534 |
+
"loss": 0.5943,
|
| 535 |
+
"step": 520
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 18.93,
|
| 539 |
+
"grad_norm": 7.7561211585998535,
|
| 540 |
+
"learning_rate": 3.4523809523809526e-05,
|
| 541 |
+
"loss": 0.5993,
|
| 542 |
+
"step": 530
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 19.0,
|
| 546 |
+
"eval_accuracy": 0.5741206030150754,
|
| 547 |
+
"eval_loss": 1.1327540874481201,
|
| 548 |
+
"eval_runtime": 5.7963,
|
| 549 |
+
"eval_samples_per_second": 274.658,
|
| 550 |
+
"eval_steps_per_second": 2.243,
|
| 551 |
+
"step": 532
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 19.29,
|
| 555 |
+
"grad_norm": 4.787009239196777,
|
| 556 |
+
"learning_rate": 3.412698412698413e-05,
|
| 557 |
+
"loss": 0.5741,
|
| 558 |
+
"step": 540
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 19.64,
|
| 562 |
+
"grad_norm": 7.622641086578369,
|
| 563 |
+
"learning_rate": 3.3730158730158734e-05,
|
| 564 |
+
"loss": 0.5803,
|
| 565 |
+
"step": 550
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 20.0,
|
| 569 |
+
"grad_norm": 6.530401706695557,
|
| 570 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 571 |
+
"loss": 0.5683,
|
| 572 |
+
"step": 560
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 20.0,
|
| 576 |
+
"eval_accuracy": 0.5791457286432161,
|
| 577 |
+
"eval_loss": 1.1423227787017822,
|
| 578 |
+
"eval_runtime": 5.7458,
|
| 579 |
+
"eval_samples_per_second": 277.073,
|
| 580 |
+
"eval_steps_per_second": 2.263,
|
| 581 |
+
"step": 560
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 20.36,
|
| 585 |
+
"grad_norm": 4.021312236785889,
|
| 586 |
+
"learning_rate": 3.2936507936507936e-05,
|
| 587 |
+
"loss": 0.5457,
|
| 588 |
+
"step": 570
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 20.71,
|
| 592 |
+
"grad_norm": 5.022521018981934,
|
| 593 |
+
"learning_rate": 3.253968253968254e-05,
|
| 594 |
+
"loss": 0.5524,
|
| 595 |
+
"step": 580
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 21.0,
|
| 599 |
+
"eval_accuracy": 0.5873115577889447,
|
| 600 |
+
"eval_loss": 1.1516565084457397,
|
| 601 |
+
"eval_runtime": 6.4553,
|
| 602 |
+
"eval_samples_per_second": 246.619,
|
| 603 |
+
"eval_steps_per_second": 2.014,
|
| 604 |
+
"step": 588
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 21.07,
|
| 608 |
+
"grad_norm": 4.585721969604492,
|
| 609 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 610 |
+
"loss": 0.5493,
|
| 611 |
+
"step": 590
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 21.43,
|
| 615 |
+
"grad_norm": 3.6502797603607178,
|
| 616 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 617 |
+
"loss": 0.5393,
|
| 618 |
+
"step": 600
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 21.79,
|
| 622 |
+
"grad_norm": 4.494349002838135,
|
| 623 |
+
"learning_rate": 3.134920634920635e-05,
|
| 624 |
+
"loss": 0.5151,
|
| 625 |
+
"step": 610
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 22.0,
|
| 629 |
+
"eval_accuracy": 0.5766331658291457,
|
| 630 |
+
"eval_loss": 1.1673014163970947,
|
| 631 |
+
"eval_runtime": 6.0498,
|
| 632 |
+
"eval_samples_per_second": 263.15,
|
| 633 |
+
"eval_steps_per_second": 2.149,
|
| 634 |
+
"step": 616
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 22.14,
|
| 638 |
+
"grad_norm": 4.292243957519531,
|
| 639 |
+
"learning_rate": 3.095238095238095e-05,
|
| 640 |
+
"loss": 0.5241,
|
| 641 |
+
"step": 620
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 22.5,
|
| 645 |
+
"grad_norm": 4.669267177581787,
|
| 646 |
+
"learning_rate": 3.055555555555556e-05,
|
| 647 |
+
"loss": 0.5095,
|
| 648 |
+
"step": 630
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 22.86,
|
| 652 |
+
