commit files to HF hub
Browse files- README.md +22 -5
- all_results.json +15 -0
- eval_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +2095 -0
README.md
CHANGED
|
@@ -1,22 +1,35 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
|
|
|
|
|
|
| 4 |
metrics:
|
| 5 |
- accuracy
|
| 6 |
model-index:
|
| 7 |
-
- name:
|
| 8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
---
|
| 10 |
|
| 11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
|
| 14 |
-
#
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [P1ayer-1/pythia-deduped-1b-chat-base](https://huggingface.co/P1ayer-1/pythia-deduped-1b-chat-base) on the
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Accuracy: 0.3842
|
| 19 |
- Loss: 4.9648
|
|
|
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -89,3 +102,7 @@ The following hyperparameters were used during training:
|
|
| 89 |
- Pytorch 2.0.0+cu117
|
| 90 |
- Datasets 2.11.0
|
| 91 |
- Tokenizers 0.13.3
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- pile-instruct/
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
| 8 |
model-index:
|
| 9 |
+
- name: layer_4,5,6,7,8
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
type: text-generation
|
| 13 |
+
name: Causal Language Modeling
|
| 14 |
+
dataset:
|
| 15 |
+
name: pile-instruct/
|
| 16 |
+
type: pile-instruct/
|
| 17 |
+
split: None
|
| 18 |
+
metrics:
|
| 19 |
+
- type: accuracy
|
| 20 |
+
value: 0.38424293893426953
|
| 21 |
+
name: Accuracy
|
| 22 |
---
|
| 23 |
|
| 24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 25 |
should probably proofread and complete it, then remove this comment. -->
|
| 26 |
|
| 27 |
+
# layer_4,5,6,7,8
|
| 28 |
|
| 29 |
+
This model is a fine-tuned version of [P1ayer-1/pythia-deduped-1b-chat-base](https://huggingface.co/P1ayer-1/pythia-deduped-1b-chat-base) on the pile-instruct/ dataset.
|
| 30 |
It achieves the following results on the evaluation set:
|
|
|
|
| 31 |
- Loss: 4.9648
|
| 32 |
+
- Accuracy: 0.3842
|
| 33 |
|
| 34 |
## Model description
|
| 35 |
|
|
|
|
| 102 |
- Pytorch 2.0.0+cu117
|
| 103 |
- Datasets 2.11.0
|
| 104 |
- Tokenizers 0.13.3
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
## Wandb Report
|
| 108 |
+
https://wandb.ai/ontocord/pythia-1b-deduped-layer-test-min-pile-instruct/runs/zad9qli2
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.99,
|
| 3 |
+
"eval_accuracy": 0.38424293893426953,
|
| 4 |
+
"eval_loss": 4.96484375,
|
| 5 |
+
"eval_runtime": 549.6112,
|
| 6 |
+
"eval_samples": 10321,
|
| 7 |
+
"eval_samples_per_second": 18.779,
|
| 8 |
+
"eval_steps_per_second": 0.295,
|
| 9 |
+
"perplexity": 143.28616006164455,
|
| 10 |
+
"train_loss": 0.0007667992834527579,
|
| 11 |
+
"train_runtime": 4.5998,
|
| 12 |
+
"train_samples": 192515,
|
| 13 |
+
"train_samples_per_second": 125223.897,
|
| 14 |
+
"train_steps_per_second": 1304.416
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.99,
|
| 3 |
+
"eval_accuracy": 0.38424293893426953,
|
| 4 |
+
"eval_loss": 4.96484375,
|
| 5 |
+
"eval_runtime": 549.6112,
|
| 6 |
+
"eval_samples": 10321,
|
| 7 |
+
"eval_samples_per_second": 18.779,
|
| 8 |
+
"eval_steps_per_second": 0.295,
|
| 9 |
+
"perplexity": 143.28616006164455
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.99,
|
| 3 |
+
"train_loss": 0.0007667992834527579,
|
| 4 |
+
"train_runtime": 4.5998,
|
| 5 |
+
"train_samples": 192515,
|
| 6 |
+
"train_samples_per_second": 125223.897,
|
| 7 |
+
"train_steps_per_second": 1304.416
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2095 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9915254237288136,
|
| 5 |
+
"global_step": 6001,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 9.974991663887962e-05,
|
| 13 |
+
"loss": 9.7321,
|
| 14 |
+
"step": 20
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 9.941647215738579e-05,
|
| 19 |
+
"loss": 8.8682,
|
| 20 |
+
"step": 40
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 9.908302767589197e-05,
|
| 25 |
+
"loss": 8.5062,
|
| 26 |
+
"step": 60
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 9.874958319439813e-05,
|
| 31 |
+
"loss": 8.2194,
|
| 32 |
+
"step": 80
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.05,
|
| 36 |
+
"learning_rate": 9.841613871290431e-05,
|
| 37 |
+
"loss": 8.0283,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06,
|
| 42 |
+
"learning_rate": 9.808269423141048e-05,
|
| 43 |
+
"loss": 7.9013,
|
| 44 |
+
"step": 120
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07,
|
| 48 |
+
"learning_rate": 9.774924974991663e-05,
|
| 49 |
+
"loss": 7.7487,
|
| 50 |
+
"step": 140
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.08,
|
| 54 |
+
"learning_rate": 9.741580526842281e-05,
|
| 55 |
+
"loss": 7.6568,
|
| 56 |
+
"step": 160
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.09,
|
| 60 |
+
"learning_rate": 9.708236078692898e-05,
|
| 61 |
+
"loss": 7.5793,
|
| 62 |
+
"step": 180
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.1,
|
| 66 |
+
"learning_rate": 9.674891630543515e-05,
|
| 67 |
+
"loss": 7.4574,
|
| 68 |
+
"step": 200
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.1,
|
| 72 |
+
"eval_accuracy": 0.16882355811759567,
|
| 73 |
+
"eval_loss": 7.49609375,
|
| 74 |
+
"eval_runtime": 975.6082,
|
| 75 |
+
"eval_samples_per_second": 10.579,
|
| 76 |
+
"eval_steps_per_second": 0.166,
