End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Mar21_02-25-49_721b7bb2497f/events.out.tfevents.1742525036.721b7bb2497f.9420.1 +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: 0.2868
|
| 22 |
+
- Accuracy: 0.9314
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.52173913043478,
|
| 3 |
+
"eval_accuracy": 0.9314285714285714,
|
| 4 |
+
"eval_loss": 0.2867558002471924,
|
| 5 |
+
"eval_runtime": 2.0591,
|
| 6 |
+
"eval_samples_per_second": 84.987,
|
| 7 |
+
"eval_steps_per_second": 5.342,
|
| 8 |
+
"total_flos": 1.5068369042520146e+18,
|
| 9 |
+
"train_loss": 0.22573306322876924,
|
| 10 |
+
"train_runtime": 1056.0428,
|
| 11 |
+
"train_samples_per_second": 46.404,
|
| 12 |
+
"train_steps_per_second": 0.724
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.52173913043478,
|
| 3 |
+
"eval_accuracy": 0.9314285714285714,
|
| 4 |
+
"eval_loss": 0.2867558002471924,
|
| 5 |
+
"eval_runtime": 2.0591,
|
| 6 |
+
"eval_samples_per_second": 84.987,
|
| 7 |
+
"eval_steps_per_second": 5.342
|
| 8 |
+
}
|
runs/Mar21_02-25-49_721b7bb2497f/events.out.tfevents.1742525036.721b7bb2497f.9420.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89b57e4af34c58bed942b810f0602c2fb819ca0c22abe610df6e3fb950b9b3cc
|
| 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": 0.22573306322876924,
|
| 5 |
+
"train_runtime": 1056.0428,
|
| 6 |
+
"train_samples_per_second": 46.404,
|
| 7 |
+
"train_steps_per_second": 0.724
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,961 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9314285714285714,
|
| 3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV68/checkpoint-720",
|
| 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": 3.7087650299072266,
|
| 14 |
+
"learning_rate": 3.896103896103896e-06,
|
| 15 |
+
"loss": 1.0651,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.4685714285714286,
|
| 21 |
+
"eval_loss": 1.0769113302230835,
|
| 22 |
+
"eval_runtime": 1.9723,
|
| 23 |
+
"eval_samples_per_second": 88.729,
|
| 24 |
+
"eval_steps_per_second": 5.577,
|
| 25 |
+
"step": 18
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.1159420289855073,
|
| 29 |
+
"grad_norm": 3.6832127571105957,
|
| 30 |
+
"learning_rate": 7.792207792207792e-06,
|
| 31 |
+
"loss": 0.9566,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.6956521739130435,
|
| 36 |
+
"grad_norm": 3.8521084785461426,
|
| 37 |
+
"learning_rate": 1.168831168831169e-05,
|
| 38 |
+
"loss": 0.9503,
|
| 39 |
+
"step": 30
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"eval_accuracy": 0.7257142857142858,
|
| 44 |
+
"eval_loss": 0.8111104965209961,
|
| 45 |
+
"eval_runtime": 1.9367,
|
| 46 |
+
"eval_samples_per_second": 90.359,
|
| 47 |
+
"eval_steps_per_second": 5.68,
|
| 48 |
+
"step": 36
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.2318840579710146,
|
| 52 |
+
"grad_norm": 5.586236476898193,
|
| 53 |
+
"learning_rate": 1.5584415584415583e-05,
|
| 54 |
+
"loss": 0.7219,
|
| 55 |
+
"step": 40
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.8115942028985508,
|
| 59 |
+
"grad_norm": 8.486871719360352,
|
| 60 |
+
"learning_rate": 1.948051948051948e-05,
|
| 61 |
+
"loss": 0.5745,
|
| 62 |
+
"step": 50
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.0,
|
| 66 |
+
"eval_accuracy": 0.7314285714285714,
|
| 67 |
+
"eval_loss": 0.4972275197505951,
|
| 68 |
+
"eval_runtime": 1.904,
|
| 69 |
+
"eval_samples_per_second": 91.91,
|
| 70 |
+
"eval_steps_per_second": 5.777,
|
| 71 |
+
"step": 54
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.3478260869565215,
|
| 75 |
+
"grad_norm": 10.05865478515625,
|
| 76 |
+
"learning_rate": 2.337662337662338e-05,
|
| 77 |
+
"loss": 0.4438,
|
| 78 |
+
"step": 60
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.927536231884058,
|
| 82 |
+
"grad_norm": 12.097763061523438,
|
| 83 |
+
"learning_rate": 2.6883116883116883e-05,
|
| 84 |
+
"loss": 0.4746,
|
| 85 |
+
"step": 70
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4.0,
