End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Mar21_01-57-17_721b7bb2497f/events.out.tfevents.1742523304.721b7bb2497f.1961.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +961 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.2829
|
| 22 |
+
- Accuracy: 0.92
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.52173913043478,
|
| 3 |
+
"eval_accuracy": 0.92,
|
| 4 |
+
"eval_loss": 0.28286808729171753,
|
| 5 |
+
"eval_runtime": 2.0374,
|
| 6 |
+
"eval_samples_per_second": 85.895,
|
| 7 |
+
"eval_steps_per_second": 5.399,
|
| 8 |
+
"total_flos": 1.5068369042520146e+18,
|
| 9 |
+
"train_loss": 0.20776909204090344,
|
| 10 |
+
"train_runtime": 1021.41,
|
| 11 |
+
"train_samples_per_second": 47.978,
|
| 12 |
+
"train_steps_per_second": 0.749
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.52173913043478,
|
| 3 |
+
"eval_accuracy": 0.92,
|
| 4 |
+
"eval_loss": 0.28286808729171753,
|
| 5 |
+
"eval_runtime": 2.0374,
|
| 6 |
+
"eval_samples_per_second": 85.895,
|
| 7 |
+
"eval_steps_per_second": 5.399
|
| 8 |
+
}
|
runs/Mar21_01-57-17_721b7bb2497f/events.out.tfevents.1742523304.721b7bb2497f.1961.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c61152531966b45873c78064fbe65d17e1cd8994de04dbccfe45380a8930d7f
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.52173913043478,
|
| 3 |
+
"total_flos": 1.5068369042520146e+18,
|
| 4 |
+
"train_loss": 0.20776909204090344,
|
| 5 |
+
"train_runtime": 1021.41,
|
| 6 |
+
"train_samples_per_second": 47.978,
|
| 7 |
+
"train_steps_per_second": 0.749
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,961 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.92,
|
| 3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV67/checkpoint-216",
|
| 4 |
+
"epoch": 42.52173913043478,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 765,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.5797101449275363,
|
| 13 |
+
"grad_norm": 3.3762147426605225,
|
| 14 |
+
"learning_rate": 5.194805194805194e-06,
|
| 15 |
+
"loss": 1.103,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.4,
|
| 21 |
+
"eval_loss": 1.0588308572769165,
|
| 22 |
+
"eval_runtime": 2.651,
|
| 23 |
+
"eval_samples_per_second": 66.013,
|
| 24 |
+
"eval_steps_per_second": 4.149,
|
| 25 |
+
"step": 18
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.1159420289855073,
|
| 29 |
+
"grad_norm": 3.310516119003296,
|
| 30 |
+
"learning_rate": 1.0389610389610389e-05,
|
| 31 |
+
"loss": 0.9714,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.6956521739130435,
|
| 36 |
+
"grad_norm": 4.865700721740723,
|
| 37 |
+
"learning_rate": 1.5584415584415587e-05,
|
| 38 |
+
"loss": 0.9007,
|
| 39 |
+
"step": 30
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"eval_accuracy": 0.8228571428571428,
|
| 44 |
+
"eval_loss": 0.5650595426559448,
|
| 45 |
+
"eval_runtime": 2.1573,
|
| 46 |
+
"eval_samples_per_second": 81.121,
|
| 47 |
+
"eval_steps_per_second": 5.099,
|
| 48 |
+
"step": 36
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.2318840579710146,
|
| 52 |
+
"grad_norm": 7.906283378601074,
|
| 53 |
+
"learning_rate": 2.0779220779220778e-05,
|
| 54 |
+
"loss": 0.6704,
|
| 55 |
+
"step": 40
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.8115942028985508,
|
| 59 |
+
"grad_norm": 9.154444694519043,
|
| 60 |
+
"learning_rate": 2.5974025974025975e-05,
|
| 61 |
+
"loss": 0.497,
|
| 62 |
+
"step": 50
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.0,
|
| 66 |
+
"eval_accuracy": 0.8742857142857143,
|
| 67 |
+
"eval_loss": 0.35587868094444275,
|
| 68 |
+
"eval_runtime": 1.8852,
|
| 69 |
+
"eval_samples_per_second": 92.83,
|
| 70 |
+
"eval_steps_per_second": 5.835,
|
| 71 |
+
"step": 54
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.3478260869565215,
|
| 75 |
+
"grad_norm": Infinity,
|
| 76 |
+
"learning_rate": 3.064935064935065e-05,
|
| 77 |
+
"loss": 0.4035,
|
| 78 |
+
"step": 60
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.927536231884058,
|
| 82 |
+
"grad_norm": 9.306342124938965,
|
| 83 |
+
"learning_rate": 3.584415584415584e-05,