"grad_norm": 4.2560882568359375,
|
| 653 |
+
"learning_rate": 3.0158730158730158e-05,
|
| 654 |
+
"loss": 0.5096,
|
| 655 |
+
"step": 640
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 23.0,
|
| 659 |
+
"eval_accuracy": 0.5797738693467337,
|
| 660 |
+
"eval_loss": 1.17599618434906,
|
| 661 |
+
"eval_runtime": 5.8279,
|
| 662 |
+
"eval_samples_per_second": 273.169,
|
| 663 |
+
"eval_steps_per_second": 2.231,
|
| 664 |
+
"step": 644
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 23.21,
|
| 668 |
+
"grad_norm": 4.331119537353516,
|
| 669 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 670 |
+
"loss": 0.5092,
|
| 671 |
+
"step": 650
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 23.57,
|
| 675 |
+
"grad_norm": 5.246380805969238,
|
| 676 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 677 |
+
"loss": 0.5016,
|
| 678 |
+
"step": 660
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 23.93,
|
| 682 |
+
"grad_norm": 4.598259449005127,
|
| 683 |
+
"learning_rate": 2.8968253968253974e-05,
|
| 684 |
+
"loss": 0.4937,
|
| 685 |
+
"step": 670
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 24.0,
|
| 689 |
+
"eval_accuracy": 0.5816582914572864,
|
| 690 |
+
"eval_loss": 1.193081259727478,
|
| 691 |
+
"eval_runtime": 5.7093,
|
| 692 |
+
"eval_samples_per_second": 278.844,
|
| 693 |
+
"eval_steps_per_second": 2.277,
|
| 694 |
+
"step": 672
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 24.29,
|
| 698 |
+
"grad_norm": 4.1697893142700195,
|
| 699 |
+
"learning_rate": 2.857142857142857e-05,
|
| 700 |
+
"loss": 0.4713,
|
| 701 |
+
"step": 680
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 24.64,
|
| 705 |
+
"grad_norm": 3.761561870574951,
|
| 706 |
+
"learning_rate": 2.8174603174603175e-05,
|
| 707 |
+
"loss": 0.469,
|
| 708 |
+
"step": 690
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 25.0,
|
| 712 |
+
"grad_norm": 3.7730281352996826,
|
| 713 |
+
"learning_rate": 2.777777777777778e-05,
|
| 714 |
+
"loss": 0.487,
|
| 715 |
+
"step": 700
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 25.0,
|
| 719 |
+
"eval_accuracy": 0.5734924623115578,
|
| 720 |
+
"eval_loss": 1.2083638906478882,
|
| 721 |
+
"eval_runtime": 6.0189,
|
| 722 |
+
"eval_samples_per_second": 264.498,
|
| 723 |
+
"eval_steps_per_second": 2.16,
|
| 724 |
+
"step": 700
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 25.36,
|
| 728 |
+
"grad_norm": 4.0658674240112305,
|
| 729 |
+
"learning_rate": 2.7380952380952383e-05,
|
| 730 |
+
"loss": 0.4587,
|
| 731 |
+
"step": 710
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 25.71,
|
| 735 |
+
"grad_norm": 3.6383256912231445,
|
| 736 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 737 |
+
"loss": 0.4597,
|
| 738 |
+
"step": 720
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 26.0,
|
| 742 |
+
"eval_accuracy": 0.571608040201005,
|
| 743 |
+
"eval_loss": 1.2270249128341675,
|
| 744 |
+
"eval_runtime": 6.2134,
|
| 745 |
+
"eval_samples_per_second": 256.222,
|
| 746 |
+
"eval_steps_per_second": 2.092,
|
| 747 |
+
"step": 728
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 26.07,
|
| 751 |
+
"grad_norm": 4.403080940246582,
|
| 752 |
+
"learning_rate": 2.6587301587301588e-05,
|
| 753 |
+
"loss": 0.4461,
|
| 754 |
+
"step": 730
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 26.43,
|
| 758 |
+
"grad_norm": 4.59806489944458,
|
| 759 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 760 |
+
"loss": 0.4487,
|
| 761 |
+
"step": 740
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 26.79,
|
| 765 |
+
"grad_norm": 3.9561307430267334,
|
| 766 |
+
"learning_rate": 2.5793650793650796e-05,
|
| 767 |
+
"loss": 0.4482,
|
| 768 |