|
| 89 |
+
"eval_accuracy": 0.7485714285714286,
|
| 90 |
+
"eval_loss": 0.478755921125412,
|
| 91 |
+
"eval_runtime": 1.9296,
|
| 92 |
+
"eval_samples_per_second": 90.693,
|
| 93 |
+
"eval_steps_per_second": 5.701,
|
| 94 |
+
"step": 72
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 4.463768115942029,
|
| 98 |
+
"grad_norm": 5.598774433135986,
|
| 99 |
+
"learning_rate": 2.991279069767442e-05,
|
| 100 |
+
"loss": 0.3793,
|
| 101 |
+
"step": 80
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 5.0,
|
| 105 |
+
"grad_norm": 3.355344295501709,
|
| 106 |
+
"learning_rate": 2.947674418604651e-05,
|
| 107 |
+
"loss": 0.4363,
|
| 108 |
+
"step": 90
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 5.0,
|
| 112 |
+
"eval_accuracy": 0.7314285714285714,
|
| 113 |
+
"eval_loss": 0.5427026748657227,
|
| 114 |
+
"eval_runtime": 1.9262,
|
| 115 |
+
"eval_samples_per_second": 90.853,
|
| 116 |
+
"eval_steps_per_second": 5.711,
|
| 117 |
+
"step": 90
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 5.579710144927536,
|
| 121 |
+
"grad_norm": 10.868864059448242,
|
| 122 |
+
"learning_rate": 2.9040697674418607e-05,
|
| 123 |
+
"loss": 0.4362,
|
| 124 |
+
"step": 100
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 6.0,
|
| 128 |
+
"eval_accuracy": 0.8685714285714285,
|
| 129 |
+
"eval_loss": 0.3581157624721527,
|
| 130 |
+
"eval_runtime": 1.8969,
|
| 131 |
+
"eval_samples_per_second": 92.257,
|
| 132 |
+
"eval_steps_per_second": 5.799,
|
| 133 |
+
"step": 108
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 6.115942028985507,
|
| 137 |
+
"grad_norm": 18.57970428466797,
|
| 138 |
+
"learning_rate": 2.86046511627907e-05,
|
| 139 |
+
"loss": 0.3911,
|
| 140 |
+
"step": 110
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.695652173913043,
|
| 144 |
+
"grad_norm": 11.689884185791016,
|
| 145 |
+
"learning_rate": 2.8168604651162793e-05,
|
| 146 |
+
"loss": 0.3476,
|
| 147 |
+
"step": 120
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 7.0,
|
| 151 |
+
"eval_accuracy": 0.8685714285714285,
|
| 152 |
+
"eval_loss": 0.3572000563144684,
|
| 153 |
+
"eval_runtime": 1.9174,
|
| 154 |
+
"eval_samples_per_second": 91.27,
|
| 155 |
+
"eval_steps_per_second": 5.737,
|
| 156 |
+
"step": 126
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 7.231884057971015,
|
| 160 |
+
"grad_norm": 7.682909965515137,
|
| 161 |
+
"learning_rate": 2.7776162790697673e-05,
|
| 162 |
+
"loss": 0.3861,
|
| 163 |
+
"step": 130
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 7.811594202898551,
|
| 167 |
+
"grad_norm": 9.239214897155762,
|
| 168 |
+
"learning_rate": 2.7340116279069766e-05,
|
| 169 |
+
"loss": 0.3113,
|
| 170 |
+
"step": 140
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 8.0,
|
| 174 |
+
"eval_accuracy": 0.7885714285714286,
|
| 175 |
+
"eval_loss": 0.4334910809993744,
|
| 176 |
+
"eval_runtime": 2.4972,
|
| 177 |
+
"eval_samples_per_second": 70.078,
|
| 178 |
+
"eval_steps_per_second": 4.405,
|
| 179 |
+
"step": 144
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 8.347826086956522,
|
| 183 |
+
"grad_norm": 10.808152198791504,
|
| 184 |
+
"learning_rate": 2.6904069767441863e-05,
|
| 185 |
+
"loss": 0.2685,
|
| 186 |
+
"step": 150
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 8.927536231884059,
|
| 190 |
+
"grad_norm": 11.47179889678955,
|
| 191 |
+
"learning_rate": 2.6468023255813956e-05,
|
| 192 |
+
"loss": 0.3943,
|
| 193 |
+
"step": 160
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 9.0,
|
| 197 |
+
"eval_accuracy": 0.8685714285714285,
|
| 198 |
+
"eval_loss": 0.27819445729255676,
|
| 199 |
+
"eval_runtime": 2.4012,
|
| 200 |
+
"eval_samples_per_second": 72.88,
|
| 201 |
+
"eval_steps_per_second": 4.581,
|
| 202 |
+
"step": 162
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 9.46376811594203,
|
| 206 |
+
"grad_norm": 6.136239051818848,
|
| 207 |
+
"learning_rate": 2.6031976744186046e-05,
|
| 208 |
+
"loss": 0.2876,
|