|
| 77 |
+
"step": 200
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.11,
|
| 81 |
+
"learning_rate": 9.641547182394132e-05,
|
| 82 |
+
"loss": 7.4258,
|
| 83 |
+
"step": 220
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.12,
|
| 87 |
+
"learning_rate": 9.608202734244749e-05,
|
| 88 |
+
"loss": 7.3778,
|
| 89 |
+
"step": 240
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.13,
|
| 93 |
+
"learning_rate": 9.574858286095365e-05,
|
| 94 |
+
"loss": 7.3281,
|
| 95 |
+
"step": 260
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.14,
|
| 99 |
+
"learning_rate": 9.541513837945982e-05,
|
| 100 |
+
"loss": 7.2731,
|
| 101 |
+
"step": 280
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.15,
|
| 105 |
+
"learning_rate": 9.508169389796599e-05,
|
| 106 |
+
"loss": 7.2491,
|
| 107 |
+
"step": 300
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.16,
|
| 111 |
+
"learning_rate": 9.474824941647216e-05,
|
| 112 |
+
"loss": 7.1846,
|
| 113 |
+
"step": 320
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.17,
|
| 117 |
+
"learning_rate": 9.441480493497833e-05,
|
| 118 |
+
"loss": 7.131,
|
| 119 |
+
"step": 340
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.18,
|
| 123 |
+
"learning_rate": 9.408136045348451e-05,
|
| 124 |
+
"loss": 7.1157,
|
| 125 |
+
"step": 360
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.19,
|
| 129 |
+
"learning_rate": 9.374791597199066e-05,
|
| 130 |
+
"loss": 7.0675,
|
| 131 |
+
"step": 380
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.2,
|
| 135 |
+
"learning_rate": 9.341447149049683e-05,
|
| 136 |
+
"loss": 7.0445,
|
| 137 |
+
"step": 400
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.2,
|
| 141 |
+
"eval_accuracy": 0.19973070589428632,
|
| 142 |
+
"eval_loss": 7.0546875,
|
| 143 |
+
"eval_runtime": 892.7377,
|
| 144 |
+
"eval_samples_per_second": 11.561,
|
| 145 |
+
"eval_steps_per_second": 0.181,
|
| 146 |
+
"step": 400
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.21,
|
| 150 |
+
"learning_rate": 9.308102700900301e-05,
|
| 151 |
+
"loss": 7.0021,
|
| 152 |
+
"step": 420
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.22,
|
| 156 |
+
"learning_rate": 9.274758252750917e-05,
|
| 157 |
+
"loss": 6.9485,
|
| 158 |
+
"step": 440
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.23,
|
| 162 |
+
"learning_rate": 9.241413804601535e-05,
|
| 163 |
+
"loss": 6.949,
|
| 164 |
+
"step": 460
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.24,
|
| 168 |
+
"learning_rate": 9.208069356452151e-05,
|
| 169 |
+
"loss": 6.9056,
|
| 170 |
+
"step": 480
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.25,
|
| 174 |
+
"learning_rate": 9.174724908302767e-05,
|
| 175 |
+
"loss": 6.8714,
|
| 176 |
+
"step": 500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.26,
|
| 180 |
+
"learning_rate": 9.141380460153385e-05,
|
| 181 |
+
"loss": 6.8231,
|
| 182 |
+
"step": 520
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.27,
|
| 186 |
+
"learning_rate": 9.108036012004002e-05,
|
| 187 |
+
"loss": 6.8117,
|
| 188 |
+
"step": 540
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.28,
|
| 192 |
+
"learning_rate": 9.074691563854619e-05,
|
| 193 |
+
"loss": 6.7348,
|
| 194 |
+
"step": 560
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.29,
|
| 198 |
+
"learning_rate": 9.041347115705235e-05,
|
| 199 |
+
"loss": 6.7406,
|
| 200 |
+
"step": 580
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.3,
|
| 204 |
+
"learning_rate": 9.008002667555852e-05,
|
| 205 |
+
"loss": 6.7483,
|
| 206 |
+
"step": 600
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.3,
|
| 210 |
+
"eval_accuracy": 0.21896009610790168,
|
| 211 |
+
"eval_loss": 6.79296875,
|
| 212 |
+
"eval_runtime": 632.1671,
|
| 213 |
+
"eval_samples_per_second": 16.326,
|
| 214 |
+
"eval_steps_per_second": 0.256,
|
| 215 |
+
"step": 600
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.31,
|
| 219 |
+
"learning_rate": 8.974658219406469e-05,
|
| 220 |
+
"loss": 6.6812,
|
| 221 |
+
"step": 620
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.32,
|
| 225 |
+
"learning_rate": 8.941313771257086e-05,
|
| 226 |
+
"loss": 6.6846,
|
| 227 |
+
"step": 640
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.33,
|
| 231 |
+
"learning_rate": 8.907969323107703e-05,
|
| 232 |
+
"loss": 6.711,
|
| 233 |
+
"step": 660
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.34,
|
| 237 |
+
"learning_rate": 8.87462487495832e-05,
|
| 238 |
+
"loss": 6.6342,
|
| 239 |
+
"step": 680
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.35,
|
| 243 |
+
"learning_rate": 8.841280426808936e-05,
|
| 244 |
+
"loss": 6.6208,
|
| 245 |
+
"step": 700
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.36,
|
| 249 |
+
"learning_rate": 8.807935978659554e-05,
|
| 250 |
+
"loss": 6.6218,
|
| 251 |
+
"step": 720
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.37,
|
| 255 |
+
"learning_rate": 8.77459153051017e-05,
|
| 256 |
+
"loss": 6.5341,
|
| 257 |
+
"step": 740
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.38,
|
| 261 |
+
"learning_rate": 8.741247082360787e-05,
|
| 262 |
+
"loss": 6.5627,
|
| 263 |
+
"step": 760
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.39,
|
| 267 |
+
"learning_rate": 8.707902634211405e-05,
|
| 268 |
+
"loss": 6.4917,
|
| 269 |
+
"step": 780
