Commit
•
a76cc48
1
Parent(s):
979c104
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May21_23-00-36_804716d6a90a/events.out.tfevents.1716334079.804716d6a90a.958.15 +3 -0
- train_results.json +8 -0
- trainer_state.json +1098 -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 [facebook/deit-base-distilled-patch16-224](https://huggingface.co/facebook/deit-base-distilled-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
- Accuracy: 0.8732
|
| 37 |
|
| 38 |
## Model description
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [facebook/deit-base-distilled-patch16-224](https://huggingface.co/facebook/deit-base-distilled-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.4993
|
| 36 |
- Accuracy: 0.8732
|
| 37 |
|
| 38 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 92.3076923076923,
|
| 3 |
+
"eval_accuracy": 0.8732394366197183,
|
| 4 |
+
"eval_loss": 0.49927499890327454,
|
| 5 |
+
"eval_runtime": 1.0275,
|
| 6 |
+
"eval_samples_per_second": 69.097,
|
| 7 |
+
"eval_steps_per_second": 2.92,
|
| 8 |
+
"total_flos": 2.8402872494292173e+18,
|
| 9 |
+
"train_loss": 0.2501216806968053,
|
| 10 |
+
"train_runtime": 1631.2353,
|
| 11 |
+
"train_samples_per_second": 24.337,
|
| 12 |
+
"train_steps_per_second": 0.184
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 92.3076923076923,
|
| 3 |
+
"eval_accuracy": 0.8732394366197183,
|
| 4 |
+
"eval_loss": 0.49927499890327454,
|
| 5 |
+
"eval_runtime": 1.0275,
|
| 6 |
+
"eval_samples_per_second": 69.097,
|
| 7 |
+
"eval_steps_per_second": 2.92
|
| 8 |
+
}
|
runs/May21_23-00-36_804716d6a90a/events.out.tfevents.1716334079.804716d6a90a.958.15
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14ea45010ff458c7d346a608ccfdadc60e503842aedeb3ee4ec88577c6877d13
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 92.3076923076923,
|
| 3 |
+
"total_flos": 2.8402872494292173e+18,
|
| 4 |
+
"train_loss": 0.2501216806968053,
|
| 5 |
+
"train_runtime": 1631.2353,
|
| 6 |
+
"train_samples_per_second": 24.337,
|
| 7 |
+
"train_steps_per_second": 0.184
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1098 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8732394366197183,
|
| 3 |
+
"best_model_checkpoint": "deit-base-distilled-patch16-224-65-fold3/checkpoint-276",
|
| 4 |
+
"epoch": 92.3076923076923,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9230769230769231,
|
| 13 |
+
"eval_accuracy": 0.5492957746478874,
|
| 14 |
+
"eval_loss": 0.7402985692024231,
|
| 15 |
+
"eval_runtime": 0.9976,
|
| 16 |
+
"eval_samples_per_second": 71.17,
|
| 17 |
+
"eval_steps_per_second": 3.007,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.8461538461538463,
|
| 22 |
+
"eval_accuracy": 0.5211267605633803,
|
| 23 |
+
"eval_loss": 0.7199354767799377,
|
| 24 |
+
"eval_runtime": 0.9611,
|
| 25 |
+
"eval_samples_per_second": 73.873,
|
| 26 |
+
"eval_steps_per_second": 3.121,
|
| 27 |
+
"step": 6
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.769230769230769,
|
| 31 |
+
"eval_accuracy": 0.5633802816901409,
|
| 32 |
+
"eval_loss": 0.7111433148384094,
|
| 33 |
+
"eval_runtime": 0.9671,
|
| 34 |
+
"eval_samples_per_second": 73.414,
|
| 35 |
+
"eval_steps_per_second": 3.102,
|
| 36 |
+
"step": 9
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.076923076923077,
|
| 40 |
+
"grad_norm": 7.598382949829102,
|
| 41 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 42 |
+
"loss": 0.7693,
|
| 43 |
+
"step": 10
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 4.0,
|
| 47 |
+
"eval_accuracy": 0.5352112676056338,
|
| 48 |
+
"eval_loss": 0.7014583945274353,
|
| 49 |
+
"eval_runtime": 0.9614,
|
| 50 |
+
"eval_samples_per_second": 73.854,
|
| 51 |
+
"eval_steps_per_second": 3.121,
|
| 52 |
+
"step": 13
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.923076923076923,
|
| 56 |
+
"eval_accuracy": 0.6197183098591549,
|
| 57 |
+
"eval_loss": 0.6470862627029419,
|
| 58 |
+
"eval_runtime": 0.965,
|
| 59 |
+
"eval_samples_per_second": 73.577,
|
| 60 |
+
"eval_steps_per_second": 3.109,
|
| 61 |
+
"step": 16
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 5.846153846153846,
|
| 65 |
+
"eval_accuracy": 0.6056338028169014,
|
| 66 |
+
"eval_loss": 0.6691372394561768,
|
| 67 |
+
"eval_runtime": 0.966,
|
| 68 |
+
"eval_samples_per_second": 73.497,
|
| 69 |
+
"eval_steps_per_second": 3.105,
|
| 70 |
+
"step": 19
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 6.153846153846154,
|
| 74 |
+
"grad_norm": 1.975386381149292,
|
| 75 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 76 |
+
"loss": 0.6542,
|
| 77 |
+
"step": 20
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.769230769230769,
|
| 81 |
+
"eval_accuracy": 0.6197183098591549,
|
| 82 |
+
"eval_loss": 0.6188070774078369,
|
| 83 |
+
"eval_runtime": 0.9657,
|
| 84 |
+
"eval_samples_per_second": 73.52,
|
| 85 |
+
"eval_steps_per_second": 3.106,
|
| 86 |
+
"step": 22
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.0,
|
| 90 |
+