| 209 |
+
"step": 170
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 10.0,
|
| 213 |
+
"grad_norm": 1.0991929769515991,
|
| 214 |
+
"learning_rate": 2.559593023255814e-05,
|
| 215 |
+
"loss": 0.2574,
|
| 216 |
+
"step": 180
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 10.0,
|
| 220 |
+
"eval_accuracy": 0.8685714285714285,
|
| 221 |
+
"eval_loss": 0.3320414125919342,
|
| 222 |
+
"eval_runtime": 1.9305,
|
| 223 |
+
"eval_samples_per_second": 90.652,
|
| 224 |
+
"eval_steps_per_second": 5.698,
|
| 225 |
+
"step": 180
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 10.579710144927537,
|
| 229 |
+
"grad_norm": 9.614068984985352,
|
| 230 |
+
"learning_rate": 2.5159883720930236e-05,
|
| 231 |
+
"loss": 0.2345,
|
| 232 |
+
"step": 190
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 11.0,
|
| 236 |
+
"eval_accuracy": 0.8342857142857143,
|
| 237 |
+
"eval_loss": 0.4383019506931305,
|
| 238 |
+
"eval_runtime": 1.9283,
|
| 239 |
+
"eval_samples_per_second": 90.753,
|
| 240 |
+
"eval_steps_per_second": 5.704,
|
| 241 |
+
"step": 198
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 11.115942028985508,
|
| 245 |
+
"grad_norm": 10.854966163635254,
|
| 246 |
+
"learning_rate": 2.4723837209302326e-05,
|
| 247 |
+
"loss": 0.2437,
|
| 248 |
+
"step": 200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 11.695652173913043,
|
| 252 |
+
"grad_norm": 8.31966495513916,
|
| 253 |
+
"learning_rate": 2.428779069767442e-05,
|
| 254 |
+
"loss": 0.3002,
|
| 255 |
+
"step": 210
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 12.0,
|
| 259 |
+
"eval_accuracy": 0.8685714285714285,
|
| 260 |
+
"eval_loss": 0.3052798807621002,
|
| 261 |
+
"eval_runtime": 1.9211,
|
| 262 |
+
"eval_samples_per_second": 91.096,
|
| 263 |
+
"eval_steps_per_second": 5.726,
|
| 264 |
+
"step": 216
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.231884057971014,
|
| 268 |
+
"grad_norm": 12.32070541381836,
|
| 269 |
+
"learning_rate": 2.3851744186046512e-05,
|
| 270 |
+
"loss": 0.2268,
|
| 271 |
+
"step": 220
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.81159420289855,
|
| 275 |
+
"grad_norm": 5.423739433288574,
|
| 276 |
+
"learning_rate": 2.3415697674418605e-05,
|
| 277 |
+
"loss": 0.2038,
|
| 278 |
+
"step": 230
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 13.0,
|
| 282 |
+
"eval_accuracy": 0.8742857142857143,
|
| 283 |
+
"eval_loss": 0.3189180791378021,
|
| 284 |
+
"eval_runtime": 1.9021,
|
| 285 |
+
"eval_samples_per_second": 92.005,
|
| 286 |
+
"eval_steps_per_second": 5.783,
|
| 287 |
+
"step": 234
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 13.347826086956522,
|
| 291 |
+
"grad_norm": 6.77400016784668,
|
| 292 |
+
"learning_rate": 2.29796511627907e-05,
|
| 293 |
+
"loss": 0.2148,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.927536231884059,
|
| 298 |
+
"grad_norm": 7.098783016204834,
|
| 299 |
+
"learning_rate": 2.2543604651162792e-05,
|
| 300 |
+
"loss": 0.2244,
|
| 301 |
+
"step": 250
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 14.0,
|
| 305 |
+
"eval_accuracy": 0.8742857142857143,
|
| 306 |
+
"eval_loss": 0.27661752700805664,
|
| 307 |
+
"eval_runtime": 1.9403,
|
| 308 |
+
"eval_samples_per_second": 90.191,
|
| 309 |
+
"eval_steps_per_second": 5.669,
|
| 310 |
+
"step": 252
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 14.46376811594203,
|
| 314 |
+
"grad_norm": 8.359834671020508,
|
| 315 |
+
"learning_rate": 2.2107558139534885e-05,
|
| 316 |
+
"loss": 0.1859,
|
| 317 |
+
"step": 260
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 15.0,
|
| 321 |
+
"grad_norm": 25.05173683166504,
|
| 322 |
+
"learning_rate": 2.1671511627906975e-05,
|
| 323 |
+
"loss": 0.2277,
|
| 324 |
+
"step": 270
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 15.0,
|
| 328 |
+
"eval_accuracy": 0.8857142857142857,
|
| 329 |
+
"eval_loss": 0.2637348771095276,
|
| 330 |
+
"eval_runtime": 2.2383,
|
| 331 |
+
"eval_samples_per_second": 78.185,