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4,
|
| 273 |
+
"learning_rate": 8.67455818606202e-05,
|
| 274 |
+
"loss": 6.4568,
|
| 275 |
+
"step": 800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4,
|
| 279 |
+
"eval_accuracy": 0.23763536355011933,
|
| 280 |
+
"eval_loss": 6.5703125,
|
| 281 |
+
"eval_runtime": 909.8761,
|
| 282 |
+
"eval_samples_per_second": 11.343,
|
| 283 |
+
"eval_steps_per_second": 0.178,
|
| 284 |
+
"step": 800
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.41,
|
| 288 |
+
"learning_rate": 8.641213737912638e-05,
|
| 289 |
+
"loss": 6.4541,
|
| 290 |
+
"step": 820
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.42,
|
| 294 |
+
"learning_rate": 8.607869289763255e-05,
|
| 295 |
+
"loss": 6.4661,
|
| 296 |
+
"step": 840
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.43,
|
| 300 |
+
"learning_rate": 8.57452484161387e-05,
|
| 301 |
+
"loss": 6.445,
|
| 302 |
+
"step": 860
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.44,
|
| 306 |
+
"learning_rate": 8.541180393464489e-05,
|
| 307 |
+
"loss": 6.4083,
|
| 308 |
+
"step": 880
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.45,
|
| 312 |
+
"learning_rate": 8.507835945315106e-05,
|
| 313 |
+
"loss": 6.4168,
|
| 314 |
+
"step": 900
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.46,
|
| 318 |
+
"learning_rate": 8.474491497165722e-05,
|
| 319 |
+
"loss": 6.3794,
|
| 320 |
+
"step": 920
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.47,
|
| 324 |
+
"learning_rate": 8.441147049016339e-05,
|
| 325 |
+
"loss": 6.3398,
|
| 326 |
+
"step": 940
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.48,
|
| 330 |
+
"learning_rate": 8.407802600866956e-05,
|
| 331 |
+
"loss": 6.3377,
|
| 332 |
+
"step": 960
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.49,
|
| 336 |
+
"learning_rate": 8.374458152717573e-05,
|
| 337 |
+
"loss": 6.3086,
|
| 338 |
+
"step": 980
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.5,
|
| 342 |
+
"learning_rate": 8.34111370456819e-05,
|
| 343 |
+
"loss": 6.2865,
|
| 344 |
+
"step": 1000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.5,
|
| 348 |
+
"eval_accuracy": 0.25518331041094305,
|
| 349 |
+
"eval_loss": 6.375,
|
| 350 |
+
"eval_runtime": 974.2435,
|
| 351 |
+
"eval_samples_per_second": 10.594,
|
| 352 |
+
"eval_steps_per_second": 0.166,
|
| 353 |
+
"step": 1000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.51,
|
| 357 |
+
"learning_rate": 8.311103701233745e-05,
|
| 358 |
+
"loss": 6.3177,
|
| 359 |
+
"step": 1020
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.52,
|
| 363 |
+
"learning_rate": 8.277759253084362e-05,
|
| 364 |
+
"loss": 6.2473,
|
| 365 |
+
"step": 1040
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.53,
|
| 369 |
+
"learning_rate": 8.244414804934978e-05,
|
| 370 |
+
"loss": 6.2194,
|
| 371 |
+
"step": 1060
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 0.54,
|
| 375 |
+
"learning_rate": 8.211070356785595e-05,
|
| 376 |
+
"loss": 6.2438,
|
| 377 |
+
"step": 1080
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.55,
|
| 381 |
+
"learning_rate": 8.177725908636212e-05,
|
| 382 |
+
"loss": 6.2338,
|
| 383 |
+
"step": 1100
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.56,
|
| 387 |
+
"learning_rate": 8.14438146048683e-05,
|
| 388 |
+
"loss": 6.1888,
|
| 389 |
+
"step": 1120
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.57,
|
| 393 |
+
"learning_rate": 8.111037012337445e-05,
|
| 394 |
+
"loss": 6.1476,
|
| 395 |
+
"step": 1140
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.58,
|
| 399 |
+
"learning_rate": 8.077692564188064e-05,
|
| 400 |
+
"loss": 6.1139,
|
| 401 |
+
"step": 1160
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.59,
|
| 405 |
+
"learning_rate": 8.04434811603868e-05,
|
| 406 |
+
"loss": 6.1182,
|
| 407 |
+
"step": 1180
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.6,
|
| 411 |
+
"learning_rate": 8.011003667889296e-05,
|
| 412 |
+
"loss": 6.1028,
|
| 413 |
+
"step": 1200
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.6,
|
| 417 |
+
"eval_accuracy": 0.27931541153780454,
|
| 418 |
+
"eval_loss": 6.1484375,
|
| 419 |
+
"eval_runtime": 865.8054,
|
| 420 |
+
"eval_samples_per_second": 11.921,
|
| 421 |
+
"eval_steps_per_second": 0.187,
|
| 422 |
+
"step": 1200
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.61,
|
| 426 |
+
"learning_rate": 7.977659219739914e-05,
|
| 427 |
+
"loss": 6.0742,
|
| 428 |
+
"step": 1220
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.62,
|
| 432 |
+
"learning_rate": 7.944314771590531e-05,
|
| 433 |
+
"loss": 6.0706,
|
| 434 |
+
"step": 1240
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.63,
|
| 438 |
+
"learning_rate": 7.910970323441148e-05,
|
| 439 |
+
"loss": 6.0479,
|
| 440 |
+
"step": 1260
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.64,
|
| 444 |
+
"learning_rate": 7.877625875291764e-05,
|
| 445 |
+
"loss": 5.9765,
|
| 446 |
+
"step": 1280
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.65,
|
| 450 |
+
"learning_rate": 7.844281427142381e-05,
|
| 451 |
+
"loss": 5.9953,
|
| 452 |
+
"step": 1300
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.66,
|
| 456 |
+
"learning_rate": 7.810936978992998e-05,
|
| 457 |
+
"loss": 6.0057,
|
| 458 |
+
"step": 1320
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.67,