|
| 84 |
+
"loss": 0.4178,
|
| 85 |
+
"step": 70
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4.0,
|
| 89 |
+
"eval_accuracy": 0.7885714285714286,
|
| 90 |
+
"eval_loss": 0.4171466827392578,
|
| 91 |
+
"eval_runtime": 1.8969,
|
| 92 |
+
"eval_samples_per_second": 92.256,
|
| 93 |
+
"eval_steps_per_second": 5.799,
|
| 94 |
+
"step": 72
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 4.463768115942029,
|
| 98 |
+
"grad_norm": 8.625231742858887,
|
| 99 |
+
"learning_rate": 3.988372093023256e-05,
|
| 100 |
+
"loss": 0.3313,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 5.0,
|
| 105 |
+
"grad_norm": 16.59040069580078,
|
| 106 |
+
"learning_rate": 3.930232558139535e-05,
|
| 107 |
+
"loss": 0.4303,
|
| 108 |
+
"step": 90
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 5.0,
|
| 112 |
+
"eval_accuracy": 0.7314285714285714,
|
| 113 |
+
"eval_loss": 0.688423752784729,
|
| 114 |
+
"eval_runtime": 1.8723,
|
| 115 |
+
"eval_samples_per_second": 93.467,
|
| 116 |
+
"eval_steps_per_second": 5.875,
|
| 117 |
+
"step": 90
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 5.579710144927536,
|
| 121 |
+
"grad_norm": 8.558308601379395,
|
| 122 |
+
"learning_rate": 3.8720930232558147e-05,
|
| 123 |
+
"loss": 0.485,
|
| 124 |
+
"step": 100
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 6.0,
|
| 128 |
+
"eval_accuracy": 0.8628571428571429,
|
| 129 |
+
"eval_loss": 0.3255015015602112,
|
| 130 |
+
"eval_runtime": 1.9531,
|
| 131 |
+
"eval_samples_per_second": 89.6,
|
| 132 |
+
"eval_steps_per_second": 5.632,
|
| 133 |
+
"step": 108
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 6.115942028985507,
|
| 137 |
+
"grad_norm": 7.873001575469971,
|
| 138 |
+
"learning_rate": 3.8139534883720935e-05,
|
| 139 |
+
"loss": 0.3853,
|
| 140 |
+
"step": 110
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.695652173913043,
|
| 144 |
+
"grad_norm": 7.555118560791016,
|
| 145 |
+
"learning_rate": 3.7558139534883724e-05,
|
| 146 |
+
"loss": 0.3345,
|
| 147 |
+
"step": 120
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 7.0,
|
| 151 |
+
"eval_accuracy": 0.9028571428571428,
|
| 152 |
+
"eval_loss": 0.26311492919921875,
|
| 153 |
+
"eval_runtime": 1.9109,
|
| 154 |
+
"eval_samples_per_second": 91.578,
|
| 155 |
+
"eval_steps_per_second": 5.756,
|
| 156 |
+
"step": 126
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 7.231884057971015,
|
| 160 |
+
"grad_norm": 8.322998046875,
|
| 161 |
+
"learning_rate": 3.703488372093023e-05,
|
| 162 |
+
"loss": 0.3761,
|
| 163 |
+
"step": 130
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 7.811594202898551,
|
| 167 |
+
"grad_norm": 6.434593677520752,
|
| 168 |
+
"learning_rate": 3.645348837209302e-05,
|
| 169 |
+
"loss": 0.3058,
|
| 170 |
+
"step": 140
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 8.0,
|
| 174 |
+
"eval_accuracy": 0.8342857142857143,
|
| 175 |
+
"eval_loss": 0.353305846452713,
|
| 176 |
+
"eval_runtime": 2.6568,
|
| 177 |
+
"eval_samples_per_second": 65.868,
|
| 178 |
+
"eval_steps_per_second": 4.14,
|
| 179 |
+
"step": 144
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 8.347826086956522,
|
| 183 |
+
"grad_norm": 6.407224178314209,
|
| 184 |
+
"learning_rate": 3.587209302325582e-05,
|
| 185 |
+
"loss": 0.2499,
|
| 186 |
+
"step": 150
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 8.927536231884059,
|
| 190 |
+
"grad_norm": 6.604982376098633,
|
| 191 |
+
"learning_rate": 3.5290697674418606e-05,
|
| 192 |
+
"loss": 0.35,
|
| 193 |
+
"step": 160
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 9.0,
|
| 197 |
+
"eval_accuracy": 0.8685714285714285,
|
| 198 |
+
"eval_loss": 0.2853247821331024,
|
| 199 |
+
"eval_runtime": 1.9251,
|
| 200 |
+
"eval_samples_per_second": 90.906,
|
| 201 |
+
"eval_steps_per_second": 5.714,
|
| 202 |
+
"step": 162
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 9.46376811594203,
|
| 206 |
+
"grad_norm": 4.880464553833008,
|
| 207 |
+
"learning_rate": 3.4709302325581395e-05,
|
| 208 |
+
"loss": 0.2751,
|
| 209 |
+
"step": 170