|
| 332 |
+
"eval_steps_per_second": 4.914,
|
| 333 |
+
"step": 270
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 15.579710144927537,
|
| 337 |
+
"grad_norm": 6.100148677825928,
|
| 338 |
+
"learning_rate": 2.123546511627907e-05,
|
| 339 |
+
"loss": 0.2318,
|
| 340 |
+
"step": 280
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 16.0,
|
| 344 |
+
"eval_accuracy": 0.8114285714285714,
|
| 345 |
+
"eval_loss": 0.46118730306625366,
|
| 346 |
+
"eval_runtime": 2.6144,
|
| 347 |
+
"eval_samples_per_second": 66.937,
|
| 348 |
+
"eval_steps_per_second": 4.207,
|
| 349 |
+
"step": 288
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 16.115942028985508,
|
| 353 |
+
"grad_norm": 6.950288772583008,
|
| 354 |
+
"learning_rate": 2.0799418604651165e-05,
|
| 355 |
+
"loss": 0.1786,
|
| 356 |
+
"step": 290
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 16.695652173913043,
|
| 360 |
+
"grad_norm": 14.06019401550293,
|
| 361 |
+
"learning_rate": 2.0363372093023254e-05,
|
| 362 |
+
"loss": 0.1908,
|
| 363 |
+
"step": 300
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 17.0,
|
| 367 |
+
"eval_accuracy": 0.8857142857142857,
|
| 368 |
+
"eval_loss": 0.31668975949287415,
|
| 369 |
+
"eval_runtime": 4.7899,
|
| 370 |
+
"eval_samples_per_second": 36.535,
|
| 371 |
+
"eval_steps_per_second": 2.297,
|
| 372 |
+
"step": 306
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 17.231884057971016,
|
| 376 |
+
"grad_norm": 14.938794136047363,
|
| 377 |
+
"learning_rate": 1.9927325581395348e-05,
|
| 378 |
+
"loss": 0.2174,
|
| 379 |
+
"step": 310
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 17.81159420289855,
|
| 383 |
+
"grad_norm": 10.554563522338867,
|
| 384 |
+
"learning_rate": 1.9491279069767444e-05,
|
| 385 |
+
"loss": 0.1932,
|
| 386 |
+
"step": 320
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 18.0,
|
| 390 |
+
"eval_accuracy": 0.9028571428571428,
|
| 391 |
+
"eval_loss": 0.29490649700164795,
|
| 392 |
+
"eval_runtime": 1.9334,
|
| 393 |
+
"eval_samples_per_second": 90.514,
|
| 394 |
+
"eval_steps_per_second": 5.689,
|
| 395 |
+
"step": 324
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 18.347826086956523,
|
| 399 |
+
"grad_norm": 11.751145362854004,
|
| 400 |
+
"learning_rate": 1.9055232558139538e-05,
|
| 401 |
+
"loss": 0.1432,
|
| 402 |
+
"step": 330
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 18.92753623188406,
|
| 406 |
+
"grad_norm": 9.6201171875,
|
| 407 |
+
"learning_rate": 1.8619186046511627e-05,
|
| 408 |
+
"loss": 0.1676,
|
| 409 |
+
"step": 340
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 19.0,
|
| 413 |
+
"eval_accuracy": 0.9085714285714286,
|
| 414 |
+
"eval_loss": 0.26273345947265625,
|
| 415 |
+
"eval_runtime": 2.1044,
|
| 416 |
+
"eval_samples_per_second": 83.159,
|
| 417 |
+
"eval_steps_per_second": 5.227,
|
| 418 |
+
"step": 342
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 19.463768115942027,
|
| 422 |
+
"grad_norm": 5.4566731452941895,
|
| 423 |
+
"learning_rate": 1.822674418604651e-05,
|
| 424 |
+
"loss": 0.1472,
|
| 425 |
+
"step": 350
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 20.0,
|
| 429 |
+
"grad_norm": 0.2691696584224701,
|
| 430 |
+
"learning_rate": 1.7790697674418608e-05,
|
| 431 |
+
"loss": 0.1442,
|
| 432 |
+
"step": 360
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 20.0,
|
| 436 |
+
"eval_accuracy": 0.9142857142857143,
|
| 437 |
+
"eval_loss": 0.2584407329559326,
|
| 438 |
+
"eval_runtime": 4.6506,
|
| 439 |
+
"eval_samples_per_second": 37.629,
|
| 440 |
+
"eval_steps_per_second": 2.365,
|
| 441 |
+
"step": 360
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 20.579710144927535,
|
| 445 |
+
"grad_norm": 14.349730491638184,
|
| 446 |
+
"learning_rate": 1.7354651162790697e-05,
|
| 447 |
+
"loss": 0.1606,
|
| 448 |
+
"step": 370
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 21.0,
|
| 452 |
+
"eval_accuracy": 0.9142857142857143,
|
| 453 |
+