+
"step": 750
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 27.0,
|
| 772 |
+
"eval_accuracy": 0.5829145728643216,
|
| 773 |
+
"eval_loss": 1.2389401197433472,
|
| 774 |
+
"eval_runtime": 5.923,
|
| 775 |
+
"eval_samples_per_second": 268.783,
|
| 776 |
+
"eval_steps_per_second": 2.195,
|
| 777 |
+
"step": 756
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 27.14,
|
| 781 |
+
"grad_norm": 4.864688396453857,
|
| 782 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 783 |
+
"loss": 0.4678,
|
| 784 |
+
"step": 760
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 27.5,
|
| 788 |
+
"grad_norm": 11.65251636505127,
|
| 789 |
+
"learning_rate": 2.5e-05,
|
| 790 |
+
"loss": 0.4347,
|
| 791 |
+
"step": 770
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 27.86,
|
| 795 |
+
"grad_norm": 4.065871715545654,
|
| 796 |
+
"learning_rate": 2.4603174603174602e-05,
|
| 797 |
+
"loss": 0.4183,
|
| 798 |
+
"step": 780
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 28.0,
|
| 802 |
+
"eval_accuracy": 0.5772613065326633,
|
| 803 |
+
"eval_loss": 1.2430446147918701,
|
| 804 |
+
"eval_runtime": 5.8698,
|
| 805 |
+
"eval_samples_per_second": 271.218,
|
| 806 |
+
"eval_steps_per_second": 2.215,
|
| 807 |
+
"step": 784
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 28.21,
|
| 811 |
+
"grad_norm": 3.8666417598724365,
|
| 812 |
+
"learning_rate": 2.4206349206349206e-05,
|
| 813 |
+
"loss": 0.42,
|
| 814 |
+
"step": 790
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 28.57,
|
| 818 |
+
"grad_norm": 4.105574131011963,
|
| 819 |
+
"learning_rate": 2.380952380952381e-05,
|
| 820 |
+
"loss": 0.4359,
|
| 821 |
+
"step": 800
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 28.93,
|
| 825 |
+
"grad_norm": 4.677915573120117,
|
| 826 |
+
"learning_rate": 2.3412698412698414e-05,
|
| 827 |
+
"loss": 0.4228,
|
| 828 |
+
"step": 810
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 29.0,
|
| 832 |
+
"eval_accuracy": 0.5741206030150754,
|
| 833 |
+
"eval_loss": 1.2637208700180054,
|
| 834 |
+
"eval_runtime": 5.5801,
|
| 835 |
+
"eval_samples_per_second": 285.302,
|
| 836 |
+
"eval_steps_per_second": 2.33,
|
| 837 |
+
"step": 812
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 29.29,
|
| 841 |
+
"grad_norm": 4.680498123168945,
|
| 842 |
+
"learning_rate": 2.3015873015873015e-05,
|
| 843 |
+
"loss": 0.4268,
|
| 844 |
+
"step": 820
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 29.64,
|
| 848 |
+
"grad_norm": 3.9155895709991455,
|
| 849 |
+
"learning_rate": 2.261904761904762e-05,
|
| 850 |
+
"loss": 0.3908,
|
| 851 |
+
"step": 830
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 30.0,
|
| 855 |
+
"grad_norm": 4.745033264160156,
|
| 856 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 857 |
+
"loss": 0.4116,
|
| 858 |
+
"step": 840
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 30.0,
|
| 862 |
+
"eval_accuracy": 0.5778894472361809,
|
| 863 |
+
"eval_loss": 1.2687839269638062,
|
| 864 |
+
"eval_runtime": 5.7938,
|
| 865 |
+
"eval_samples_per_second": 274.776,
|
| 866 |
+
"eval_steps_per_second": 2.244,
|
| 867 |
+
"step": 840
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 30.36,
|
| 871 |
+
"grad_norm": 4.2474236488342285,
|
| 872 |
+
"learning_rate": 2.1825396825396827e-05,
|
| 873 |
+
"loss": 0.3889,
|
| 874 |
+
"step": 850
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 30.71,
|
| 878 |
+
"grad_norm": 4.243285655975342,
|
| 879 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 880 |
+
"loss": 0.3942,
|
| 881 |
+
"step": 860
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 31.0,
|
| 885 |
+
"eval_accuracy": 0.5879396984924623,
|
| 886 |
+