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 10.0,
|
| 213 |
+
"grad_norm": 1.717911958694458,
|
| 214 |
+
"learning_rate": 3.412790697674419e-05,
|
| 215 |
+
"loss": 0.2535,
|
| 216 |
+
"step": 180
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 10.0,
|
| 220 |
+
"eval_accuracy": 0.9142857142857143,
|
| 221 |
+
"eval_loss": 0.25287744402885437,
|
| 222 |
+
"eval_runtime": 1.9003,
|
| 223 |
+
"eval_samples_per_second": 92.092,
|
| 224 |
+
"eval_steps_per_second": 5.789,
|
| 225 |
+
"step": 180
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 10.579710144927537,
|
| 229 |
+
"grad_norm": 5.554988384246826,
|
| 230 |
+
"learning_rate": 3.354651162790698e-05,
|
| 231 |
+
"loss": 0.21,
|
| 232 |
+
"step": 190
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 11.0,
|
| 236 |
+
"eval_accuracy": 0.84,
|
| 237 |
+
"eval_loss": 0.3805539309978485,
|
| 238 |
+
"eval_runtime": 1.9566,
|
| 239 |
+
"eval_samples_per_second": 89.442,
|
| 240 |
+
"eval_steps_per_second": 5.622,
|
| 241 |
+
"step": 198
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 11.115942028985508,
|
| 245 |
+
"grad_norm": 10.970183372497559,
|
| 246 |
+
"learning_rate": 3.2965116279069774e-05,
|
| 247 |
+
"loss": 0.2346,
|
| 248 |
+
"step": 200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 11.695652173913043,
|
| 252 |
+
"grad_norm": 7.596284866333008,
|
| 253 |
+
"learning_rate": 3.238372093023256e-05,
|
| 254 |
+
"loss": 0.2414,
|
| 255 |
+
"step": 210
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 12.0,
|
| 259 |
+
"eval_accuracy": 0.92,
|
| 260 |
+
"eval_loss": 0.28286808729171753,
|
| 261 |
+
"eval_runtime": 1.889,
|
| 262 |
+
"eval_samples_per_second": 92.64,
|
| 263 |
+
"eval_steps_per_second": 5.823,
|
| 264 |
+
"step": 216
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.231884057971014,
|
| 268 |
+
"grad_norm": 8.505143165588379,
|
| 269 |
+
"learning_rate": 3.180232558139535e-05,
|
| 270 |
+
"loss": 0.2106,
|
| 271 |
+
"step": 220
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.81159420289855,
|
| 275 |
+
"grad_norm": 3.423171043395996,
|
| 276 |
+
"learning_rate": 3.122093023255815e-05,
|
| 277 |
+
"loss": 0.1978,
|
| 278 |
+
"step": 230
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 13.0,
|
| 282 |
+
"eval_accuracy": 0.9142857142857143,
|
| 283 |
+
"eval_loss": 0.3011384904384613,
|
| 284 |
+
"eval_runtime": 2.3291,
|
| 285 |
+
"eval_samples_per_second": 75.137,
|
| 286 |
+
"eval_steps_per_second": 4.723,
|
| 287 |
+
"step": 234
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 13.347826086956522,
|
| 291 |
+
"grad_norm": 3.8812015056610107,
|
| 292 |
+
"learning_rate": 3.0639534883720936e-05,
|
| 293 |
+
"loss": 0.1973,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.927536231884059,
|
| 298 |
+
"grad_norm": 8.166449546813965,
|
| 299 |
+
"learning_rate": 3.0058139534883725e-05,
|
| 300 |
+
"loss": 0.1683,
|
| 301 |
+
"step": 250
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 14.0,
|
| 305 |
+
"eval_accuracy": 0.9085714285714286,
|
| 306 |
+
"eval_loss": 0.2486036717891693,
|
| 307 |
+
"eval_runtime": 2.3479,
|
| 308 |
+
"eval_samples_per_second": 74.533,
|
| 309 |
+
"eval_steps_per_second": 4.685,
|
| 310 |
+
"step": 252
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 14.46376811594203,
|
| 314 |
+
"grad_norm": 8.5850830078125,
|
| 315 |
+
"learning_rate": 2.9476744186046513e-05,
|
| 316 |
+
"loss": 0.155,
|
| 317 |
+
"step": 260
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 15.0,
|
| 321 |
+
"grad_norm": Infinity,
|
| 322 |
+
"learning_rate": 2.8953488372093022e-05,
|
| 323 |
+
"loss": 0.2351,
|
| 324 |
+
"step": 270
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 15.0,
|
| 328 |
+
"eval_accuracy": 0.8628571428571429,
|
| 329 |
+
"eval_loss": 0.3612350821495056,
|
| 330 |
+
"eval_runtime": 1.9339,
|
| 331 |
+
"eval_samples_per_second": 90.489,
|
| 332 |
+
"eval_steps_per_second": 5.688,
|
| 333 |
+
"step": 270
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 15.579710144927537,
|
| 337 |
+
"grad_norm": 5.161411762237549,
|