"eval_accuracy": 0.5774647887323944,
|
| 91 |
+
"eval_loss": 0.69666588306427,
|
| 92 |
+
"eval_runtime": 0.9856,
|
| 93 |
+
"eval_samples_per_second": 72.037,
|
| 94 |
+
"eval_steps_per_second": 3.044,
|
| 95 |
+
"step": 26
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.923076923076923,
|
| 99 |
+
"eval_accuracy": 0.7323943661971831,
|
| 100 |
+
"eval_loss": 0.5731548070907593,
|
| 101 |
+
"eval_runtime": 0.9661,
|
| 102 |
+
"eval_samples_per_second": 73.491,
|
| 103 |
+
"eval_steps_per_second": 3.105,
|
| 104 |
+
"step": 29
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 9.23076923076923,
|
| 108 |
+
"grad_norm": 5.027002811431885,
|
| 109 |
+
"learning_rate": 5e-05,
|
| 110 |
+
"loss": 0.5935,
|
| 111 |
+
"step": 30
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 9.846153846153847,
|
| 115 |
+
"eval_accuracy": 0.704225352112676,
|
| 116 |
+
"eval_loss": 0.5184200406074524,
|
| 117 |
+
"eval_runtime": 0.9721,
|
| 118 |
+
"eval_samples_per_second": 73.035,
|
| 119 |
+
"eval_steps_per_second": 3.086,
|
| 120 |
+
"step": 32
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 10.76923076923077,
|
| 124 |
+
"eval_accuracy": 0.7183098591549296,
|
| 125 |
+
"eval_loss": 0.6031274795532227,
|
| 126 |
+
"eval_runtime": 0.976,
|
| 127 |
+
"eval_samples_per_second": 72.75,
|
| 128 |
+
"eval_steps_per_second": 3.074,
|
| 129 |
+
"step": 35
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 12.0,
|
| 133 |
+
"eval_accuracy": 0.647887323943662,
|
| 134 |
+
"eval_loss": 0.6670598983764648,
|
| 135 |
+
"eval_runtime": 0.9859,
|
| 136 |
+
"eval_samples_per_second": 72.016,
|
| 137 |
+
"eval_steps_per_second": 3.043,
|
| 138 |
+
"step": 39
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 12.307692307692308,
|
| 142 |
+
"grad_norm": 7.941720008850098,
|
| 143 |
+
"learning_rate": 4.814814814814815e-05,
|
| 144 |
+
"loss": 0.549,
|
| 145 |
+
"step": 40
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 12.923076923076923,
|
| 149 |
+
"eval_accuracy": 0.7183098591549296,
|
| 150 |
+
"eval_loss": 0.5281336903572083,
|
| 151 |
+
"eval_runtime": 1.0091,
|
| 152 |
+
"eval_samples_per_second": 70.358,
|
| 153 |
+
"eval_steps_per_second": 2.973,
|
| 154 |
+
"step": 42
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 13.846153846153847,
|
| 158 |
+
"eval_accuracy": 0.7183098591549296,
|
| 159 |
+
"eval_loss": 0.5792147517204285,
|
| 160 |
+
"eval_runtime": 1.0732,
|
| 161 |
+
"eval_samples_per_second": 66.158,
|
| 162 |
+
"eval_steps_per_second": 2.795,
|
| 163 |
+
"step": 45
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 14.76923076923077,
|
| 167 |
+
"eval_accuracy": 0.7464788732394366,
|
| 168 |
+
"eval_loss": 0.5389305353164673,
|
| 169 |
+
"eval_runtime": 1.015,
|
| 170 |
+
"eval_samples_per_second": 69.952,
|
| 171 |
+
"eval_steps_per_second": 2.956,
|
| 172 |
+
"step": 48
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 15.384615384615385,
|
| 176 |
+
"grad_norm": 3.035529613494873,
|
| 177 |
+
"learning_rate": 4.62962962962963e-05,
|
| 178 |
+
"loss": 0.4778,
|
| 179 |
+
"step": 50
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 16.0,
|
| 183 |
+
"eval_accuracy": 0.704225352112676,
|
| 184 |
+
"eval_loss": 0.600963830947876,
|
| 185 |
+
"eval_runtime": 0.9862,
|
| 186 |
+
"eval_samples_per_second": 71.992,
|
| 187 |
+
"eval_steps_per_second": 3.042,
|
| 188 |
+
"step": 52
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 16.923076923076923,
|
| 192 |
+
"eval_accuracy": 0.7605633802816901,
|
| 193 |
+
"eval_loss": 0.5244532227516174,
|
| 194 |
+
"eval_runtime": 0.9743,
|
| 195 |
+
"eval_samples_per_second": 72.876,
|
| 196 |
+
"eval_steps_per_second": 3.079,
|
| 197 |
+
"step": 55
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 17.846153846153847,
|
| 201 |
+
"eval_accuracy": 0.7183098591549296,
|
| 202 |
+
"eval_loss": 0.5491234064102173,
|
| 203 |
+
"eval_runtime": 0.997,
|
| 204 |
+
"eval_samples_per_second": 71.212,
|
| 205 |
+
"eval_steps_per_second": 3.009,
|
| 206 |
+
"step": 58
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 18.46153846153846,
|
| 210 |
+
"grad_norm": 2.951366662979126,
|
| 211 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 212 |
+
"loss": 0.4039,
|
| 213 |
+
"step": 60
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 18.76923076923077,
|
| 217 |
+
"eval_accuracy": 0.7464788732394366,
|
| 218 |
+
"eval_loss": 0.55897057056427,
|
| 219 |
+
"eval_runtime": 0.9926,
|
| 220 |
+
"eval_samples_per_second": 71.527,
|
| 221 |
+
"eval_steps_per_second": 3.022,
|
| 222 |
+
"step": 61
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 20.0,
|
| 226 |
+
"eval_accuracy": 0.7323943661971831,
|
| 227 |
+
"eval_loss": 0.4886135458946228,
|
| 228 |
+
"eval_runtime": 0.9968,
|
| 229 |
+
"eval_samples_per_second": 71.225,
|
| 230 |
+
"eval_steps_per_second": 3.01,
|
| 231 |
+
"step": 65
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 20.923076923076923,
|
| 235 |
+