"eval_loss": 1.2986161708831787,
|
| 887 |
+
"eval_runtime": 5.7249,
|
| 888 |
+
"eval_samples_per_second": 278.081,
|
| 889 |
+
"eval_steps_per_second": 2.271,
|
| 890 |
+
"step": 868
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 31.07,
|
| 894 |
+
"grad_norm": 4.418118000030518,
|
| 895 |
+
"learning_rate": 2.1031746031746032e-05,
|
| 896 |
+
"loss": 0.3868,
|
| 897 |
+
"step": 870
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 31.43,
|
| 901 |
+
"grad_norm": 4.63828182220459,
|
| 902 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 903 |
+
"loss": 0.3935,
|
| 904 |
+
"step": 880
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 31.79,
|
| 908 |
+
"grad_norm": 4.932127952575684,
|
| 909 |
+
"learning_rate": 2.023809523809524e-05,
|
| 910 |
+
"loss": 0.3815,
|
| 911 |
+
"step": 890
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 32.0,
|
| 915 |
+
"eval_accuracy": 0.5766331658291457,
|
| 916 |
+
"eval_loss": 1.2911109924316406,
|
| 917 |
+
"eval_runtime": 5.5086,
|
| 918 |
+
"eval_samples_per_second": 289.004,
|
| 919 |
+
"eval_steps_per_second": 2.36,
|
| 920 |
+
"step": 896
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 32.14,
|
| 924 |
+
"grad_norm": 5.120588779449463,
|
| 925 |
+
"learning_rate": 1.984126984126984e-05,
|
| 926 |
+
"loss": 0.3814,
|
| 927 |
+
"step": 900
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 32.5,
|
| 931 |
+
"grad_norm": 4.348486423492432,
|
| 932 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 933 |
+
"loss": 0.3891,
|
| 934 |
+
"step": 910
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 32.86,
|
| 938 |
+
"grad_norm": 5.042387008666992,
|
| 939 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 940 |
+
"loss": 0.3828,
|
| 941 |
+
"step": 920
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 33.0,
|
| 945 |
+
"eval_accuracy": 0.5772613065326633,
|
| 946 |
+
"eval_loss": 1.311328411102295,
|
| 947 |
+
"eval_runtime": 5.8619,
|
| 948 |
+
"eval_samples_per_second": 271.584,
|
| 949 |
+
"eval_steps_per_second": 2.218,
|
| 950 |
+
"step": 924
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 33.21,
|
| 954 |
+
"grad_norm": 4.366072177886963,
|
| 955 |
+
"learning_rate": 1.8650793650793654e-05,
|
| 956 |
+
"loss": 0.3692,
|
| 957 |
+
"step": 930
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 33.57,
|
| 961 |
+
"grad_norm": 3.8073909282684326,
|
| 962 |
+
"learning_rate": 1.8253968253968254e-05,
|
| 963 |
+
"loss": 0.3617,
|
| 964 |
+
"step": 940
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 33.93,
|
| 968 |
+
"grad_norm": 4.39243221282959,
|
| 969 |
+
"learning_rate": 1.785714285714286e-05,
|
| 970 |
+
"loss": 0.3791,
|
| 971 |
+
"step": 950
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 34.0,
|
| 975 |
+
"eval_accuracy": 0.5766331658291457,
|
| 976 |
+
"eval_loss": 1.3316831588745117,
|
| 977 |
+
"eval_runtime": 5.7518,
|
| 978 |
+
"eval_samples_per_second": 276.782,
|
| 979 |
+
"eval_steps_per_second": 2.26,
|
| 980 |
+
"step": 952
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 34.29,
|
| 984 |
+
"grad_norm": 4.339027404785156,
|
| 985 |
+
"learning_rate": 1.746031746031746e-05,
|
| 986 |
+
"loss": 0.3716,
|
| 987 |
+
"step": 960
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 34.64,
|
| 991 |
+
"grad_norm": 4.211394786834717,
|
| 992 |
+
"learning_rate": 1.7063492063492063e-05,
|
| 993 |
+
"loss": 0.3644,
|
| 994 |
+
"step": 970
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 35.0,
|
| 998 |
+
"grad_norm": 4.398725509643555,
|
| 999 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 1000 |
+
"loss": 0.3701,
|
| 1001 |
+
"step": 980
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 35.0,