| 338 |
+
"learning_rate": 2.8372093023255818e-05,
|
| 339 |
+
"loss": 0.264,
|
| 340 |
+
"step": 280
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 16.0,
|
| 344 |
+
"eval_accuracy": 0.88,
|
| 345 |
+
"eval_loss": 0.3643089234828949,
|
| 346 |
+
"eval_runtime": 1.933,
|
| 347 |
+
"eval_samples_per_second": 90.533,
|
| 348 |
+
"eval_steps_per_second": 5.691,
|
| 349 |
+
"step": 288
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 16.115942028985508,
|
| 353 |
+
"grad_norm": 11.968738555908203,
|
| 354 |
+
"learning_rate": 2.7790697674418607e-05,
|
| 355 |
+
"loss": 0.1644,
|
| 356 |
+
"step": 290
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 16.695652173913043,
|
| 360 |
+
"grad_norm": 7.685915470123291,
|
| 361 |
+
"learning_rate": 2.72093023255814e-05,
|
| 362 |
+
"loss": 0.1714,
|
| 363 |
+
"step": 300
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 17.0,
|
| 367 |
+
"eval_accuracy": 0.9085714285714286,
|
| 368 |
+
"eval_loss": 0.2481045126914978,
|
| 369 |
+
"eval_runtime": 1.8847,
|
| 370 |
+
"eval_samples_per_second": 92.855,
|
| 371 |
+
"eval_steps_per_second": 5.837,
|
| 372 |
+
"step": 306
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 17.231884057971016,
|
| 376 |
+
"grad_norm": 12.054829597473145,
|
| 377 |
+
"learning_rate": 2.6627906976744187e-05,
|
| 378 |
+
"loss": 0.174,
|
| 379 |
+
"step": 310
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 17.81159420289855,
|
| 383 |
+
"grad_norm": 10.886869430541992,
|
| 384 |
+
"learning_rate": 2.604651162790698e-05,
|
| 385 |
+
"loss": 0.1714,
|
| 386 |
+
"step": 320
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 18.0,
|
| 390 |
+
"eval_accuracy": 0.8914285714285715,
|
| 391 |
+
"eval_loss": 0.3478798568248749,
|
| 392 |
+
"eval_runtime": 1.897,
|
| 393 |
+
"eval_samples_per_second": 92.251,
|
| 394 |
+
"eval_steps_per_second": 5.799,
|
| 395 |
+
"step": 324
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 18.347826086956523,
|
| 399 |
+
"grad_norm": 6.273326396942139,
|
| 400 |
+
"learning_rate": 2.546511627906977e-05,
|
| 401 |
+
"loss": 0.1274,
|
| 402 |
+
"step": 330
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 18.92753623188406,
|
| 406 |
+
"grad_norm": 21.078323364257812,
|
| 407 |
+
"learning_rate": 2.488372093023256e-05,
|
| 408 |
+
"loss": 0.1886,
|
| 409 |
+
"step": 340
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 19.0,
|
| 413 |
+
"eval_accuracy": 0.9028571428571428,
|
| 414 |
+
"eval_loss": 0.26444798707962036,
|
| 415 |
+
"eval_runtime": 2.6918,
|
| 416 |
+
"eval_samples_per_second": 65.013,
|
| 417 |
+
"eval_steps_per_second": 4.087,
|
| 418 |
+
"step": 342
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 19.463768115942027,
|
| 422 |
+
"grad_norm": 10.183343887329102,
|
| 423 |
+
"learning_rate": 2.430232558139535e-05,
|
| 424 |
+
"loss": 0.1328,
|
| 425 |
+
"step": 350
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 20.0,
|
| 429 |
+
"grad_norm": 0.006570586934685707,
|
| 430 |
+
"learning_rate": 2.3720930232558144e-05,
|
| 431 |
+
"loss": 0.1522,
|
| 432 |
+
"step": 360
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 20.0,
|
| 436 |
+
"eval_accuracy": 0.8971428571428571,
|
| 437 |
+
"eval_loss": 0.25868120789527893,
|
| 438 |
+
"eval_runtime": 2.0121,
|
| 439 |
+
"eval_samples_per_second": 86.974,
|
| 440 |
+
"eval_steps_per_second": 5.467,
|
| 441 |
+
"step": 360
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 20.579710144927535,
|
| 445 |
+
"grad_norm": 12.063268661499023,
|
| 446 |
+
"learning_rate": 2.3139534883720933e-05,
|
| 447 |
+
"loss": 0.1468,
|
| 448 |
+
"step": 370
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 21.0,
|
| 452 |
+
"eval_accuracy": 0.9028571428571428,
|
| 453 |
+
"eval_loss": 0.28319162130355835,
|
| 454 |
+
"eval_runtime": 1.9232,
|
| 455 |
+
"eval_samples_per_second": 90.993,
|
| 456 |
+
"eval_steps_per_second": 5.72,
|
| 457 |
+
"step": 378
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 21.115942028985508,
|
| 461 |
+
"grad_norm": 5.412624835968018,
|
| 462 |
+