"eval_loss": 0.2625867426395416,
|
| 454 |
+
"eval_runtime": 2.0381,
|
| 455 |
+
"eval_samples_per_second": 85.864,
|
| 456 |
+
"eval_steps_per_second": 5.397,
|
| 457 |
+
"step": 378
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 21.115942028985508,
|
| 461 |
+
"grad_norm": 6.261329650878906,
|
| 462 |
+
"learning_rate": 1.691860465116279e-05,
|
| 463 |
+
"loss": 0.1291,
|
| 464 |
+
"step": 380
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 21.695652173913043,
|
| 468 |
+
"grad_norm": 11.669342994689941,
|
| 469 |
+
"learning_rate": 1.6482558139534884e-05,
|
| 470 |
+
"loss": 0.1624,
|
| 471 |
+
"step": 390
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 22.0,
|
| 475 |
+
"eval_accuracy": 0.9257142857142857,
|
| 476 |
+
"eval_loss": 0.2351078987121582,
|
| 477 |
+
"eval_runtime": 2.7162,
|
| 478 |
+
"eval_samples_per_second": 64.427,
|
| 479 |
+
"eval_steps_per_second": 4.05,
|
| 480 |
+
"step": 396
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 22.231884057971016,
|
| 484 |
+
"grad_norm": 9.660877227783203,
|
| 485 |
+
"learning_rate": 1.6046511627906977e-05,
|
| 486 |
+
"loss": 0.1447,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 22.81159420289855,
|
| 491 |
+
"grad_norm": 15.922120094299316,
|
| 492 |
+
"learning_rate": 1.561046511627907e-05,
|
| 493 |
+
"loss": 0.1735,
|
| 494 |
+
"step": 410
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 23.0,
|
| 498 |
+
"eval_accuracy": 0.9257142857142857,
|
| 499 |
+
"eval_loss": 0.27460697293281555,
|
| 500 |
+
"eval_runtime": 2.0043,
|
| 501 |
+
"eval_samples_per_second": 87.311,
|
| 502 |
+
"eval_steps_per_second": 5.488,
|
| 503 |
+
"step": 414
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 23.347826086956523,
|
| 507 |
+
"grad_norm": 7.466541767120361,
|
| 508 |
+
"learning_rate": 1.5174418604651163e-05,
|
| 509 |
+
"loss": 0.0943,
|
| 510 |
+
"step": 420
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 23.92753623188406,
|
| 514 |
+
"grad_norm": 7.257338047027588,
|
| 515 |
+
"learning_rate": 1.4738372093023255e-05,
|
| 516 |
+
"loss": 0.1604,
|
| 517 |
+
"step": 430
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 24.0,
|
| 521 |
+
"eval_accuracy": 0.8914285714285715,
|
| 522 |
+
"eval_loss": 0.3236704170703888,
|
| 523 |
+
"eval_runtime": 1.9055,
|
| 524 |
+
"eval_samples_per_second": 91.842,
|
| 525 |
+
"eval_steps_per_second": 5.773,
|
| 526 |
+
"step": 432
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 24.463768115942027,
|
| 530 |
+
"grad_norm": 8.80977725982666,
|
| 531 |
+
"learning_rate": 1.430232558139535e-05,
|
| 532 |
+
"loss": 0.1495,
|
| 533 |
+
"step": 440
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 25.0,
|
| 537 |
+
"grad_norm": 1.1711186170578003,
|
| 538 |
+
"learning_rate": 1.3866279069767441e-05,
|
| 539 |
+
"loss": 0.122,
|
| 540 |
+
"step": 450
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 25.0,
|
| 544 |
+
"eval_accuracy": 0.8914285714285715,
|
| 545 |
+
"eval_loss": 0.28520700335502625,
|
| 546 |
+
"eval_runtime": 1.9027,
|
| 547 |
+
"eval_samples_per_second": 91.974,
|
| 548 |
+
"eval_steps_per_second": 5.781,
|
| 549 |
+
"step": 450
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 25.579710144927535,
|
| 553 |
+
"grad_norm": 10.047913551330566,
|
| 554 |
+
"learning_rate": 1.3430232558139536e-05,
|
| 555 |
+
"loss": 0.1447,
|
| 556 |
+
"step": 460
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 26.0,
|
| 560 |
+
"eval_accuracy": 0.92,
|
| 561 |
+
"eval_loss": 0.2593800127506256,
|
| 562 |
+
"eval_runtime": 1.9095,
|
| 563 |
+
"eval_samples_per_second": 91.646,
|
| 564 |
+
"eval_steps_per_second": 5.761,
|
| 565 |
+
"step": 468
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 26.115942028985508,
|
| 569 |
+
"grad_norm": 4.829050064086914,
|
| 570 |
+
"learning_rate": 1.2994186046511628e-05,
|
| 571 |
+
"loss": 0.1318,
|
| 572 |
+
"step": 470
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 26.695652173913043,