|
| 1005 |
+
"eval_accuracy": 0.5772613065326633,
|
| 1006 |
+
"eval_loss": 1.3383643627166748,
|
| 1007 |
+
"eval_runtime": 5.7692,
|
| 1008 |
+
"eval_samples_per_second": 275.948,
|
| 1009 |
+
"eval_steps_per_second": 2.253,
|
| 1010 |
+
"step": 980
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 35.36,
|
| 1014 |
+
"grad_norm": 5.39536714553833,
|
| 1015 |
+
"learning_rate": 1.626984126984127e-05,
|
| 1016 |
+
"loss": 0.3625,
|
| 1017 |
+
"step": 990
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 35.71,
|
| 1021 |
+
"grad_norm": 4.096094131469727,
|
| 1022 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 1023 |
+
"loss": 0.3566,
|
| 1024 |
+
"step": 1000
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 36.0,
|
| 1028 |
+
"eval_accuracy": 0.5753768844221105,
|
| 1029 |
+
"eval_loss": 1.3406319618225098,
|
| 1030 |
+
"eval_runtime": 5.7402,
|
| 1031 |
+
"eval_samples_per_second": 277.342,
|
| 1032 |
+
"eval_steps_per_second": 2.265,
|
| 1033 |
+
"step": 1008
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 36.07,
|
| 1037 |
+
"grad_norm": 4.44684362411499,
|
| 1038 |
+
"learning_rate": 1.5476190476190476e-05,
|
| 1039 |
+
"loss": 0.3621,
|
| 1040 |
+
"step": 1010
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 36.43,
|
| 1044 |
+
"grad_norm": 4.413670063018799,
|
| 1045 |
+
"learning_rate": 1.5079365079365079e-05,
|
| 1046 |
+
"loss": 0.3435,
|
| 1047 |
+
"step": 1020
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 36.79,
|
| 1051 |
+
"grad_norm": 4.409304618835449,
|
| 1052 |
+
"learning_rate": 1.4682539682539683e-05,
|
| 1053 |
+
"loss": 0.3551,
|
| 1054 |
+
"step": 1030
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 37.0,
|
| 1058 |
+
"eval_accuracy": 0.5766331658291457,
|
| 1059 |
+
"eval_loss": 1.34104323387146,
|
| 1060 |
+
"eval_runtime": 5.5746,
|
| 1061 |
+
"eval_samples_per_second": 285.58,
|
| 1062 |
+
"eval_steps_per_second": 2.332,
|
| 1063 |
+
"step": 1036
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 37.14,
|
| 1067 |
+
"grad_norm": 4.187081813812256,
|
| 1068 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 1069 |
+
"loss": 0.3558,
|
| 1070 |
+
"step": 1040
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 37.5,
|
| 1074 |
+
"grad_norm": 4.235673427581787,
|
| 1075 |
+
"learning_rate": 1.388888888888889e-05,
|
| 1076 |
+
"loss": 0.3392,
|
| 1077 |
+
"step": 1050
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 37.86,
|
| 1081 |
+
"grad_norm": 4.1030497550964355,
|
| 1082 |
+
"learning_rate": 1.3492063492063492e-05,
|
| 1083 |
+
"loss": 0.3487,
|
| 1084 |
+
"step": 1060
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 38.0,
|
| 1088 |
+
"eval_accuracy": 0.5866834170854272,
|
| 1089 |
+
"eval_loss": 1.3364226818084717,
|
| 1090 |
+
"eval_runtime": 5.6497,
|
| 1091 |
+
"eval_samples_per_second": 281.783,
|
| 1092 |
+
"eval_steps_per_second": 2.301,
|
| 1093 |
+
"step": 1064
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 38.21,
|
| 1097 |
+
"grad_norm": 5.5322394371032715,
|
| 1098 |
+
"learning_rate": 1.3095238095238096e-05,
|
| 1099 |
+
"loss": 0.3359,
|
| 1100 |
+
"step": 1070
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 38.57,
|
| 1104 |
+
"grad_norm": 4.478665828704834,
|
| 1105 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 1106 |
+
"loss": 0.3445,
|
| 1107 |
+
"step": 1080
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 38.93,
|
| 1111 |
+
"grad_norm": 4.051063060760498,
|
| 1112 |
+
"learning_rate": 1.2301587301587301e-05,
|
| 1113 |
+
"loss": 0.3463,
|
| 1114 |
+
"step": 1090
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 39.0,
|
| 1118 |
+