"learning_rate": 2.2558139534883722e-05,
|
| 463 |
+
"loss": 0.1137,
|
| 464 |
+
"step": 380
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 21.695652173913043,
|
| 468 |
+
"grad_norm": 10.886478424072266,
|
| 469 |
+
"learning_rate": 2.1976744186046514e-05,
|
| 470 |
+
"loss": 0.1364,
|
| 471 |
+
"step": 390
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 22.0,
|
| 475 |
+
"eval_accuracy": 0.9028571428571428,
|
| 476 |
+
"eval_loss": 0.28295958042144775,
|
| 477 |
+
"eval_runtime": 1.9359,
|
| 478 |
+
"eval_samples_per_second": 90.399,
|
| 479 |
+
"eval_steps_per_second": 5.682,
|
| 480 |
+
"step": 396
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 22.231884057971016,
|
| 484 |
+
"grad_norm": 15.01888370513916,
|
| 485 |
+
"learning_rate": 2.1395348837209303e-05,
|
| 486 |
+
"loss": 0.1216,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 22.81159420289855,
|
| 491 |
+
"grad_norm": 6.896238803863525,
|
| 492 |
+
"learning_rate": 2.0813953488372095e-05,
|
| 493 |
+
"loss": 0.1294,
|
| 494 |
+
"step": 410
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 23.0,
|
| 498 |
+
"eval_accuracy": 0.8914285714285715,
|
| 499 |
+
"eval_loss": 0.295413076877594,
|
| 500 |
+
"eval_runtime": 1.9151,
|
| 501 |
+
"eval_samples_per_second": 91.377,
|
| 502 |
+
"eval_steps_per_second": 5.744,
|
| 503 |
+
"step": 414
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 23.347826086956523,
|
| 507 |
+
"grad_norm": 9.244548797607422,
|
| 508 |
+
"learning_rate": 2.0232558139534887e-05,
|
| 509 |
+
"loss": 0.0822,
|
| 510 |
+
"step": 420
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 23.92753623188406,
|
| 514 |
+
"grad_norm": 4.16889762878418,
|
| 515 |
+
"learning_rate": 1.9651162790697676e-05,
|
| 516 |
+
"loss": 0.122,
|
| 517 |
+
"step": 430
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 24.0,
|
| 521 |
+
"eval_accuracy": 0.8742857142857143,
|
| 522 |
+
"eval_loss": 0.38006919622421265,
|
| 523 |
+
"eval_runtime": 1.9716,
|
| 524 |
+
"eval_samples_per_second": 88.759,
|
| 525 |
+
"eval_steps_per_second": 5.579,
|
| 526 |
+
"step": 432
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 24.463768115942027,
|
| 530 |
+
"grad_norm": 10.813605308532715,
|
| 531 |
+
"learning_rate": 1.9069767441860468e-05,
|
| 532 |
+
"loss": 0.1355,
|
| 533 |
+
"step": 440
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 25.0,
|
| 537 |
+
"grad_norm": 2.4495270252227783,
|
| 538 |
+
"learning_rate": 1.8488372093023256e-05,
|
| 539 |
+
"loss": 0.1114,
|
| 540 |
+
"step": 450
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 25.0,
|
| 544 |
+
"eval_accuracy": 0.9028571428571428,
|
| 545 |
+
"eval_loss": 0.3375105857849121,
|
| 546 |
+
"eval_runtime": 2.7139,
|
| 547 |
+
"eval_samples_per_second": 64.484,
|
| 548 |
+
"eval_steps_per_second": 4.053,
|
| 549 |
+
"step": 450
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 25.579710144927535,
|
| 553 |
+
"grad_norm": 6.9816718101501465,
|
| 554 |
+
"learning_rate": 1.790697674418605e-05,
|
| 555 |
+
"loss": 0.12,
|
| 556 |
+
"step": 460
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 26.0,
|
| 560 |
+
"eval_accuracy": 0.8742857142857143,
|
| 561 |
+
"eval_loss": 0.3695812225341797,
|
| 562 |
+
"eval_runtime": 2.0343,
|
| 563 |
+
"eval_samples_per_second": 86.026,
|
| 564 |
+
"eval_steps_per_second": 5.407,
|
| 565 |
+
"step": 468
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 26.115942028985508,
|
| 569 |
+
"grad_norm": 3.3524887561798096,
|
| 570 |
+
"learning_rate": 1.7325581395348837e-05,
|
| 571 |
+
"loss": 0.0994,
|
| 572 |
+
"step": 470
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 26.695652173913043,
|
| 576 |
+
"grad_norm": 4.345656394958496,
|
| 577 |
+
"learning_rate": 1.674418604651163e-05,
|
| 578 |
+
"loss": 0.121,
|
| 579 |
+
"step": 480
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 27.0,
|
| 583 |
+
"eval_accuracy": 0.8857142857142857,
|
| 584 |
+
"eval_loss": 0.3459794819355011,
|
| 585 |
+
"eval_runtime": 1.922,
|
| 586 |
+
"eval_samples_per_second": 91.052,
|
| 587 |