|
| 576 |
+
"grad_norm": 3.806666851043701,
|
| 577 |
+
"learning_rate": 1.2558139534883723e-05,
|
| 578 |
+
"loss": 0.1265,
|
| 579 |
+
"step": 480
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 27.0,
|
| 583 |
+
"eval_accuracy": 0.9028571428571428,
|
| 584 |
+
"eval_loss": 0.28569361567497253,
|
| 585 |
+
"eval_runtime": 1.9576,
|
| 586 |
+
"eval_samples_per_second": 89.395,
|
| 587 |
+
"eval_steps_per_second": 5.619,
|
| 588 |
+
"step": 486
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 27.231884057971016,
|
| 592 |
+
"grad_norm": 13.416851043701172,
|
| 593 |
+
"learning_rate": 1.2122093023255814e-05,
|
| 594 |
+
"loss": 0.1198,
|
| 595 |
+
"step": 490
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 27.81159420289855,
|
| 599 |
+
"grad_norm": 16.207693099975586,
|
| 600 |
+
"learning_rate": 1.1686046511627907e-05,
|
| 601 |
+
"loss": 0.1265,
|
| 602 |
+
"step": 500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 28.0,
|
| 606 |
+
"eval_accuracy": 0.8742857142857143,
|
| 607 |
+
"eval_loss": 0.32380279898643494,
|
| 608 |
+
"eval_runtime": 2.636,
|
| 609 |
+
"eval_samples_per_second": 66.388,
|
| 610 |
+
"eval_steps_per_second": 4.173,
|
| 611 |
+
"step": 504
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 28.347826086956523,
|
| 615 |
+
"grad_norm": 6.766117572784424,
|
| 616 |
+
"learning_rate": 1.125e-05,
|
| 617 |
+
"loss": 0.0857,
|
| 618 |
+
"step": 510
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 28.92753623188406,
|
| 622 |
+
"grad_norm": 2.6189181804656982,
|
| 623 |
+
"learning_rate": 1.0813953488372092e-05,
|
| 624 |
+
"loss": 0.122,
|
| 625 |
+
"step": 520
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 29.0,
|
| 629 |
+
"eval_accuracy": 0.8857142857142857,
|
| 630 |
+
"eval_loss": 0.30293118953704834,
|
| 631 |
+
"eval_runtime": 2.1444,
|
| 632 |
+
"eval_samples_per_second": 81.609,
|
| 633 |
+
"eval_steps_per_second": 5.13,
|
| 634 |
+
"step": 522
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 29.463768115942027,
|
| 638 |
+
"grad_norm": 7.471499443054199,
|
| 639 |
+
"learning_rate": 1.0377906976744187e-05,
|
| 640 |
+
"loss": 0.1182,
|
| 641 |
+
"step": 530
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 30.0,
|
| 645 |
+
"grad_norm": 0.14097607135772705,
|
| 646 |
+
"learning_rate": 9.941860465116279e-06,
|
| 647 |
+
"loss": 0.0929,
|
| 648 |
+
"step": 540
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 30.0,
|
| 652 |
+
"eval_accuracy": 0.9028571428571428,
|
| 653 |
+
"eval_loss": 0.2936091423034668,
|
| 654 |
+
"eval_runtime": 1.9297,
|
| 655 |
+
"eval_samples_per_second": 90.69,
|
| 656 |
+
"eval_steps_per_second": 5.7,
|
| 657 |
+
"step": 540
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 30.579710144927535,
|
| 661 |
+
"grad_norm": 9.875027656555176,
|
| 662 |
+
"learning_rate": 9.505813953488372e-06,
|
| 663 |
+
"loss": 0.1276,
|
| 664 |
+
"step": 550
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 31.0,
|
| 668 |
+
"eval_accuracy": 0.9142857142857143,
|
| 669 |
+
"eval_loss": 0.2777470052242279,
|
| 670 |
+
"eval_runtime": 1.9462,
|
| 671 |
+
"eval_samples_per_second": 89.918,
|
| 672 |
+
"eval_steps_per_second": 5.652,
|
| 673 |
+
"step": 558
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 31.115942028985508,
|
| 677 |
+
"grad_norm": 10.335461616516113,
|
| 678 |
+
"learning_rate": 9.069767441860465e-06,
|
| 679 |
+
"loss": 0.1278,
|
| 680 |
+
"step": 560
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 31.695652173913043,
|
| 684 |
+
"grad_norm": 5.649544715881348,
|
| 685 |
+
"learning_rate": 8.633720930232558e-06,
|
| 686 |
+
"loss": 0.1118,
|
| 687 |
+
"step": 570
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 32.0,
|
| 691 |
+
"eval_accuracy": 0.9142857142857143,
|
| 692 |
+
"eval_loss": 0.2812146544456482,
|
| 693 |
+
"eval_runtime": 1.9255,
|
| 694 |
+
"eval_samples_per_second": 90.885,
|
| 695 |
+
"eval_steps_per_second": 5.713,
|
| 696 |
+