"eval_accuracy": 0.5810301507537688,
|
| 1119 |
+
"eval_loss": 1.3495649099349976,
|
| 1120 |
+
"eval_runtime": 5.6668,
|
| 1121 |
+
"eval_samples_per_second": 280.934,
|
| 1122 |
+
"eval_steps_per_second": 2.294,
|
| 1123 |
+
"step": 1092
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 39.29,
|
| 1127 |
+
"grad_norm": 3.2469401359558105,
|
| 1128 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 1129 |
+
"loss": 0.3348,
|
| 1130 |
+
"step": 1100
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 39.64,
|
| 1134 |
+
"grad_norm": 3.4843826293945312,
|
| 1135 |
+
"learning_rate": 1.1507936507936508e-05,
|
| 1136 |
+
"loss": 0.3162,
|
| 1137 |
+
"step": 1110
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 40.0,
|
| 1141 |
+
"grad_norm": 4.144495010375977,
|
| 1142 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 1143 |
+
"loss": 0.3242,
|
| 1144 |
+
"step": 1120
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 40.0,
|
| 1148 |
+
"eval_accuracy": 0.574748743718593,
|
| 1149 |
+
"eval_loss": 1.3639838695526123,
|
| 1150 |
+
"eval_runtime": 5.8819,
|
| 1151 |
+
"eval_samples_per_second": 270.661,
|
| 1152 |
+
"eval_steps_per_second": 2.21,
|
| 1153 |
+
"step": 1120
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 40.36,
|
| 1157 |
+
"grad_norm": 4.064784526824951,
|
| 1158 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 1159 |
+
"loss": 0.3376,
|
| 1160 |
+
"step": 1130
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 40.71,
|
| 1164 |
+
"grad_norm": 3.94600248336792,
|
| 1165 |
+
"learning_rate": 1.0317460317460318e-05,
|
| 1166 |
+
"loss": 0.3308,
|
| 1167 |
+
"step": 1140
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 41.0,
|
| 1171 |
+
"eval_accuracy": 0.571608040201005,
|
| 1172 |
+
"eval_loss": 1.3626537322998047,
|
| 1173 |
+
"eval_runtime": 5.5788,
|
| 1174 |
+
"eval_samples_per_second": 285.366,
|
| 1175 |
+
"eval_steps_per_second": 2.33,
|
| 1176 |
+
"step": 1148
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 41.07,
|
| 1180 |
+
"grad_norm": 3.8364109992980957,
|
| 1181 |
+
"learning_rate": 9.92063492063492e-06,
|
| 1182 |
+
"loss": 0.3196,
|
| 1183 |
+
"step": 1150
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 41.43,
|
| 1187 |
+
"grad_norm": 3.929502010345459,
|
| 1188 |
+
"learning_rate": 9.523809523809523e-06,
|
| 1189 |
+
"loss": 0.3224,
|
| 1190 |
+
"step": 1160
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 41.79,
|
| 1194 |
+
"grad_norm": 4.359261989593506,
|
| 1195 |
+
"learning_rate": 9.126984126984127e-06,
|
| 1196 |
+
"loss": 0.3255,
|
| 1197 |
+
"step": 1170
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 42.0,
|
| 1201 |
+
"eval_accuracy": 0.5804020100502513,
|
| 1202 |
+
"eval_loss": 1.379508376121521,
|
| 1203 |
+
"eval_runtime": 5.7679,
|
| 1204 |
+
"eval_samples_per_second": 276.011,
|
| 1205 |
+
"eval_steps_per_second": 2.254,
|
| 1206 |
+
"step": 1176
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 42.14,
|
| 1210 |
+
"grad_norm": 4.006887912750244,
|
| 1211 |
+
"learning_rate": 8.73015873015873e-06,
|
| 1212 |
+
"loss": 0.3268,
|
| 1213 |
+
"step": 1180
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 42.5,
|
| 1217 |
+
"grad_norm": 3.5152969360351562,
|
| 1218 |
+
"learning_rate": 8.333333333333334e-06,
|
| 1219 |
+
"loss": 0.3166,
|
| 1220 |
+
"step": 1190
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 42.86,
|
| 1224 |
+
"grad_norm": 3.868173122406006,
|
| 1225 |
+
"learning_rate": 7.936507936507936e-06,
|
| 1226 |
+
"loss": 0.3295,
|
| 1227 |
+
"step": 1200
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 43.0,
|
| 1231 |
+
"eval_accuracy": 0.5797738693467337,
|
| 1232 |
+