|
| 462 |
+
"learning_rate": 7.777592530843615e-05,
|
| 463 |
+
"loss": 6.0033,
|
| 464 |
+
"step": 1340
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.68,
|
| 468 |
+
"learning_rate": 7.744248082694232e-05,
|
| 469 |
+
"loss": 5.9667,
|
| 470 |
+
"step": 1360
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.69,
|
| 474 |
+
"learning_rate": 7.710903634544848e-05,
|
| 475 |
+
"loss": 5.9367,
|
| 476 |
+
"step": 1380
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.7,
|
| 480 |
+
"learning_rate": 7.677559186395465e-05,
|
| 481 |
+
"loss": 5.8888,
|
| 482 |
+
"step": 1400
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.7,
|
| 486 |
+
"eval_accuracy": 0.29820689430587377,
|
| 487 |
+
"eval_loss": 5.95703125,
|
| 488 |
+
"eval_runtime": 778.5891,
|
| 489 |
+
"eval_samples_per_second": 13.256,
|
| 490 |
+
"eval_steps_per_second": 0.208,
|
| 491 |
+
"step": 1400
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.71,
|
| 495 |
+
"learning_rate": 7.644214738246082e-05,
|
| 496 |
+
"loss": 5.8829,
|
| 497 |
+
"step": 1420
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.72,
|
| 501 |
+
"learning_rate": 7.610870290096699e-05,
|
| 502 |
+
"loss": 5.8627,
|
| 503 |
+
"step": 1440
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.73,
|
| 507 |
+
"learning_rate": 7.577525841947316e-05,
|
| 508 |
+
"loss": 5.8251,
|
| 509 |
+
"step": 1460
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 0.74,
|
| 513 |
+
"learning_rate": 7.544181393797934e-05,
|
| 514 |
+
"loss": 5.8655,
|
| 515 |
+
"step": 1480
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.75,
|
| 519 |
+
"learning_rate": 7.510836945648549e-05,
|
| 520 |
+
"loss": 5.8127,
|
| 521 |
+
"step": 1500
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.76,
|
| 525 |
+
"learning_rate": 7.477492497499166e-05,
|
| 526 |
+
"loss": 5.8305,
|
| 527 |
+
"step": 1520
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.77,
|
| 531 |
+
"learning_rate": 7.444148049349784e-05,
|
| 532 |
+
"loss": 5.8159,
|
| 533 |
+
"step": 1540
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.78,
|
| 537 |
+
"learning_rate": 7.4108036012004e-05,
|
| 538 |
+
"loss": 5.8354,
|
| 539 |
+
"step": 1560
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.79,
|
| 543 |
+
"learning_rate": 7.377459153051018e-05,
|
| 544 |
+
"loss": 5.7823,
|
| 545 |
+
"step": 1580
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.8,
|
| 549 |
+
"learning_rate": 7.344114704901635e-05,
|
| 550 |
+
"loss": 5.7362,
|
| 551 |
+
"step": 1600
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.8,
|
| 555 |
+
"eval_accuracy": 0.3121272705508336,
|
| 556 |
+
"eval_loss": 5.80078125,
|
| 557 |
+
"eval_runtime": 525.7065,
|
| 558 |
+
"eval_samples_per_second": 19.633,
|
| 559 |
+
"eval_steps_per_second": 0.308,
|
| 560 |
+
"step": 1600
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.81,
|
| 564 |
+
"learning_rate": 7.310770256752251e-05,
|
| 565 |
+
"loss": 5.7201,
|
| 566 |
+
"step": 1620
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.82,
|
| 570 |
+
"learning_rate": 7.277425808602868e-05,
|
| 571 |
+
"loss": 5.7253,
|
| 572 |
+
"step": 1640
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.83,
|
| 576 |
+
"learning_rate": 7.244081360453485e-05,
|
| 577 |
+
"loss": 5.7258,
|
| 578 |
+
"step": 1660
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.84,
|
| 582 |
+
"learning_rate": 7.210736912304102e-05,
|
| 583 |
+
"loss": 5.6782,
|
| 584 |
+
"step": 1680
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.85,
|
| 588 |
+
"learning_rate": 7.177392464154719e-05,
|
| 589 |
+
"loss": 5.7047,
|
| 590 |
+
"step": 1700
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.86,
|
| 594 |
+
"learning_rate": 7.144048016005335e-05,
|
| 595 |
+
"loss": 5.6893,
|
| 596 |
+
"step": 1720
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.87,
|
| 600 |
+
"learning_rate": 7.110703567855952e-05,
|
| 601 |
+
"loss": 5.6119,
|
| 602 |
+
"step": 1740
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.88,
|
| 606 |
+
"learning_rate": 7.077359119706569e-05,
|
| 607 |
+
"loss": 5.6852,
|
| 608 |
+
"step": 1760
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.89,
|
| 612 |
+
"learning_rate": 7.044014671557186e-05,
|
| 613 |
+
"loss": 5.6021,
|
| 614 |
+
"step": 1780
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.9,
|
| 618 |
+
"learning_rate": 7.010670223407803e-05,
|
| 619 |
+
"loss": 5.6507,
|
| 620 |
+
"step": 1800
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.9,
|
| 624 |
+
"eval_accuracy": 0.32381624594057856,
|
| 625 |
+
"eval_loss": 5.6796875,
|
| 626 |
+
"eval_runtime": 634.7933,
|
| 627 |
+
"eval_samples_per_second": 16.259,
|
| 628 |
+
"eval_steps_per_second": 0.255,
|
| 629 |
+
"step": 1800
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.91,
|
| 633 |
+
"learning_rate": 6.97732577525842e-05,
|
| 634 |
+
"loss": 5.6308,
|
| 635 |
+
"step": 1820
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.92,
|
| 639 |
+
"learning_rate": 6.943981327109038e-05,
|
| 640 |
+
"loss": 5.6368,
|
| 641 |
+
"step": 1840
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.93,
|
| 645 |
+
"learning_rate": 6.910636878959653e-05,
|
| 646 |
+
"loss": 5.6018,
|
| 647 |
+
"step": 1860
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.94,
|