+
"eval_steps_per_second": 5.723,
|
| 588 |
+
"step": 486
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 27.231884057971016,
|
| 592 |
+
"grad_norm": 11.216938972473145,
|
| 593 |
+
"learning_rate": 1.616279069767442e-05,
|
| 594 |
+
"loss": 0.0978,
|
| 595 |
+
"step": 490
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 27.81159420289855,
|
| 599 |
+
"grad_norm": 7.896313190460205,
|
| 600 |
+
"learning_rate": 1.558139534883721e-05,
|
| 601 |
+
"loss": 0.1056,
|
| 602 |
+
"step": 500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 28.0,
|
| 606 |
+
"eval_accuracy": 0.9028571428571428,
|
| 607 |
+
"eval_loss": 0.32958656549453735,
|
| 608 |
+
"eval_runtime": 1.9313,
|
| 609 |
+
"eval_samples_per_second": 90.611,
|
| 610 |
+
"eval_steps_per_second": 5.696,
|
| 611 |
+
"step": 504
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 28.347826086956523,
|
| 615 |
+
"grad_norm": 1.8364324569702148,
|
| 616 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 617 |
+
"loss": 0.0804,
|
| 618 |
+
"step": 510
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 28.92753623188406,
|
| 622 |
+
"grad_norm": 2.498980760574341,
|
| 623 |
+
"learning_rate": 1.441860465116279e-05,
|
| 624 |
+
"loss": 0.0941,
|
| 625 |
+
"step": 520
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 29.0,
|
| 629 |
+
"eval_accuracy": 0.8971428571428571,
|
| 630 |
+
"eval_loss": 0.39773935079574585,
|
| 631 |
+
"eval_runtime": 1.9043,
|
| 632 |
+
"eval_samples_per_second": 91.895,
|
| 633 |
+
"eval_steps_per_second": 5.776,
|
| 634 |
+
"step": 522
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 29.463768115942027,
|
| 638 |
+
"grad_norm": 3.4750986099243164,
|
| 639 |
+
"learning_rate": 1.3837209302325583e-05,
|
| 640 |
+
"loss": 0.1048,
|
| 641 |
+
"step": 530
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 30.0,
|
| 645 |
+
"grad_norm": 0.39407047629356384,
|
| 646 |
+
"learning_rate": 1.3255813953488372e-05,
|
| 647 |
+
"loss": 0.0882,
|
| 648 |
+
"step": 540
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 30.0,
|
| 652 |
+
"eval_accuracy": 0.8914285714285715,
|
| 653 |
+
"eval_loss": 0.3436221778392792,
|
| 654 |
+
"eval_runtime": 2.1059,
|
| 655 |
+
"eval_samples_per_second": 83.098,
|
| 656 |
+
"eval_steps_per_second": 5.223,
|
| 657 |
+
"step": 540
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 30.579710144927535,
|
| 661 |
+
"grad_norm": 11.971905708312988,
|
| 662 |
+
"learning_rate": 1.2674418604651164e-05,
|
| 663 |
+
"loss": 0.1131,
|
| 664 |
+
"step": 550
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 31.0,
|
| 668 |
+
"eval_accuracy": 0.8914285714285715,
|
| 669 |
+
"eval_loss": 0.33969902992248535,
|
| 670 |
+
"eval_runtime": 2.7032,
|
| 671 |
+
"eval_samples_per_second": 64.737,
|
| 672 |
+
"eval_steps_per_second": 4.069,
|
| 673 |
+
"step": 558
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 31.115942028985508,
|
| 677 |
+
"grad_norm": 6.7430925369262695,
|
| 678 |
+
"learning_rate": 1.2093023255813954e-05,
|
| 679 |
+
"loss": 0.0945,
|
| 680 |
+
"step": 560
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 31.695652173913043,
|
| 684 |
+
"grad_norm": 5.155873775482178,
|
| 685 |
+
"learning_rate": 1.1511627906976746e-05,
|
| 686 |
+
"loss": 0.0959,
|
| 687 |
+
"step": 570
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 32.0,
|
| 691 |
+
"eval_accuracy": 0.8971428571428571,
|
| 692 |
+
"eval_loss": 0.34200194478034973,
|
| 693 |
+
"eval_runtime": 1.9416,
|
| 694 |
+
"eval_samples_per_second": 90.13,
|
| 695 |
+
"eval_steps_per_second": 5.665,
|
| 696 |
+
"step": 576
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 32.231884057971016,
|
| 700 |
+
"grad_norm": 1.955030918121338,
|
| 701 |
+
"learning_rate": 1.0930232558139535e-05,
|
| 702 |
+
"loss": 0.0719,
|
| 703 |
+
"step": 580
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 32.81159420289855,
|
| 707 |
+
"grad_norm": 8.749945640563965,
|
| 708 |
+
"learning_rate": 1.0348837209302327e-05,
|
| 709 |
+
"loss": 0.1058,
|
| 710 |
+
"step": 590