"eval_accuracy": 0.7323943661971831,
|
| 236 |
+
"eval_loss": 0.5049741268157959,
|
| 237 |
+
"eval_runtime": 0.9885,
|
| 238 |
+
"eval_samples_per_second": 71.822,
|
| 239 |
+
"eval_steps_per_second": 3.035,
|
| 240 |
+
"step": 68
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 21.53846153846154,
|
| 244 |
+
"grad_norm": 2.950965404510498,
|
| 245 |
+
"learning_rate": 4.259259259259259e-05,
|
| 246 |
+
"loss": 0.3409,
|
| 247 |
+
"step": 70
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 21.846153846153847,
|
| 251 |
+
"eval_accuracy": 0.7464788732394366,
|
| 252 |
+
"eval_loss": 0.49122846126556396,
|
| 253 |
+
"eval_runtime": 0.9814,
|
| 254 |
+
"eval_samples_per_second": 72.349,
|
| 255 |
+
"eval_steps_per_second": 3.057,
|
| 256 |
+
"step": 71
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 22.76923076923077,
|
| 260 |
+
"eval_accuracy": 0.7746478873239436,
|
| 261 |
+
"eval_loss": 0.492927223443985,
|
| 262 |
+
"eval_runtime": 0.9891,
|
| 263 |
+
"eval_samples_per_second": 71.783,
|
| 264 |
+
"eval_steps_per_second": 3.033,
|
| 265 |
+
"step": 74
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 24.0,
|
| 269 |
+
"eval_accuracy": 0.7746478873239436,
|
| 270 |
+
"eval_loss": 0.5365388989448547,
|
| 271 |
+
"eval_runtime": 0.9902,
|
| 272 |
+
"eval_samples_per_second": 71.704,
|
| 273 |
+
"eval_steps_per_second": 3.03,
|
| 274 |
+
"step": 78
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 24.615384615384617,
|
| 278 |
+
"grad_norm": 8.019288063049316,
|
| 279 |
+
"learning_rate": 4.074074074074074e-05,
|
| 280 |
+
"loss": 0.3202,
|
| 281 |
+
"step": 80
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 24.923076923076923,
|
| 285 |
+
"eval_accuracy": 0.8028169014084507,
|
| 286 |
+
"eval_loss": 0.4685072600841522,
|
| 287 |
+
"eval_runtime": 0.983,
|
| 288 |
+
"eval_samples_per_second": 72.224,
|
| 289 |
+
"eval_steps_per_second": 3.052,
|
| 290 |
+
"step": 81
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 25.846153846153847,
|
| 294 |
+
"eval_accuracy": 0.8169014084507042,
|
| 295 |
+
"eval_loss": 0.44036272168159485,
|
| 296 |
+
"eval_runtime": 0.9931,
|
| 297 |
+
"eval_samples_per_second": 71.493,
|
| 298 |
+
"eval_steps_per_second": 3.021,
|
| 299 |
+
"step": 84
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 26.76923076923077,
|
| 303 |
+
"eval_accuracy": 0.8028169014084507,
|
| 304 |
+
"eval_loss": 0.463866263628006,
|
| 305 |
+
"eval_runtime": 0.9949,
|
| 306 |
+
"eval_samples_per_second": 71.364,
|
| 307 |
+
"eval_steps_per_second": 3.015,
|
| 308 |
+
"step": 87
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 27.692307692307693,
|
| 312 |
+
"grad_norm": 2.1704952716827393,
|
| 313 |
+
"learning_rate": 3.888888888888889e-05,
|
| 314 |
+
"loss": 0.2466,
|
| 315 |
+
"step": 90
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 28.0,
|
| 319 |
+
"eval_accuracy": 0.7605633802816901,
|
| 320 |
+
"eval_loss": 0.5491413474082947,
|
| 321 |
+
"eval_runtime": 0.9936,
|
| 322 |
+
"eval_samples_per_second": 71.461,
|
| 323 |
+
"eval_steps_per_second": 3.019,
|
| 324 |
+
"step": 91
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 28.923076923076923,
|
| 328 |
+
"eval_accuracy": 0.7605633802816901,
|
| 329 |
+
"eval_loss": 0.5169638991355896,
|
| 330 |
+
"eval_runtime": 1.0004,
|
| 331 |
+
"eval_samples_per_second": 70.97,
|
| 332 |
+
"eval_steps_per_second": 2.999,
|
| 333 |
+
"step": 94
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 29.846153846153847,
|
| 337 |
+
"eval_accuracy": 0.8028169014084507,
|
| 338 |
+
"eval_loss": 0.4444153606891632,
|
| 339 |
+
"eval_runtime": 0.9937,
|
| 340 |
+
"eval_samples_per_second": 71.452,
|
| 341 |
+
"eval_steps_per_second": 3.019,
|
| 342 |
+
"step": 97
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 30.76923076923077,
|
| 346 |
+
"grad_norm": 3.7423977851867676,
|
| 347 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 348 |
+
"loss": 0.2433,
|
| 349 |
+
"step": 100
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 30.76923076923077,
|
| 353 |
+
"eval_accuracy": 0.8309859154929577,
|
| 354 |
+
"eval_loss": 0.4516857862472534,
|
| 355 |
+
"eval_runtime": 0.9934,
|
| 356 |
+
"eval_samples_per_second": 71.473,
|
| 357 |
+
"eval_steps_per_second": 3.02,
|
| 358 |
+
"step": 100
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 32.0,
|
| 362 |
+
"eval_accuracy": 0.7605633802816901,
|
| 363 |
+
"eval_loss": 0.7796855568885803,
|
| 364 |
+
"eval_runtime": 0.9972,
|
| 365 |
+
"eval_samples_per_second": 71.202,
|
| 366 |
+
"eval_steps_per_second": 3.009,
|
| 367 |
+
"step": 104
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 32.92307692307692,
|
| 371 |
+
"eval_accuracy": 0.8169014084507042,
|
| 372 |
+
"eval_loss": 0.43210193514823914,
|
| 373 |
+
"eval_runtime": 0.9973,
|
| 374 |
+
"eval_samples_per_second": 71.192,
|
| 375 |
+
"eval_steps_per_second": 3.008,
|
| 376 |
+
"step": 107
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 33.84615384615385,