"eval_loss": 1.374683141708374,
|
| 1233 |
+
"eval_runtime": 5.5853,
|
| 1234 |
+
"eval_samples_per_second": 285.032,
|
| 1235 |
+
"eval_steps_per_second": 2.328,
|
| 1236 |
+
"step": 1204
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 43.21,
|
| 1240 |
+
"grad_norm": 3.3808319568634033,
|
| 1241 |
+
"learning_rate": 7.5396825396825394e-06,
|
| 1242 |
+
"loss": 0.3201,
|
| 1243 |
+
"step": 1210
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 43.57,
|
| 1247 |
+
"grad_norm": 4.094415187835693,
|
| 1248 |
+
"learning_rate": 7.142857142857143e-06,
|
| 1249 |
+
"loss": 0.3223,
|
| 1250 |
+
"step": 1220
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 43.93,
|
| 1254 |
+
"grad_norm": 3.389286518096924,
|
| 1255 |
+
"learning_rate": 6.746031746031746e-06,
|
| 1256 |
+
"loss": 0.3147,
|
| 1257 |
+
"step": 1230
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 44.0,
|
| 1261 |
+
"eval_accuracy": 0.5860552763819096,
|
| 1262 |
+
"eval_loss": 1.3746650218963623,
|
| 1263 |
+
"eval_runtime": 6.0707,
|
| 1264 |
+
"eval_samples_per_second": 262.244,
|
| 1265 |
+
"eval_steps_per_second": 2.141,
|
| 1266 |
+
"step": 1232
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 44.29,
|
| 1270 |
+
"grad_norm": 3.764704942703247,
|
| 1271 |
+
"learning_rate": 6.349206349206349e-06,
|
| 1272 |
+
"loss": 0.3107,
|
| 1273 |
+
"step": 1240
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 44.64,
|
| 1277 |
+
"grad_norm": 4.65610933303833,
|
| 1278 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 1279 |
+
"loss": 0.3058,
|
| 1280 |
+
"step": 1250
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 45.0,
|
| 1284 |
+
"grad_norm": 4.179033279418945,
|
| 1285 |
+
"learning_rate": 5.555555555555556e-06,
|
| 1286 |
+
"loss": 0.3125,
|
| 1287 |
+
"step": 1260
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 45.0,
|
| 1291 |
+
"eval_accuracy": 0.5816582914572864,
|
| 1292 |
+
"eval_loss": 1.3839383125305176,
|
| 1293 |
+
"eval_runtime": 5.8047,
|
| 1294 |
+
"eval_samples_per_second": 274.262,
|
| 1295 |
+
"eval_steps_per_second": 2.24,
|
| 1296 |
+
"step": 1260
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 45.36,
|
| 1300 |
+
"grad_norm": 7.927102088928223,
|
| 1301 |
+
"learning_rate": 5.158730158730159e-06,
|
| 1302 |
+
"loss": 0.3077,
|
| 1303 |
+
"step": 1270
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 45.71,
|
| 1307 |
+
"grad_norm": 3.8612356185913086,
|
| 1308 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 1309 |
+
"loss": 0.3276,
|
| 1310 |
+
"step": 1280
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 46.0,
|
| 1314 |
+
"eval_accuracy": 0.5841708542713567,
|
| 1315 |
+
"eval_loss": 1.3805787563323975,
|
| 1316 |
+
"eval_runtime": 5.5264,
|
| 1317 |
+
"eval_samples_per_second": 288.072,
|
| 1318 |
+
"eval_steps_per_second": 2.352,
|
| 1319 |
+
"step": 1288
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 46.07,
|
| 1323 |
+
"grad_norm": 4.099473476409912,
|
| 1324 |
+
"learning_rate": 4.365079365079365e-06,
|
| 1325 |
+
"loss": 0.3076,
|
| 1326 |
+
"step": 1290
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 46.43,
|
| 1330 |
+
"grad_norm": 3.8270063400268555,
|
| 1331 |
+
"learning_rate": 3.968253968253968e-06,
|
| 1332 |
+
"loss": 0.3097,
|
| 1333 |
+
"step": 1300
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 46.79,
|
| 1337 |
+
"grad_norm": 3.907658576965332,
|
| 1338 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 1339 |
+
"loss": 0.2989,
|
| 1340 |
+
"step": 1310
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 47.0,
|
| 1344 |
+
"eval_accuracy": 0.5885678391959799,
|
| 1345 |
+
"eval_loss": 1.3905527591705322,
|
| 1346 |
+