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 33.0,
|
| 714 |
+
"eval_accuracy": 0.8971428571428571,
|
| 715 |
+
"eval_loss": 0.3312114477157593,
|
| 716 |
+
"eval_runtime": 1.9046,
|
| 717 |
+
"eval_samples_per_second": 91.883,
|
| 718 |
+
"eval_steps_per_second": 5.776,
|
| 719 |
+
"step": 594
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 33.34782608695652,
|
| 723 |
+
"grad_norm": 8.832493782043457,
|
| 724 |
+
"learning_rate": 9.767441860465117e-06,
|
| 725 |
+
"loss": 0.0619,
|
| 726 |
+
"step": 600
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 33.927536231884055,
|
| 730 |
+
"grad_norm": 5.939527988433838,
|
| 731 |
+
"learning_rate": 9.186046511627908e-06,
|
| 732 |
+
"loss": 0.0707,
|
| 733 |
+
"step": 610
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 34.0,
|
| 737 |
+
"eval_accuracy": 0.8857142857142857,
|
| 738 |
+
"eval_loss": 0.39173805713653564,
|
| 739 |
+
"eval_runtime": 1.9652,
|
| 740 |
+
"eval_samples_per_second": 89.048,
|
| 741 |
+
"eval_steps_per_second": 5.597,
|
| 742 |
+
"step": 612
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 34.46376811594203,
|
| 746 |
+
"grad_norm": 12.942173957824707,
|
| 747 |
+
"learning_rate": 8.604651162790698e-06,
|
| 748 |
+
"loss": 0.0672,
|
| 749 |
+
"step": 620
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 35.0,
|
| 753 |
+
"grad_norm": 0.0012061038287356496,
|
| 754 |
+
"learning_rate": 8.023255813953488e-06,
|
| 755 |
+
"loss": 0.0885,
|
| 756 |
+
"step": 630
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 35.0,
|
| 760 |
+
"eval_accuracy": 0.88,
|
| 761 |
+
"eval_loss": 0.38554221391677856,
|
| 762 |
+
"eval_runtime": 1.9365,
|
| 763 |
+
"eval_samples_per_second": 90.369,
|
| 764 |
+
"eval_steps_per_second": 5.68,
|
| 765 |
+
"step": 630
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 35.57971014492754,
|
| 769 |
+
"grad_norm": 6.067467212677002,
|
| 770 |
+
"learning_rate": 7.44186046511628e-06,
|
| 771 |
+
"loss": 0.0788,
|
| 772 |
+
"step": 640
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 36.0,
|
| 776 |
+
"eval_accuracy": 0.8857142857142857,
|
| 777 |
+
"eval_loss": 0.351922869682312,
|
| 778 |
+
"eval_runtime": 2.1646,
|
| 779 |
+
"eval_samples_per_second": 80.845,
|
| 780 |
+
"eval_steps_per_second": 5.082,
|
| 781 |
+
"step": 648
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 36.11594202898551,
|
| 785 |
+
"grad_norm": 9.353009223937988,
|
| 786 |
+
"learning_rate": 6.86046511627907e-06,
|
| 787 |
+
"loss": 0.0696,
|
| 788 |
+
"step": 650
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 36.69565217391305,
|
| 792 |
+
"grad_norm": 12.263450622558594,
|
| 793 |
+
"learning_rate": 6.279069767441861e-06,
|
| 794 |
+
"loss": 0.112,
|
| 795 |
+
"step": 660
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 37.0,
|
| 799 |
+
"eval_accuracy": 0.8914285714285715,
|
| 800 |
+
"eval_loss": 0.34730595350265503,
|
| 801 |
+
"eval_runtime": 2.7239,
|
| 802 |
+
"eval_samples_per_second": 64.245,
|
| 803 |
+
"eval_steps_per_second": 4.038,
|
| 804 |
+
"step": 666
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 37.231884057971016,
|
| 808 |
+
"grad_norm": 2.59305477142334,
|
| 809 |
+
"learning_rate": 5.697674418604652e-06,
|
| 810 |
+
"loss": 0.0503,
|
| 811 |
+
"step": 670
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 37.81159420289855,
|
| 815 |
+
"grad_norm": 11.254059791564941,
|
| 816 |
+
"learning_rate": 5.116279069767442e-06,
|
| 817 |
+
"loss": 0.0588,
|
| 818 |
+
"step": 680
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 38.0,
|
| 822 |
+
"eval_accuracy": 0.9028571428571428,
|
| 823 |
+
"eval_loss": 0.3717617392539978,
|
| 824 |
+
"eval_runtime": 1.969,
|
| 825 |
+
"eval_samples_per_second": 88.876,
|
| 826 |
+
"eval_steps_per_second": 5.586,
|
| 827 |
+
"step": 684
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 38.34782608695652,
|
| 831 |
+
"grad_norm": 1.6452008485794067,
|
| 832 |
+
"learning_rate": 4.5348837209302326e-06,
|
| 833 |
+
"loss": 0.0613,
|
| 834 |
+
"step": 690
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 38.927536231884055,