|
| 380 |
+
"grad_norm": 4.042857646942139,
|
| 381 |
+
"learning_rate": 3.518518518518519e-05,
|
| 382 |
+
"loss": 0.2535,
|
| 383 |
+
"step": 110
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 33.84615384615385,
|
| 387 |
+
"eval_accuracy": 0.7746478873239436,
|
| 388 |
+
"eval_loss": 0.5956353545188904,
|
| 389 |
+
"eval_runtime": 0.9936,
|
| 390 |
+
"eval_samples_per_second": 71.457,
|
| 391 |
+
"eval_steps_per_second": 3.019,
|
| 392 |
+
"step": 110
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 34.76923076923077,
|
| 396 |
+
"eval_accuracy": 0.7887323943661971,
|
| 397 |
+
"eval_loss": 0.46952158212661743,
|
| 398 |
+
"eval_runtime": 1.0048,
|
| 399 |
+
"eval_samples_per_second": 70.661,
|
| 400 |
+
"eval_steps_per_second": 2.986,
|
| 401 |
+
"step": 113
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 36.0,
|
| 405 |
+
"eval_accuracy": 0.6901408450704225,
|
| 406 |
+
"eval_loss": 0.8105931878089905,
|
| 407 |
+
"eval_runtime": 0.9983,
|
| 408 |
+
"eval_samples_per_second": 71.118,
|
| 409 |
+
"eval_steps_per_second": 3.005,
|
| 410 |
+
"step": 117
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 36.92307692307692,
|
| 414 |
+
"grad_norm": 3.739529609680176,
|
| 415 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 416 |
+
"loss": 0.2215,
|
| 417 |
+
"step": 120
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 36.92307692307692,
|
| 421 |
+
"eval_accuracy": 0.7464788732394366,
|
| 422 |
+
"eval_loss": 0.7118771076202393,
|
| 423 |
+
"eval_runtime": 0.9979,
|
| 424 |
+
"eval_samples_per_second": 71.147,
|
| 425 |
+
"eval_steps_per_second": 3.006,
|
| 426 |
+
"step": 120
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 37.84615384615385,
|
| 430 |
+
"eval_accuracy": 0.8028169014084507,
|
| 431 |
+
"eval_loss": 0.4752153158187866,
|
| 432 |
+
"eval_runtime": 0.9912,
|
| 433 |
+
"eval_samples_per_second": 71.631,
|
| 434 |
+
"eval_steps_per_second": 3.027,
|
| 435 |
+
"step": 123
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 38.76923076923077,
|
| 439 |
+
"eval_accuracy": 0.8169014084507042,
|
| 440 |
+
"eval_loss": 0.4784373342990875,
|
| 441 |
+
"eval_runtime": 0.9957,
|
| 442 |
+
"eval_samples_per_second": 71.31,
|
| 443 |
+
"eval_steps_per_second": 3.013,
|
| 444 |
+
"step": 126
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 40.0,
|
| 448 |
+
"grad_norm": 3.5269320011138916,
|
| 449 |
+
"learning_rate": 3.148148148148148e-05,
|
| 450 |
+
"loss": 0.2143,
|
| 451 |
+
"step": 130
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 40.0,
|
| 455 |
+
"eval_accuracy": 0.8028169014084507,
|
| 456 |
+
"eval_loss": 0.4772735834121704,
|
| 457 |
+
"eval_runtime": 1.0003,
|
| 458 |
+
"eval_samples_per_second": 70.981,
|
| 459 |
+
"eval_steps_per_second": 2.999,
|
| 460 |
+
"step": 130
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 40.92307692307692,
|
| 464 |
+
"eval_accuracy": 0.8169014084507042,
|
| 465 |
+
"eval_loss": 0.5580780506134033,
|
| 466 |
+
"eval_runtime": 0.9949,
|
| 467 |
+
"eval_samples_per_second": 71.366,
|
| 468 |
+
"eval_steps_per_second": 3.015,
|
| 469 |
+
"step": 133
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 41.84615384615385,
|
| 473 |
+
"eval_accuracy": 0.8028169014084507,
|
| 474 |
+
"eval_loss": 0.6098394989967346,
|
| 475 |
+
"eval_runtime": 0.9964,
|
| 476 |
+
"eval_samples_per_second": 71.254,
|
| 477 |
+
"eval_steps_per_second": 3.011,
|
| 478 |
+
"step": 136
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 42.76923076923077,
|
| 482 |
+
"eval_accuracy": 0.8169014084507042,
|
| 483 |
+
"eval_loss": 0.5193454623222351,
|
| 484 |
+
"eval_runtime": 0.9881,
|
| 485 |
+
"eval_samples_per_second": 71.858,
|
| 486 |
+
"eval_steps_per_second": 3.036,
|
| 487 |
+
"step": 139
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 43.07692307692308,
|
| 491 |
+
"grad_norm": 2.791985273361206,
|
| 492 |
+
"learning_rate": 2.962962962962963e-05,
|
| 493 |
+
"loss": 0.1726,
|
| 494 |
+
"step": 140
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 44.0,
|
| 498 |
+
"eval_accuracy": 0.8450704225352113,
|
| 499 |
+
"eval_loss": 0.4306134879589081,
|
| 500 |
+
"eval_runtime": 0.9976,
|
| 501 |
+
"eval_samples_per_second": 71.173,
|
| 502 |
+
"eval_steps_per_second": 3.007,
|
| 503 |
+
"step": 143
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 44.92307692307692,
|
| 507 |
+
"eval_accuracy": 0.8591549295774648,
|
| 508 |
+
"eval_loss": 0.4233701527118683,
|
| 509 |
+
"eval_runtime": 0.9983,
|
| 510 |
+
"eval_samples_per_second": 71.12,
|
| 511 |
+
"eval_steps_per_second": 3.005,
|
| 512 |
+
"step": 146
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 45.84615384615385,
|
| 516 |
+
"eval_accuracy": 0.8169014084507042,
|
| 517 |
+
"eval_loss": 0.5263967514038086,
|
| 518 |
+
"eval_runtime": 0.9942,
|
| 519 |
+
"eval_samples_per_second": 71.411,
|
| 520 |
+
"eval_steps_per_second": 3.017,
|
| 521 |
+
"step": 149