| 651 |
+
"learning_rate": 6.87729243081027e-05,
|
| 652 |
+
"loss": 5.5617,
|
| 653 |
+
"step": 1880
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.95,
|
| 657 |
+
"learning_rate": 6.843947982660888e-05,
|
| 658 |
+
"loss": 5.5805,
|
| 659 |
+
"step": 1900
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.96,
|
| 663 |
+
"learning_rate": 6.810603534511503e-05,
|
| 664 |
+
"loss": 5.5503,
|
| 665 |
+
"step": 1920
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.97,
|
| 669 |
+
"learning_rate": 6.777259086362121e-05,
|
| 670 |
+
"loss": 5.555,
|
| 671 |
+
"step": 1940
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 0.98,
|
| 675 |
+
"learning_rate": 6.743914638212738e-05,
|
| 676 |
+
"loss": 5.5774,
|
| 677 |
+
"step": 1960
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.99,
|
| 681 |
+
"learning_rate": 6.710570190063354e-05,
|
| 682 |
+
"loss": 5.5667,
|
| 683 |
+
"step": 1980
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.0,
|
| 687 |
+
"learning_rate": 6.677225741913972e-05,
|
| 688 |
+
"loss": 5.565,
|
| 689 |
+
"step": 2000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.0,
|
| 693 |
+
"eval_accuracy": 0.33180064057103564,
|
| 694 |
+
"eval_loss": 5.578125,
|
| 695 |
+
"eval_runtime": 846.4215,
|
| 696 |
+
"eval_samples_per_second": 12.194,
|
| 697 |
+
"eval_steps_per_second": 0.191,
|
| 698 |
+
"step": 2000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.01,
|
| 702 |
+
"learning_rate": 6.647215738579527e-05,
|
| 703 |
+
"loss": 5.5107,
|
| 704 |
+
"step": 2020
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.02,
|
| 708 |
+
"learning_rate": 6.613871290430144e-05,
|
| 709 |
+
"loss": 5.5002,
|
| 710 |
+
"step": 2040
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.03,
|
| 714 |
+
"learning_rate": 6.58052684228076e-05,
|
| 715 |
+
"loss": 5.4849,
|
| 716 |
+
"step": 2060
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.04,
|
| 720 |
+
"learning_rate": 6.547182394131377e-05,
|
| 721 |
+
"loss": 5.5153,
|
| 722 |
+
"step": 2080
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.05,
|
| 726 |
+
"learning_rate": 6.513837945981994e-05,
|
| 727 |
+
"loss": 5.4517,
|
| 728 |
+
"step": 2100
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 1.06,
|
| 732 |
+
"learning_rate": 6.480493497832611e-05,
|
| 733 |
+
"loss": 5.4576,
|
| 734 |
+
"step": 2120
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.07,
|
| 738 |
+
"learning_rate": 6.447149049683228e-05,
|
| 739 |
+
"loss": 5.454,
|
| 740 |
+
"step": 2140
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.08,
|
| 744 |
+
"learning_rate": 6.413804601533845e-05,
|
| 745 |
+
"loss": 5.432,
|
| 746 |
+
"step": 2160
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.09,
|
| 750 |
+
"learning_rate": 6.380460153384463e-05,
|
| 751 |
+
"loss": 5.4429,
|
| 752 |
+
"step": 2180
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.1,
|
| 756 |
+
"learning_rate": 6.347115705235078e-05,
|
| 757 |
+
"loss": 5.4688,
|
| 758 |
+
"step": 2200
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.1,
|
| 762 |
+
"eval_accuracy": 0.33922944511943365,
|
| 763 |
+
"eval_loss": 5.49609375,
|
| 764 |
+
"eval_runtime": 969.4851,
|
| 765 |
+
"eval_samples_per_second": 10.646,
|
| 766 |
+
"eval_steps_per_second": 0.167,
|
| 767 |
+
"step": 2200
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.11,
|
| 771 |
+
"learning_rate": 6.313771257085695e-05,
|
| 772 |
+
"loss": 5.4398,
|
| 773 |
+
"step": 2220
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.12,
|
| 777 |
+
"learning_rate": 6.280426808936313e-05,
|
| 778 |
+
"loss": 5.3972,
|
| 779 |
+
"step": 2240
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.13,
|
| 783 |
+
"learning_rate": 6.247082360786929e-05,
|
| 784 |
+
"loss": 5.4121,
|
| 785 |
+
"step": 2260
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.14,
|
| 789 |
+
"learning_rate": 6.213737912637547e-05,
|
| 790 |
+
"loss": 5.4003,
|
| 791 |
+
"step": 2280
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 1.15,
|
| 795 |
+
"learning_rate": 6.180393464488164e-05,
|
| 796 |
+
"loss": 5.3916,
|
| 797 |
+
"step": 2300
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.16,
|
| 801 |
+
"learning_rate": 6.147049016338779e-05,
|
| 802 |
+
"loss": 5.3822,
|
| 803 |
+
"step": 2320
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 1.17,
|
| 807 |
+
"learning_rate": 6.113704568189397e-05,
|
| 808 |
+
"loss": 5.419,
|
| 809 |
+
"step": 2340
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.18,
|
| 813 |
+
"learning_rate": 6.080360120040014e-05,
|
| 814 |
+
"loss": 5.3861,
|
| 815 |
+
"step": 2360
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.19,
|
| 819 |
+
"learning_rate": 6.04701567189063e-05,
|
| 820 |
+
"loss": 5.4055,
|
| 821 |
+
"step": 2380
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.2,
|
| 825 |
+
"learning_rate": 6.0136712237412475e-05,
|
| 826 |
+
"loss": 5.4044,
|
| 827 |
+
"step": 2400
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.2,
|
| 831 |
+
"eval_accuracy": 0.3456189203935214,
|
| 832 |
+
"eval_loss": 5.421875,
|
| 833 |
+
"eval_runtime": 894.657,
|
| 834 |
+
"eval_samples_per_second": 11.536,
|
| 835 |
+
"eval_steps_per_second": 0.181,
|
| 836 |
+
"step": 2400
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.21,
|
| 840 |
+