|
| 838 |
+
"grad_norm": 7.862698078155518,
|
| 839 |
+
"learning_rate": 3.953488372093024e-06,
|
| 840 |
+
"loss": 0.0963,
|
| 841 |
+
"step": 700
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 39.0,
|
| 845 |
+
"eval_accuracy": 0.88,
|
| 846 |
+
"eval_loss": 0.4021558463573456,
|
| 847 |
+
"eval_runtime": 1.8965,
|
| 848 |
+
"eval_samples_per_second": 92.275,
|
| 849 |
+
"eval_steps_per_second": 5.8,
|
| 850 |
+
"step": 702
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 39.46376811594203,
|
| 854 |
+
"grad_norm": 5.627374649047852,
|
| 855 |
+
"learning_rate": 3.372093023255814e-06,
|
| 856 |
+
"loss": 0.0698,
|
| 857 |
+
"step": 710
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 40.0,
|
| 861 |
+
"grad_norm": 0.006953197065740824,
|
| 862 |
+
"learning_rate": 2.790697674418605e-06,
|
| 863 |
+
"loss": 0.0681,
|
| 864 |
+
"step": 720
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 40.0,
|
| 868 |
+
"eval_accuracy": 0.8971428571428571,
|
| 869 |
+
"eval_loss": 0.3574194312095642,
|
| 870 |
+
"eval_runtime": 1.9589,
|
| 871 |
+
"eval_samples_per_second": 89.336,
|
| 872 |
+
"eval_steps_per_second": 5.615,
|
| 873 |
+
"step": 720
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 40.57971014492754,
|
| 877 |
+
"grad_norm": 8.213858604431152,
|
| 878 |
+
"learning_rate": 2.2093023255813954e-06,
|
| 879 |
+
"loss": 0.0841,
|
| 880 |
+
"step": 730
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 41.0,
|
| 884 |
+
"eval_accuracy": 0.88,
|
| 885 |
+
"eval_loss": 0.3620830476284027,
|
| 886 |
+
"eval_runtime": 1.8779,
|
| 887 |
+
"eval_samples_per_second": 93.191,
|
| 888 |
+
"eval_steps_per_second": 5.858,
|
| 889 |
+
"step": 738
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 41.11594202898551,
|
| 893 |
+
"grad_norm": 2.5701255798339844,
|
| 894 |
+
"learning_rate": 1.6279069767441862e-06,
|
| 895 |
+
"loss": 0.0518,
|
| 896 |
+
"step": 740
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 41.69565217391305,
|
| 900 |
+
"grad_norm": 8.129728317260742,
|
| 901 |
+
"learning_rate": 1.0465116279069768e-06,
|
| 902 |
+
"loss": 0.0739,
|
| 903 |
+
"step": 750
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 42.0,
|
| 907 |
+
"eval_accuracy": 0.8914285714285715,
|
| 908 |
+
"eval_loss": 0.37818607687950134,
|
| 909 |
+
"eval_runtime": 1.9244,
|
| 910 |
+
"eval_samples_per_second": 90.936,
|
| 911 |
+
"eval_steps_per_second": 5.716,
|
| 912 |
+
"step": 756
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 42.231884057971016,
|
| 916 |
+
"grad_norm": 12.24697494506836,
|
| 917 |
+
"learning_rate": 4.651162790697675e-07,
|
| 918 |
+
"loss": 0.0649,
|
| 919 |
+
"step": 760
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 42.52173913043478,
|
| 923 |
+
"eval_accuracy": 0.8914285714285715,
|
| 924 |
+
"eval_loss": 0.3765674829483032,
|
| 925 |
+
"eval_runtime": 2.2844,
|
| 926 |
+
"eval_samples_per_second": 76.607,
|
| 927 |
+
"eval_steps_per_second": 4.815,
|
| 928 |
+
"step": 765
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 42.52173913043478,
|
| 932 |
+
"step": 765,
|
| 933 |
+
"total_flos": 1.5068369042520146e+18,
|
| 934 |
+
"train_loss": 0.20776909204090344,
|
| 935 |
+
"train_runtime": 1021.41,
|
| 936 |
+
"train_samples_per_second": 47.978,
|
| 937 |
+
"train_steps_per_second": 0.749
|
| 938 |
+
}
|
| 939 |
+
],
|
| 940 |
+
"logging_steps": 10,
|
| 941 |
+
"max_steps": 765,
|
| 942 |
+
"num_input_tokens_seen": 0,
|
| 943 |
+
"num_train_epochs": 45,
|
| 944 |
+
"save_steps": 500,
|
| 945 |
+
"stateful_callbacks": {
|
| 946 |
+
"TrainerControl": {
|
| 947 |
+
"args": {
|
| 948 |
+
"should_epoch_stop": false,
|
| 949 |
+
"should_evaluate": false,
|
| 950 |
+
"should_log": false,
|
| 951 |
+
"should_save": true,
|
| 952 |
+
"should_training_stop": true
|
| 953 |
+
},
|
| 954 |
+
"attributes": {}
|
| 955 |
+
}
|
| 956 |
+
},
|
| 957 |
+
"total_flos": 1.5068369042520146e+18,
|
| 958 |
+
"train_batch_size": 16,
|
| 959 |
+
"trial_name": null,
|
| 960 |
+
"trial_params": null
|
| 961 |
+
}
|