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 46.15384615384615,
|
| 525 |
+
"grad_norm": 3.0404417514801025,
|
| 526 |
+
"learning_rate": 2.777777777777778e-05,
|
| 527 |
+
"loss": 0.1684,
|
| 528 |
+
"step": 150
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 46.76923076923077,
|
| 532 |
+
"eval_accuracy": 0.8028169014084507,
|
| 533 |
+
"eval_loss": 0.7302807569503784,
|
| 534 |
+
"eval_runtime": 0.9944,
|
| 535 |
+
"eval_samples_per_second": 71.403,
|
| 536 |
+
"eval_steps_per_second": 3.017,
|
| 537 |
+
"step": 152
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 48.0,
|
| 541 |
+
"eval_accuracy": 0.8169014084507042,
|
| 542 |
+
"eval_loss": 0.5079183578491211,
|
| 543 |
+
"eval_runtime": 0.993,
|
| 544 |
+
"eval_samples_per_second": 71.499,
|
| 545 |
+
"eval_steps_per_second": 3.021,
|
| 546 |
+
"step": 156
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 48.92307692307692,
|
| 550 |
+
"eval_accuracy": 0.8169014084507042,
|
| 551 |
+
"eval_loss": 0.5391997694969177,
|
| 552 |
+
"eval_runtime": 0.9941,
|
| 553 |
+
"eval_samples_per_second": 71.421,
|
| 554 |
+
"eval_steps_per_second": 3.018,
|
| 555 |
+
"step": 159
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 49.23076923076923,
|
| 559 |
+
"grad_norm": 5.0057854652404785,
|
| 560 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 561 |
+
"loss": 0.1604,
|
| 562 |
+
"step": 160
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 49.84615384615385,
|
| 566 |
+
"eval_accuracy": 0.8169014084507042,
|
| 567 |
+
"eval_loss": 0.3951334059238434,
|
| 568 |
+
"eval_runtime": 0.9969,
|
| 569 |
+
"eval_samples_per_second": 71.222,
|
| 570 |
+
"eval_steps_per_second": 3.009,
|
| 571 |
+
"step": 162
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 50.76923076923077,
|
| 575 |
+
"eval_accuracy": 0.8028169014084507,
|
| 576 |
+
"eval_loss": 0.43108686804771423,
|
| 577 |
+
"eval_runtime": 1.0007,
|
| 578 |
+
"eval_samples_per_second": 70.95,
|
| 579 |
+
"eval_steps_per_second": 2.998,
|
| 580 |
+
"step": 165
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 52.0,
|
| 584 |
+
"eval_accuracy": 0.8028169014084507,
|
| 585 |
+
"eval_loss": 0.40819185972213745,
|
| 586 |
+
"eval_runtime": 1.0027,
|
| 587 |
+
"eval_samples_per_second": 70.81,
|
| 588 |
+
"eval_steps_per_second": 2.992,
|
| 589 |
+
"step": 169
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 52.30769230769231,
|
| 593 |
+
"grad_norm": 2.250169515609741,
|
| 594 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 595 |
+
"loss": 0.1457,
|
| 596 |
+
"step": 170
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 52.92307692307692,
|
| 600 |
+
"eval_accuracy": 0.7887323943661971,
|
| 601 |
+
"eval_loss": 0.4173259139060974,
|
| 602 |
+
"eval_runtime": 1.0625,
|
| 603 |
+
"eval_samples_per_second": 66.825,
|
| 604 |
+
"eval_steps_per_second": 2.824,
|
| 605 |
+
"step": 172
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 53.84615384615385,
|
| 609 |
+
"eval_accuracy": 0.8309859154929577,
|
| 610 |
+
"eval_loss": 0.4310729205608368,
|
| 611 |
+
"eval_runtime": 1.0019,
|
| 612 |
+
"eval_samples_per_second": 70.864,
|
| 613 |
+
"eval_steps_per_second": 2.994,
|
| 614 |
+
"step": 175
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 54.76923076923077,
|
| 618 |
+
"eval_accuracy": 0.8028169014084507,
|
| 619 |
+
"eval_loss": 0.42128264904022217,
|
| 620 |
+
"eval_runtime": 1.0021,
|
| 621 |
+
"eval_samples_per_second": 70.851,
|
| 622 |
+
"eval_steps_per_second": 2.994,
|
| 623 |
+
"step": 178
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 55.38461538461539,
|
| 627 |
+
"grad_norm": 3.804135799407959,
|
| 628 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 629 |
+
"loss": 0.1549,
|
| 630 |
+
"step": 180
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 56.0,
|
| 634 |
+
"eval_accuracy": 0.8450704225352113,
|
| 635 |
+
"eval_loss": 0.4712894558906555,
|
| 636 |
+
"eval_runtime": 0.9989,
|
| 637 |
+
"eval_samples_per_second": 71.076,
|
| 638 |
+
"eval_steps_per_second": 3.003,
|
| 639 |
+
"step": 182
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 56.92307692307692,
|
| 643 |
+
"eval_accuracy": 0.8028169014084507,
|
| 644 |
+
"eval_loss": 0.749286949634552,
|
| 645 |
+
"eval_runtime": 0.9921,
|
| 646 |
+
"eval_samples_per_second": 71.569,
|
| 647 |
+
"eval_steps_per_second": 3.024,
|
| 648 |
+
"step": 185
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 57.84615384615385,
|
| 652 |
+
"eval_accuracy": 0.8450704225352113,
|
| 653 |
+
"eval_loss": 0.5160987973213196,
|
| 654 |
+
"eval_runtime": 0.9878,
|
| 655 |
+
"eval_samples_per_second": 71.876,
|
| 656 |
+
"eval_steps_per_second": 3.037,
|
| 657 |
+
"step": 188
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 58.46153846153846,
|
| 661 |
+
"grad_norm": 2.6206445693969727,
|
| 662 |
+
"learning_rate": 2.037037037037037e-05,
|
| 663 |
+
"loss": 0.1391,
|
| 664 |
+
"step": 190
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 58.76923076923077,