"learning_rate": 5.980326775591864e-05,
|
| 841 |
+
"loss": 5.3611,
|
| 842 |
+
"step": 2420
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.22,
|
| 846 |
+
"learning_rate": 5.9469823274424805e-05,
|
| 847 |
+
"loss": 5.2802,
|
| 848 |
+
"step": 2440
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.23,
|
| 852 |
+
"learning_rate": 5.913637879293098e-05,
|
| 853 |
+
"loss": 5.3635,
|
| 854 |
+
"step": 2460
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.24,
|
| 858 |
+
"learning_rate": 5.8802934311437154e-05,
|
| 859 |
+
"loss": 5.3047,
|
| 860 |
+
"step": 2480
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.25,
|
| 864 |
+
"learning_rate": 5.8469489829943315e-05,
|
| 865 |
+
"loss": 5.3242,
|
| 866 |
+
"step": 2500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.26,
|
| 870 |
+
"learning_rate": 5.813604534844948e-05,
|
| 871 |
+
"loss": 5.3219,
|
| 872 |
+
"step": 2520
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.27,
|
| 876 |
+
"learning_rate": 5.780260086695566e-05,
|
| 877 |
+
"loss": 5.3164,
|
| 878 |
+
"step": 2540
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.28,
|
| 882 |
+
"learning_rate": 5.746915638546182e-05,
|
| 883 |
+
"loss": 5.3344,
|
| 884 |
+
"step": 2560
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.29,
|
| 888 |
+
"learning_rate": 5.7135711903967994e-05,
|
| 889 |
+
"loss": 5.2985,
|
| 890 |
+
"step": 2580
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.3,
|
| 894 |
+
"learning_rate": 5.680226742247416e-05,
|
| 895 |
+
"loss": 5.3323,
|
| 896 |
+
"step": 2600
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 1.3,
|
| 900 |
+
"eval_accuracy": 0.3515776689895772,
|
| 901 |
+
"eval_loss": 5.359375,
|
| 902 |
+
"eval_runtime": 915.4215,
|
| 903 |
+
"eval_samples_per_second": 11.275,
|
| 904 |
+
"eval_steps_per_second": 0.177,
|
| 905 |
+
"step": 2600
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.31,
|
| 909 |
+
"learning_rate": 5.646882294098032e-05,
|
| 910 |
+
"loss": 5.3498,
|
| 911 |
+
"step": 2620
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.32,
|
| 915 |
+
"learning_rate": 5.61353784594865e-05,
|
| 916 |
+
"loss": 5.3374,
|
| 917 |
+
"step": 2640
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.33,
|
| 921 |
+
"learning_rate": 5.580193397799267e-05,
|
| 922 |
+
"loss": 5.258,
|
| 923 |
+
"step": 2660
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 1.34,
|
| 927 |
+
"learning_rate": 5.5468489496498834e-05,
|
| 928 |
+
"loss": 5.3057,
|
| 929 |
+
"step": 2680
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.35,
|
| 933 |
+
"learning_rate": 5.5135045015005e-05,
|
| 934 |
+
"loss": 5.3096,
|
| 935 |
+
"step": 2700
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 1.36,
|
| 939 |
+
"learning_rate": 5.480160053351118e-05,
|
| 940 |
+
"loss": 5.3103,
|
| 941 |
+
"step": 2720
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.37,
|
| 945 |
+
"learning_rate": 5.446815605201734e-05,
|
| 946 |
+
"loss": 5.294,
|
| 947 |
+
"step": 2740
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.38,
|
| 951 |
+
"learning_rate": 5.413471157052351e-05,
|
| 952 |
+
"loss": 5.2703,
|
| 953 |
+
"step": 2760
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 1.39,
|
| 957 |
+
"learning_rate": 5.380126708902968e-05,
|
| 958 |
+
"loss": 5.2838,
|
| 959 |
+
"step": 2780
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 1.4,
|
| 963 |
+
"learning_rate": 5.346782260753584e-05,
|
| 964 |
+
"loss": 5.2598,
|
| 965 |
+
"step": 2800
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 1.4,
|
| 969 |
+
"eval_accuracy": 0.3561769305915198,
|
| 970 |
+
"eval_loss": 5.3046875,
|
| 971 |
+
"eval_runtime": 850.0259,
|
| 972 |
+
"eval_samples_per_second": 12.142,
|
| 973 |
+
"eval_steps_per_second": 0.191,
|
| 974 |
+
"step": 2800
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.41,
|
| 978 |
+
"learning_rate": 5.3134378126042017e-05,
|
| 979 |
+
"loss": 5.2614,
|
| 980 |
+
"step": 2820
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 1.42,
|
| 984 |
+
"learning_rate": 5.280093364454819e-05,
|
| 985 |
+
"loss": 5.2358,
|
| 986 |
+
"step": 2840
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 1.43,
|
| 990 |
+
"learning_rate": 5.246748916305435e-05,
|
| 991 |
+
"loss": 5.2465,
|
| 992 |
+
"step": 2860
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.44,
|
| 996 |
+
"learning_rate": 5.213404468156052e-05,
|
| 997 |
+
"loss": 5.2342,
|
| 998 |
+
"step": 2880
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.45,
|
| 1002 |
+
"learning_rate": 5.1800600200066695e-05,
|
| 1003 |
+
"loss": 5.2434,
|
| 1004 |
+
"step": 2900
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.46,
|
| 1008 |
+
"learning_rate": 5.1467155718572856e-05,
|
| 1009 |
+
"loss": 5.2131,
|
| 1010 |
+
"step": 2920
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.47,
|
| 1014 |
+
"learning_rate": 5.113371123707903e-05,
|
| 1015 |
+
"loss": 5.2167,
|
| 1016 |
+
"step": 2940
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.48,
|
| 1020 |
+
"learning_rate": 5.08002667555852e-05,
|
| 1021 |
+
"loss": 5.1727,
|
| 1022 |
+
"step": 2960
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 1.49,
|
| 1026 |
+
"learning_rate": 5.046682227409136e-05,
|
| 1027 |
+
"loss": 5.2107,
|
| 1028 |
+
"step": 2980
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 1.5,