"step": 576
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 32.231884057971016,
|
| 700 |
+
"grad_norm": 3.817037582397461,
|
| 701 |
+
"learning_rate": 8.197674418604652e-06,
|
| 702 |
+
"loss": 0.1021,
|
| 703 |
+
"step": 580
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 32.81159420289855,
|
| 707 |
+
"grad_norm": 8.079113960266113,
|
| 708 |
+
"learning_rate": 7.761627906976745e-06,
|
| 709 |
+
"loss": 0.1058,
|
| 710 |
+
"step": 590
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 33.0,
|
| 714 |
+
"eval_accuracy": 0.92,
|
| 715 |
+
"eval_loss": 0.2924804389476776,
|
| 716 |
+
"eval_runtime": 1.9242,
|
| 717 |
+
"eval_samples_per_second": 90.947,
|
| 718 |
+
"eval_steps_per_second": 5.717,
|
| 719 |
+
"step": 594
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 33.34782608695652,
|
| 723 |
+
"grad_norm": 7.013218402862549,
|
| 724 |
+
"learning_rate": 7.325581395348837e-06,
|
| 725 |
+
"loss": 0.0822,
|
| 726 |
+
"step": 600
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 33.927536231884055,
|
| 730 |
+
"grad_norm": 5.116663455963135,
|
| 731 |
+
"learning_rate": 6.88953488372093e-06,
|
| 732 |
+
"loss": 0.0824,
|
| 733 |
+
"step": 610
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 34.0,
|
| 737 |
+
"eval_accuracy": 0.8914285714285715,
|
| 738 |
+
"eval_loss": 0.35194164514541626,
|
| 739 |
+
"eval_runtime": 2.0353,
|
| 740 |
+
"eval_samples_per_second": 85.982,
|
| 741 |
+
"eval_steps_per_second": 5.405,
|
| 742 |
+
"step": 612
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 34.46376811594203,
|
| 746 |
+
"grad_norm": 7.844626426696777,
|
| 747 |
+
"learning_rate": 6.453488372093024e-06,
|
| 748 |
+
"loss": 0.0959,
|
| 749 |
+
"step": 620
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 35.0,
|
| 753 |
+
"grad_norm": 0.001561504672281444,
|
| 754 |
+
"learning_rate": 6.017441860465116e-06,
|
| 755 |
+
"loss": 0.1084,
|
| 756 |
+
"step": 630
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 35.0,
|
| 760 |
+
"eval_accuracy": 0.92,
|
| 761 |
+
"eval_loss": 0.28469112515449524,
|
| 762 |
+
"eval_runtime": 2.7076,
|
| 763 |
+
"eval_samples_per_second": 64.634,
|
| 764 |
+
"eval_steps_per_second": 4.063,
|
| 765 |
+
"step": 630
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 35.57971014492754,
|
| 769 |
+
"grad_norm": 6.455016136169434,
|
| 770 |
+
"learning_rate": 5.581395348837209e-06,
|
| 771 |
+
"loss": 0.1074,
|
| 772 |
+
"step": 640
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 36.0,
|
| 776 |
+
"eval_accuracy": 0.9142857142857143,
|
| 777 |
+
"eval_loss": 0.27351856231689453,
|
| 778 |
+
"eval_runtime": 1.9489,
|
| 779 |
+
"eval_samples_per_second": 89.796,
|
| 780 |
+
"eval_steps_per_second": 5.644,
|
| 781 |
+
"step": 648
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 36.11594202898551,
|
| 785 |
+
"grad_norm": 4.332115650177002,
|
| 786 |
+
"learning_rate": 5.145348837209302e-06,
|
| 787 |
+
"loss": 0.0896,
|
| 788 |
+
"step": 650
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 36.69565217391305,
|
| 792 |
+
"grad_norm": 17.33997917175293,
|
| 793 |
+
"learning_rate": 4.709302325581396e-06,
|
| 794 |
+
"loss": 0.1415,
|
| 795 |
+
"step": 660
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 37.0,
|
| 799 |
+
"eval_accuracy": 0.9257142857142857,
|
| 800 |
+
"eval_loss": 0.27236348390579224,
|
| 801 |
+
"eval_runtime": 1.9276,
|
| 802 |
+
"eval_samples_per_second": 90.787,
|
| 803 |
+
"eval_steps_per_second": 5.707,
|
| 804 |
+
"step": 666
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 37.231884057971016,
|
| 808 |
+
"grad_norm": 5.769250392913818,
|
| 809 |
+
"learning_rate": 4.273255813953489e-06,
|
| 810 |
+
"loss": 0.0783,
|
| 811 |
+
"step": 670
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 37.81159420289855,
|
| 815 |
+
"grad_norm": 8.946736335754395,
|
| 816 |
+
"learning_rate": 3.837209302325582e-06,
|
| 817 |
+
"loss": 0.0702,
|
| 818 |
+
"step": 680