|
| 668 |
+
"eval_accuracy": 0.8169014084507042,
|
| 669 |
+
"eval_loss": 0.46853163838386536,
|
| 670 |
+
"eval_runtime": 0.9872,
|
| 671 |
+
"eval_samples_per_second": 71.917,
|
| 672 |
+
"eval_steps_per_second": 3.039,
|
| 673 |
+
"step": 191
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 60.0,
|
| 677 |
+
"eval_accuracy": 0.8028169014084507,
|
| 678 |
+
"eval_loss": 0.6968369483947754,
|
| 679 |
+
"eval_runtime": 0.9894,
|
| 680 |
+
"eval_samples_per_second": 71.763,
|
| 681 |
+
"eval_steps_per_second": 3.032,
|
| 682 |
+
"step": 195
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 60.92307692307692,
|
| 686 |
+
"eval_accuracy": 0.8309859154929577,
|
| 687 |
+
"eval_loss": 0.5836513638496399,
|
| 688 |
+
"eval_runtime": 0.9857,
|
| 689 |
+
"eval_samples_per_second": 72.029,
|
| 690 |
+
"eval_steps_per_second": 3.043,
|
| 691 |
+
"step": 198
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 61.53846153846154,
|
| 695 |
+
"grad_norm": 1.732333779335022,
|
| 696 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 697 |
+
"loss": 0.1272,
|
| 698 |
+
"step": 200
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 61.84615384615385,
|
| 702 |
+
"eval_accuracy": 0.8169014084507042,
|
| 703 |
+
"eval_loss": 0.5863271951675415,
|
| 704 |
+
"eval_runtime": 0.9938,
|
| 705 |
+
"eval_samples_per_second": 71.442,
|
| 706 |
+
"eval_steps_per_second": 3.019,
|
| 707 |
+
"step": 201
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 62.76923076923077,
|
| 711 |
+
"eval_accuracy": 0.8309859154929577,
|
| 712 |
+
"eval_loss": 0.5459548830986023,
|
| 713 |
+
"eval_runtime": 0.9852,
|
| 714 |
+
"eval_samples_per_second": 72.065,
|
| 715 |
+
"eval_steps_per_second": 3.045,
|
| 716 |
+
"step": 204
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 64.0,
|
| 720 |
+
"eval_accuracy": 0.8309859154929577,
|
| 721 |
+
"eval_loss": 0.6198180317878723,
|
| 722 |
+
"eval_runtime": 0.9942,
|
| 723 |
+
"eval_samples_per_second": 71.415,
|
| 724 |
+
"eval_steps_per_second": 3.018,
|
| 725 |
+
"step": 208
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 64.61538461538461,
|
| 729 |
+
"grad_norm": 3.675976514816284,
|
| 730 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 731 |
+
"loss": 0.1341,
|
| 732 |
+
"step": 210
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 64.92307692307692,
|
| 736 |
+
"eval_accuracy": 0.8591549295774648,
|
| 737 |
+
"eval_loss": 0.558370053768158,
|
| 738 |
+
"eval_runtime": 1.01,
|
| 739 |
+
"eval_samples_per_second": 70.294,
|
| 740 |
+
"eval_steps_per_second": 2.97,
|
| 741 |
+
"step": 211
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 65.84615384615384,
|
| 745 |
+
"eval_accuracy": 0.8450704225352113,
|
| 746 |
+
"eval_loss": 0.6429418325424194,
|
| 747 |
+
"eval_runtime": 0.992,
|
| 748 |
+
"eval_samples_per_second": 71.571,
|
| 749 |
+
"eval_steps_per_second": 3.024,
|
| 750 |
+
"step": 214
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 66.76923076923077,
|
| 754 |
+
"eval_accuracy": 0.8028169014084507,
|
| 755 |
+
"eval_loss": 0.8592150211334229,
|
| 756 |
+
"eval_runtime": 1.0035,
|
| 757 |
+
"eval_samples_per_second": 70.751,
|
| 758 |
+
"eval_steps_per_second": 2.989,
|
| 759 |
+
"step": 217
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 67.6923076923077,
|
| 763 |
+
"grad_norm": 1.6102505922317505,
|
| 764 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 765 |
+
"loss": 0.1144,
|
| 766 |
+
"step": 220
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 68.0,
|
| 770 |
+
"eval_accuracy": 0.8028169014084507,
|
| 771 |
+
"eval_loss": 0.8471563458442688,
|
| 772 |
+
"eval_runtime": 0.989,
|
| 773 |
+
"eval_samples_per_second": 71.787,
|
| 774 |
+
"eval_steps_per_second": 3.033,
|
| 775 |
+
"step": 221
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 68.92307692307692,
|
| 779 |
+
"eval_accuracy": 0.8169014084507042,
|
| 780 |
+
"eval_loss": 0.83599853515625,
|
| 781 |
+
"eval_runtime": 0.9959,
|
| 782 |
+
"eval_samples_per_second": 71.291,
|
| 783 |
+
"eval_steps_per_second": 3.012,
|
| 784 |
+
"step": 224
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 69.84615384615384,
|
| 788 |
+
"eval_accuracy": 0.8169014084507042,
|
| 789 |
+
"eval_loss": 0.6696776151657104,
|
| 790 |
+
"eval_runtime": 0.9955,
|
| 791 |
+
"eval_samples_per_second": 71.323,
|
| 792 |
+
"eval_steps_per_second": 3.014,
|
| 793 |
+
"step": 227
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 70.76923076923077,
|
| 797 |
+
"grad_norm": 2.509876012802124,
|
| 798 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 799 |
+
"loss": 0.1321,
|
| 800 |
+
"step": 230
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 70.76923076923077,
|
| 804 |
+
"eval_accuracy": 0.8028169014084507,
|
| 805 |
+
"eval_loss": 0.6625356674194336,
|
| 806 |
+
"eval_runtime": 0.9946,
|
| 807 |
+
"eval_samples_per_second": 71.389,
|
| 808 |
+
"eval_steps_per_second": 3.016,
|
| 809 |
+
"step": 230