|
| 1032 |
+
"learning_rate": 5.0133377792597535e-05,
|
| 1033 |
+
"loss": 5.2159,
|
| 1034 |
+
"step": 3000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.5,
|
| 1038 |
+
"eval_accuracy": 0.3596428603716168,
|
| 1039 |
+
"eval_loss": 5.2578125,
|
| 1040 |
+
"eval_runtime": 533.7663,
|
| 1041 |
+
"eval_samples_per_second": 19.336,
|
| 1042 |
+
"eval_steps_per_second": 0.304,
|
| 1043 |
+
"step": 3000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 1.51,
|
| 1047 |
+
"learning_rate": 4.9833277759253086e-05,
|
| 1048 |
+
"loss": 5.2447,
|
| 1049 |
+
"step": 3020
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 1.52,
|
| 1053 |
+
"learning_rate": 4.9499833277759254e-05,
|
| 1054 |
+
"loss": 5.2568,
|
| 1055 |
+
"step": 3040
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.53,
|
| 1059 |
+
"learning_rate": 4.916638879626542e-05,
|
| 1060 |
+
"loss": 5.1794,
|
| 1061 |
+
"step": 3060
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.54,
|
| 1065 |
+
"learning_rate": 4.88329443147716e-05,
|
| 1066 |
+
"loss": 5.1844,
|
| 1067 |
+
"step": 3080
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.55,
|
| 1071 |
+
"learning_rate": 4.849949983327776e-05,
|
| 1072 |
+
"loss": 5.2218,
|
| 1073 |
+
"step": 3100
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.56,
|
| 1077 |
+
"learning_rate": 4.8166055351783926e-05,
|
| 1078 |
+
"loss": 5.2347,
|
| 1079 |
+
"step": 3120
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.57,
|
| 1083 |
+
"learning_rate": 4.78326108702901e-05,
|
| 1084 |
+
"loss": 5.1653,
|
| 1085 |
+
"step": 3140
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 1.58,
|
| 1089 |
+
"learning_rate": 4.749916638879627e-05,
|
| 1090 |
+
"loss": 5.2375,
|
| 1091 |
+
"step": 3160
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.59,
|
| 1095 |
+
"learning_rate": 4.716572190730244e-05,
|
| 1096 |
+
"loss": 5.19,
|
| 1097 |
+
"step": 3180
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.6,
|
| 1101 |
+
"learning_rate": 4.6832277425808605e-05,
|
| 1102 |
+
"loss": 5.1992,
|
| 1103 |
+
"step": 3200
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 1.6,
|
| 1107 |
+
"eval_accuracy": 0.36381432991685037,
|
| 1108 |
+
"eval_loss": 5.21484375,
|
| 1109 |
+
"eval_runtime": 1001.3018,
|
| 1110 |
+
"eval_samples_per_second": 10.308,
|
| 1111 |
+
"eval_steps_per_second": 0.162,
|
| 1112 |
+
"step": 3200
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.61,
|
| 1116 |
+
"learning_rate": 4.649883294431477e-05,
|
| 1117 |
+
"loss": 5.2072,
|
| 1118 |
+
"step": 3220
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.62,
|
| 1122 |
+
"learning_rate": 4.616538846282094e-05,
|
| 1123 |
+
"loss": 5.1528,
|
| 1124 |
+
"step": 3240
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.63,
|
| 1128 |
+
"learning_rate": 4.5831943981327116e-05,
|
| 1129 |
+
"loss": 5.1562,
|
| 1130 |
+
"step": 3260
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.64,
|
| 1134 |
+
"learning_rate": 4.549849949983328e-05,
|
| 1135 |
+
"loss": 5.1613,
|
| 1136 |
+
"step": 3280
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.65,
|
| 1140 |
+
"learning_rate": 4.5165055018339445e-05,
|
| 1141 |
+
"loss": 5.1552,
|
| 1142 |
+
"step": 3300
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.66,
|
| 1146 |
+
"learning_rate": 4.483161053684562e-05,
|
| 1147 |
+
"loss": 5.1224,
|
| 1148 |
+
"step": 3320
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 1.67,
|
| 1152 |
+
"learning_rate": 4.449816605535179e-05,
|
| 1153 |
+
"loss": 5.1118,
|
| 1154 |
+
"step": 3340
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.67,
|
| 1158 |
+
"learning_rate": 4.4164721573857956e-05,
|
| 1159 |
+
"loss": 5.138,
|
| 1160 |
+
"step": 3360
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.68,
|
| 1164 |
+
"learning_rate": 4.3831277092364124e-05,
|
| 1165 |
+
"loss": 5.1441,
|
| 1166 |
+
"step": 3380
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 1.69,
|
| 1170 |
+
"learning_rate": 4.349783261087029e-05,
|
| 1171 |
+
"loss": 5.1429,
|
| 1172 |
+
"step": 3400
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.69,
|
| 1176 |
+
"eval_accuracy": 0.36717262536004136,
|
| 1177 |
+
"eval_loss": 5.1796875,
|
| 1178 |
+
"eval_runtime": 930.4199,
|
| 1179 |
+
"eval_samples_per_second": 11.093,
|
| 1180 |
+
"eval_steps_per_second": 0.174,
|
| 1181 |
+
"step": 3400
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 1.7,
|
| 1185 |
+
"learning_rate": 4.316438812937646e-05,
|
| 1186 |
+
"loss": 5.1378,
|
| 1187 |
+
"step": 3420
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 1.71,
|
| 1191 |
+
"learning_rate": 4.2830943647882634e-05,
|
| 1192 |
+
"loss": 5.0948,
|
| 1193 |
+
"step": 3440
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.72,
|
| 1197 |
+
"learning_rate": 4.2497499166388796e-05,
|
| 1198 |
+
"loss": 5.108,
|
| 1199 |
+
"step": 3460
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.73,
|
| 1203 |
+
"learning_rate": 4.2164054684894964e-05,
|
| 1204 |
+
"loss": 5.0677,
|
| 1205 |
+
"step": 3480
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 1.74,
|
| 1209 |
+
"learning_rate": 4.183061020340114e-05,
|
| 1210 |
+
"loss": 5.1182,
|
| 1211 |
+
"step": 3500
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 1.75,
|
| 1215 |
+
"learning_rate": 4.1497165721907306e-05,