"eval_runtime": 5.6646,
|
| 1347 |
+
"eval_samples_per_second": 281.046,
|
| 1348 |
+
"eval_steps_per_second": 2.295,
|
| 1349 |
+
"step": 1316
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 47.14,
|
| 1353 |
+
"grad_norm": 4.716944694519043,
|
| 1354 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 1355 |
+
"loss": 0.3173,
|
| 1356 |
+
"step": 1320
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 47.5,
|
| 1360 |
+
"grad_norm": 3.6047604084014893,
|
| 1361 |
+
"learning_rate": 2.777777777777778e-06,
|
| 1362 |
+
"loss": 0.3172,
|
| 1363 |
+
"step": 1330
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 47.86,
|
| 1367 |
+
"grad_norm": 4.362003803253174,
|
| 1368 |
+
"learning_rate": 2.3809523809523808e-06,
|
| 1369 |
+
"loss": 0.2941,
|
| 1370 |
+
"step": 1340
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 48.0,
|
| 1374 |
+
"eval_accuracy": 0.5866834170854272,
|
| 1375 |
+
"eval_loss": 1.3876359462738037,
|
| 1376 |
+
"eval_runtime": 5.5384,
|
| 1377 |
+
"eval_samples_per_second": 287.447,
|
| 1378 |
+
"eval_steps_per_second": 2.347,
|
| 1379 |
+
"step": 1344
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 48.21,
|
| 1383 |
+
"grad_norm": 4.251241207122803,
|
| 1384 |
+
"learning_rate": 1.984126984126984e-06,
|
| 1385 |
+
"loss": 0.3138,
|
| 1386 |
+
"step": 1350
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 48.57,
|
| 1390 |
+
"grad_norm": 3.9117441177368164,
|
| 1391 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 1392 |
+
"loss": 0.3042,
|
| 1393 |
+
"step": 1360
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 48.93,
|
| 1397 |
+
"grad_norm": 3.703327178955078,
|
| 1398 |
+
"learning_rate": 1.1904761904761904e-06,
|
| 1399 |
+
"loss": 0.3131,
|
| 1400 |
+
"step": 1370
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 49.0,
|
| 1404 |
+
"eval_accuracy": 0.582286432160804,
|
| 1405 |
+
"eval_loss": 1.3895643949508667,
|
| 1406 |
+
"eval_runtime": 5.5003,
|
| 1407 |
+
"eval_samples_per_second": 289.44,
|
| 1408 |
+
"eval_steps_per_second": 2.364,
|
| 1409 |
+
"step": 1372
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 49.29,
|
| 1413 |
+
"grad_norm": 6.085123538970947,
|
| 1414 |
+
"learning_rate": 7.936507936507937e-07,
|
| 1415 |
+
"loss": 0.2984,
|
| 1416 |
+
"step": 1380
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 49.64,
|
| 1420 |
+
"grad_norm": 3.707207679748535,
|
| 1421 |
+
"learning_rate": 3.9682539682539683e-07,
|
| 1422 |
+
"loss": 0.3092,
|
| 1423 |
+
"step": 1390
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 50.0,
|
| 1427 |
+
"grad_norm": 4.588694095611572,
|
| 1428 |
+
"learning_rate": 0.0,
|
| 1429 |
+
"loss": 0.2975,
|
| 1430 |
+
"step": 1400
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 50.0,
|
| 1434 |
+
"eval_accuracy": 0.5835427135678392,
|
| 1435 |
+
"eval_loss": 1.3905625343322754,
|
| 1436 |
+
"eval_runtime": 5.7976,
|
| 1437 |
+
"eval_samples_per_second": 274.595,
|
| 1438 |
+
"eval_steps_per_second": 2.242,
|
| 1439 |
+
"step": 1400
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 50.0,
|
| 1443 |
+
"step": 1400,
|
| 1444 |
+
"total_flos": 1.7807825640923136e+19,
|
| 1445 |
+
"train_loss": 0.5565686808313642,
|
| 1446 |
+
"train_runtime": 3248.0342,
|
| 1447 |
+
"train_samples_per_second": 220.564,
|
| 1448 |
+
"train_steps_per_second": 0.431
|
| 1449 |
}
|
| 1450 |
],
|
| 1451 |
"logging_steps": 10,
|
| 1452 |
+
"max_steps": 1400,
|
| 1453 |
"num_input_tokens_seen": 0,
|
| 1454 |
+
"num_train_epochs": 50,
|
| 1455 |
"save_steps": 500,
|
| 1456 |
+
"total_flos": 1.7807825640923136e+19,
|
| 1457 |
"train_batch_size": 128,
|
| 1458 |
"trial_name": null,
|
| 1459 |
"trial_params": null
|