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 72.0,
|
| 813 |
+
"eval_accuracy": 0.8309859154929577,
|
| 814 |
+
"eval_loss": 0.7227580547332764,
|
| 815 |
+
"eval_runtime": 0.9869,
|
| 816 |
+
"eval_samples_per_second": 71.944,
|
| 817 |
+
"eval_steps_per_second": 3.04,
|
| 818 |
+
"step": 234
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 72.92307692307692,
|
| 822 |
+
"eval_accuracy": 0.8309859154929577,
|
| 823 |
+
"eval_loss": 0.6792935729026794,
|
| 824 |
+
"eval_runtime": 0.9957,
|
| 825 |
+
"eval_samples_per_second": 71.304,
|
| 826 |
+
"eval_steps_per_second": 3.013,
|
| 827 |
+
"step": 237
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 73.84615384615384,
|
| 831 |
+
"grad_norm": 3.0507752895355225,
|
| 832 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 833 |
+
"loss": 0.1206,
|
| 834 |
+
"step": 240
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 73.84615384615384,
|
| 838 |
+
"eval_accuracy": 0.8591549295774648,
|
| 839 |
+
"eval_loss": 0.5570999979972839,
|
| 840 |
+
"eval_runtime": 0.9893,
|
| 841 |
+
"eval_samples_per_second": 71.771,
|
| 842 |
+
"eval_steps_per_second": 3.033,
|
| 843 |
+
"step": 240
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 74.76923076923077,
|
| 847 |
+
"eval_accuracy": 0.8450704225352113,
|
| 848 |
+
"eval_loss": 0.5105892419815063,
|
| 849 |
+
"eval_runtime": 1.0212,
|
| 850 |
+
"eval_samples_per_second": 69.527,
|
| 851 |
+
"eval_steps_per_second": 2.938,
|
| 852 |
+
"step": 243
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 76.0,
|
| 856 |
+
"eval_accuracy": 0.8309859154929577,
|
| 857 |
+
"eval_loss": 0.6685615181922913,
|
| 858 |
+
"eval_runtime": 1.0036,
|
| 859 |
+
"eval_samples_per_second": 70.748,
|
| 860 |
+
"eval_steps_per_second": 2.989,
|
| 861 |
+
"step": 247
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 76.92307692307692,
|
| 865 |
+
"grad_norm": 3.6703972816467285,
|
| 866 |
+
"learning_rate": 9.259259259259259e-06,
|
| 867 |
+
"loss": 0.131,
|
| 868 |
+
"step": 250
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 76.92307692307692,
|
| 872 |
+
"eval_accuracy": 0.8309859154929577,
|
| 873 |
+
"eval_loss": 0.7132428288459778,
|
| 874 |
+
"eval_runtime": 0.9879,
|
| 875 |
+
"eval_samples_per_second": 71.873,
|
| 876 |
+
"eval_steps_per_second": 3.037,
|
| 877 |
+
"step": 250
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 77.84615384615384,
|
| 881 |
+
"eval_accuracy": 0.8450704225352113,
|
| 882 |
+
"eval_loss": 0.5945414900779724,
|
| 883 |
+
"eval_runtime": 0.9943,
|
| 884 |
+
"eval_samples_per_second": 71.404,
|
| 885 |
+
"eval_steps_per_second": 3.017,
|
| 886 |
+
"step": 253
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 78.76923076923077,
|
| 890 |
+
"eval_accuracy": 0.7746478873239436,
|
| 891 |
+
"eval_loss": 0.5515730977058411,
|
| 892 |
+
"eval_runtime": 1.0089,
|
| 893 |
+
"eval_samples_per_second": 70.372,
|
| 894 |
+
"eval_steps_per_second": 2.973,
|
| 895 |
+
"step": 256
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 80.0,
|
| 899 |
+
"grad_norm": 1.8021163940429688,
|
| 900 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 901 |
+
"loss": 0.1009,
|
| 902 |
+
"step": 260
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 80.0,
|
| 906 |
+
"eval_accuracy": 0.7605633802816901,
|
| 907 |
+
"eval_loss": 0.547419011592865,
|
| 908 |
+
"eval_runtime": 1.0062,
|
| 909 |
+
"eval_samples_per_second": 70.563,
|
| 910 |
+
"eval_steps_per_second": 2.982,
|
| 911 |
+
"step": 260
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 80.92307692307692,
|
| 915 |
+
"eval_accuracy": 0.7887323943661971,
|
| 916 |
+
"eval_loss": 0.5218686461448669,
|
| 917 |
+
"eval_runtime": 0.9971,
|
| 918 |
+
"eval_samples_per_second": 71.207,
|
| 919 |
+
"eval_steps_per_second": 3.009,
|
| 920 |
+
"step": 263
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 81.84615384615384,
|
| 924 |
+
"eval_accuracy": 0.8450704225352113,
|
| 925 |
+
"eval_loss": 0.5374658107757568,
|
| 926 |
+
"eval_runtime": 0.9899,
|
| 927 |
+
"eval_samples_per_second": 71.721,
|
| 928 |
+
"eval_steps_per_second": 3.03,
|
| 929 |
+
"step": 266
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 82.76923076923077,
|
| 933 |
+
"eval_accuracy": 0.8450704225352113,
|
| 934 |
+
"eval_loss": 0.5133435726165771,
|
| 935 |
+
"eval_runtime": 0.9965,
|
| 936 |
+
"eval_samples_per_second": 71.253,
|
| 937 |
+
"eval_steps_per_second": 3.011,
|
| 938 |
+
"step": 269
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 83.07692307692308,
|
| 942 |
+
"grad_norm": 1.6209955215454102,
|
| 943 |
+
"learning_rate": 5.555555555555556e-06,
|
| 944 |
+
"loss": 0.1084,
|
| 945 |
+
"step": 270
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 84.0,
|
| 949 |
+
"eval_accuracy": 0.8450704225352113,
|
| 950 |
+
"eval_loss": 0.49109983444213867,
|
| 951 |
+
"eval_runtime": 0.9882,
|
| 952 |
+
"eval_samples_per_second": 71.848,
|
| 953 |
+
"eval_steps_per_second": 3.036,
|