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 38.0,
|
| 822 |
+
"eval_accuracy": 0.92,
|
| 823 |
+
"eval_loss": 0.28733909130096436,
|
| 824 |
+
"eval_runtime": 1.9313,
|
| 825 |
+
"eval_samples_per_second": 90.614,
|
| 826 |
+
"eval_steps_per_second": 5.696,
|
| 827 |
+
"step": 684
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 38.34782608695652,
|
| 831 |
+
"grad_norm": 3.792422294616699,
|
| 832 |
+
"learning_rate": 3.4011627906976744e-06,
|
| 833 |
+
"loss": 0.0708,
|
| 834 |
+
"step": 690
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 38.927536231884055,
|
| 838 |
+
"grad_norm": 10.4917573928833,
|
| 839 |
+
"learning_rate": 2.965116279069767e-06,
|
| 840 |
+
"loss": 0.0987,
|
| 841 |
+
"step": 700
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 39.0,
|
| 845 |
+
"eval_accuracy": 0.92,
|
| 846 |
+
"eval_loss": 0.292362242937088,
|
| 847 |
+
"eval_runtime": 1.9062,
|
| 848 |
+
"eval_samples_per_second": 91.808,
|
| 849 |
+
"eval_steps_per_second": 5.771,
|
| 850 |
+
"step": 702
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 39.46376811594203,
|
| 854 |
+
"grad_norm": 5.309926509857178,
|
| 855 |
+
"learning_rate": 2.5290697674418604e-06,
|
| 856 |
+
"loss": 0.0749,
|
| 857 |
+
"step": 710
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 40.0,
|
| 861 |
+
"grad_norm": 0.004375269636511803,
|
| 862 |
+
"learning_rate": 2.0930232558139536e-06,
|
| 863 |
+
"loss": 0.0637,
|
| 864 |
+
"step": 720
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 40.0,
|
| 868 |
+
"eval_accuracy": 0.9314285714285714,
|
| 869 |
+
"eval_loss": 0.2867558002471924,
|
| 870 |
+
"eval_runtime": 1.9415,
|
| 871 |
+
"eval_samples_per_second": 90.138,
|
| 872 |
+
"eval_steps_per_second": 5.666,
|
| 873 |
+
"step": 720
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 40.57971014492754,
|
| 877 |
+
"grad_norm": 8.937047004699707,
|
| 878 |
+
"learning_rate": 1.6569767441860467e-06,
|
| 879 |
+
"loss": 0.1183,
|
| 880 |
+
"step": 730
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 41.0,
|
| 884 |
+
"eval_accuracy": 0.92,
|
| 885 |
+
"eval_loss": 0.28917399048805237,
|
| 886 |
+
"eval_runtime": 1.9784,
|
| 887 |
+
"eval_samples_per_second": 88.457,
|
| 888 |
+
"eval_steps_per_second": 5.56,
|
| 889 |
+
"step": 738
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 41.11594202898551,
|
| 893 |
+
"grad_norm": 2.839174747467041,
|
| 894 |
+
"learning_rate": 1.2209302325581397e-06,
|
| 895 |
+
"loss": 0.0732,
|
| 896 |
+
"step": 740
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 41.69565217391305,
|
| 900 |
+
"grad_norm": 9.90665054321289,
|
| 901 |
+
"learning_rate": 7.848837209302327e-07,
|
| 902 |
+
"loss": 0.096,
|
| 903 |
+
"step": 750
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 42.0,
|
| 907 |
+
"eval_accuracy": 0.9142857142857143,
|
| 908 |
+
"eval_loss": 0.29103556275367737,
|
| 909 |
+
"eval_runtime": 2.0391,
|
| 910 |
+
"eval_samples_per_second": 85.824,
|
| 911 |
+
"eval_steps_per_second": 5.395,
|
| 912 |
+
"step": 756
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 42.231884057971016,
|
| 916 |
+
"grad_norm": 10.19823932647705,
|
| 917 |
+
"learning_rate": 3.4883720930232557e-07,
|
| 918 |
+
"loss": 0.0719,
|
| 919 |
+
"step": 760
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 42.52173913043478,
|
| 923 |
+
"eval_accuracy": 0.9142857142857143,
|
| 924 |
+
"eval_loss": 0.28974097967147827,
|
| 925 |
+
"eval_runtime": 2.2909,
|
| 926 |
+
"eval_samples_per_second": 76.39,
|
| 927 |
+
"eval_steps_per_second": 4.802,
|
| 928 |
+
"step": 765
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 42.52173913043478,
|
| 932 |
+
"step": 765,
|
| 933 |
+
"total_flos": 1.5068369042520146e+18,
|
| 934 |
+
"train_loss": 0.22573306322876924,
|
| 935 |
+
"train_runtime": 1056.0428,
|
| 936 |
+
"train_samples_per_second": 46.404,
|
| 937 |
+
"train_steps_per_second": 0.724
|
| 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 |
+
}
|