| 954 |
+
"step": 273
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 84.92307692307692,
|
| 958 |
+
"eval_accuracy": 0.8732394366197183,
|
| 959 |
+
"eval_loss": 0.49927499890327454,
|
| 960 |
+
"eval_runtime": 1.0035,
|
| 961 |
+
"eval_samples_per_second": 70.749,
|
| 962 |
+
"eval_steps_per_second": 2.989,
|
| 963 |
+
"step": 276
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 85.84615384615384,
|
| 967 |
+
"eval_accuracy": 0.8591549295774648,
|
| 968 |
+
"eval_loss": 0.5418108701705933,
|
| 969 |
+
"eval_runtime": 0.9918,
|
| 970 |
+
"eval_samples_per_second": 71.588,
|
| 971 |
+
"eval_steps_per_second": 3.025,
|
| 972 |
+
"step": 279
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 86.15384615384616,
|
| 976 |
+
"grad_norm": 2.604188919067383,
|
| 977 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 978 |
+
"loss": 0.0851,
|
| 979 |
+
"step": 280
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 86.76923076923077,
|
| 983 |
+
"eval_accuracy": 0.8450704225352113,
|
| 984 |
+
"eval_loss": 0.6010194420814514,
|
| 985 |
+
"eval_runtime": 0.998,
|
| 986 |
+
"eval_samples_per_second": 71.143,
|
| 987 |
+
"eval_steps_per_second": 3.006,
|
| 988 |
+
"step": 282
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 88.0,
|
| 992 |
+
"eval_accuracy": 0.8450704225352113,
|
| 993 |
+
"eval_loss": 0.6304585933685303,
|
| 994 |
+
"eval_runtime": 0.9947,
|
| 995 |
+
"eval_samples_per_second": 71.377,
|
| 996 |
+
"eval_steps_per_second": 3.016,
|
| 997 |
+
"step": 286
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 88.92307692307692,
|
| 1001 |
+
"eval_accuracy": 0.8450704225352113,
|
| 1002 |
+
"eval_loss": 0.6016303300857544,
|
| 1003 |
+
"eval_runtime": 0.9919,
|
| 1004 |
+
"eval_samples_per_second": 71.58,
|
| 1005 |
+
"eval_steps_per_second": 3.025,
|
| 1006 |
+
"step": 289
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 89.23076923076923,
|
| 1010 |
+
"grad_norm": 2.3317203521728516,
|
| 1011 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 1012 |
+
"loss": 0.1071,
|
| 1013 |
+
"step": 290
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 89.84615384615384,
|
| 1017 |
+
"eval_accuracy": 0.8591549295774648,
|
| 1018 |
+
"eval_loss": 0.577277660369873,
|
| 1019 |
+
"eval_runtime": 0.9927,
|
| 1020 |
+
"eval_samples_per_second": 71.522,
|
| 1021 |
+
"eval_steps_per_second": 3.022,
|
| 1022 |
+
"step": 292
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 90.76923076923077,
|
| 1026 |
+
"eval_accuracy": 0.8732394366197183,
|
| 1027 |
+
"eval_loss": 0.5610097646713257,
|
| 1028 |
+
"eval_runtime": 1.0072,
|
| 1029 |
+
"eval_samples_per_second": 70.491,
|
| 1030 |
+
"eval_steps_per_second": 2.978,
|
| 1031 |
+
"step": 295
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 92.0,
|
| 1035 |
+
"eval_accuracy": 0.8732394366197183,
|
| 1036 |
+
"eval_loss": 0.5522463917732239,
|
| 1037 |
+
"eval_runtime": 0.9964,
|
| 1038 |
+
"eval_samples_per_second": 71.258,
|
| 1039 |
+
"eval_steps_per_second": 3.011,
|
| 1040 |
+
"step": 299
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 92.3076923076923,
|
| 1044 |
+
"grad_norm": 2.7261242866516113,
|
| 1045 |
+
"learning_rate": 0.0,
|
| 1046 |
+
"loss": 0.1139,
|
| 1047 |
+
"step": 300
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 92.3076923076923,
|
| 1051 |
+
"eval_accuracy": 0.8732394366197183,
|
| 1052 |
+
"eval_loss": 0.5514280796051025,
|
| 1053 |
+
"eval_runtime": 1.0006,
|
| 1054 |
+
"eval_samples_per_second": 70.955,
|
| 1055 |
+
"eval_steps_per_second": 2.998,
|
| 1056 |
+
"step": 300
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 92.3076923076923,
|
| 1060 |
+
"step": 300,
|
| 1061 |
+
"total_flos": 2.8402872494292173e+18,
|
| 1062 |
+
"train_loss": 0.2501216806968053,
|
| 1063 |
+
"train_runtime": 1631.2353,
|
| 1064 |
+
"train_samples_per_second": 24.337,
|
| 1065 |
+
"train_steps_per_second": 0.184
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 92.3076923076923,
|
| 1069 |
+
"eval_accuracy": 0.8732394366197183,
|
| 1070 |
+
"eval_loss": 0.49927499890327454,
|
| 1071 |
+
"eval_runtime": 1.0275,
|
| 1072 |
+
"eval_samples_per_second": 69.097,
|
| 1073 |
+
"eval_steps_per_second": 2.92,
|
| 1074 |
+
"step": 300
|
| 1075 |
+
}
|
| 1076 |
+
],
|
| 1077 |
+
"logging_steps": 10,
|
| 1078 |
+
"max_steps": 300,
|
| 1079 |
+
"num_input_tokens_seen": 0,
|
| 1080 |
+
"num_train_epochs": 100,
|
| 1081 |
+
"save_steps": 500,
|
| 1082 |
+
"stateful_callbacks": {
|
| 1083 |
+
"TrainerControl": {
|
| 1084 |
+
"args": {
|
| 1085 |
+
"should_epoch_stop": false,
|
| 1086 |
+
"should_evaluate": false,
|
| 1087 |
+
"should_log": false,
|
| 1088 |
+
"should_save": true,
|
| 1089 |
+
"should_training_stop": true
|
| 1090 |
+
},
|
| 1091 |
+
"attributes": {}
|
| 1092 |
+
}
|
| 1093 |
+
},
|
| 1094 |
+
"total_flos": 2.8402872494292173e+18,
|
| 1095 |
+
"train_batch_size": 32,
|
| 1096 |
+
"trial_name": null,
|
| 1097 |
+
"trial_params": null
|
| 1098 |
+
}
|