|
| 1216 |
+
"loss": 5.1199,
|
| 1217 |
+
"step": 3520
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.76,
|
| 1221 |
+
"learning_rate": 4.1163721240413474e-05,
|
| 1222 |
+
"loss": 5.1359,
|
| 1223 |
+
"step": 3540
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 1.77,
|
| 1227 |
+
"learning_rate": 4.083027675891964e-05,
|
| 1228 |
+
"loss": 5.1278,
|
| 1229 |
+
"step": 3560
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 1.78,
|
| 1233 |
+
"learning_rate": 4.049683227742581e-05,
|
| 1234 |
+
"loss": 5.1227,
|
| 1235 |
+
"step": 3580
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.79,
|
| 1239 |
+
"learning_rate": 4.016338779593198e-05,
|
| 1240 |
+
"loss": 5.095,
|
| 1241 |
+
"step": 3600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.79,
|
| 1245 |
+
"eval_accuracy": 0.3696338259978766,
|
| 1246 |
+
"eval_loss": 5.14453125,
|
| 1247 |
+
"eval_runtime": 540.5754,
|
| 1248 |
+
"eval_samples_per_second": 19.093,
|
| 1249 |
+
"eval_steps_per_second": 0.3,
|
| 1250 |
+
"step": 3600
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 1.8,
|
| 1254 |
+
"learning_rate": 3.982994331443815e-05,
|
| 1255 |
+
"loss": 5.1139,
|
| 1256 |
+
"step": 3620
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.81,
|
| 1260 |
+
"learning_rate": 3.9496498832944314e-05,
|
| 1261 |
+
"loss": 5.0827,
|
| 1262 |
+
"step": 3640
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.82,
|
| 1266 |
+
"learning_rate": 3.916305435145048e-05,
|
| 1267 |
+
"loss": 5.0918,
|
| 1268 |
+
"step": 3660
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 1.83,
|
| 1272 |
+
"learning_rate": 3.882960986995666e-05,
|
| 1273 |
+
"loss": 5.0927,
|
| 1274 |
+
"step": 3680
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.84,
|
| 1278 |
+
"learning_rate": 3.8496165388462825e-05,
|
| 1279 |
+
"loss": 5.0904,
|
| 1280 |
+
"step": 3700
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.85,
|
| 1284 |
+
"learning_rate": 3.816272090696899e-05,
|
| 1285 |
+
"loss": 5.0245,
|
| 1286 |
+
"step": 3720
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 1.86,
|
| 1290 |
+
"learning_rate": 3.7829276425475154e-05,
|
| 1291 |
+
"loss": 5.1174,
|
| 1292 |
+
"step": 3740
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 1.87,
|
| 1296 |
+
"learning_rate": 3.749583194398133e-05,
|
| 1297 |
+
"loss": 5.1325,
|
| 1298 |
+
"step": 3760
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.88,
|
| 1302 |
+
"learning_rate": 3.71623874624875e-05,
|
| 1303 |
+
"loss": 5.0408,
|
| 1304 |
+
"step": 3780
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.89,
|
| 1308 |
+
"learning_rate": 3.6828942980993665e-05,
|
| 1309 |
+
"loss": 5.0646,
|
| 1310 |
+
"step": 3800
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 1.89,
|
| 1314 |
+
"eval_accuracy": 0.37147085161338145,
|
| 1315 |
+
"eval_loss": 5.1171875,
|
| 1316 |
+
"eval_runtime": 1017.1553,
|
| 1317 |
+
"eval_samples_per_second": 10.147,
|
| 1318 |
+
"eval_steps_per_second": 0.159,
|
| 1319 |
+
"step": 3800
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.9,
|
| 1323 |
+
"learning_rate": 3.649549849949983e-05,
|
| 1324 |
+
"loss": 5.0465,
|
| 1325 |
+
"step": 3820
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.91,
|
| 1329 |
+
"learning_rate": 3.6162054018006e-05,
|
| 1330 |
+
"loss": 5.1008,
|
| 1331 |
+
"step": 3840
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 1.92,
|
| 1335 |
+
"learning_rate": 3.582860953651217e-05,
|
| 1336 |
+
"loss": 5.0641,
|
| 1337 |
+
"step": 3860
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 1.93,
|
| 1341 |
+
"learning_rate": 3.5495165055018344e-05,
|
| 1342 |
+
"loss": 5.085,
|
| 1343 |
+
"step": 3880
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.94,
|
| 1347 |
+
"learning_rate": 3.516172057352451e-05,
|
| 1348 |
+
"loss": 5.0498,
|
| 1349 |
+
"step": 3900
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.95,
|
| 1353 |
+
"learning_rate": 3.482827609203067e-05,
|
| 1354 |
+
"loss": 5.1401,
|
| 1355 |
+
"step": 3920
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.96,
|
| 1359 |
+
"learning_rate": 3.449483161053685e-05,
|
| 1360 |
+
"loss": 5.0446,
|
| 1361 |
+
"step": 3940
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.97,
|
| 1365 |
+
"learning_rate": 3.4161387129043016e-05,
|
| 1366 |
+
"loss": 5.0538,
|
| 1367 |
+
"step": 3960
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.98,
|
| 1371 |
+
"learning_rate": 3.3827942647549183e-05,
|
| 1372 |
+
"loss": 5.0641,
|
| 1373 |
+
"step": 3980
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.99,
|
| 1377 |
+
"learning_rate": 3.349449816605535e-05,
|
| 1378 |
+
"loss": 5.059,
|
| 1379 |
+
"step": 4000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 1.99,
|
| 1383 |
+
"eval_accuracy": 0.37421836716060286,
|
| 1384 |
+
"eval_loss": 5.0859375,
|
| 1385 |
+
"eval_runtime": 943.8518,
|
| 1386 |
+
"eval_samples_per_second": 10.935,
|
| 1387 |
+
"eval_steps_per_second": 0.172,
|
| 1388 |
+
"step": 4000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 2.0,
|
| 1392 |
+
"learning_rate": 3.319439813271091e-05,
|
| 1393 |
+
"loss": 5.0309,
|
| 1394 |
+
"step": 4020
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 2.01,
|
| 1398 |
+
"learning_rate": 3.286095365121708e-05,
|
| 1399 |
+
"loss": 5.0337,
|
| 1400 |
+
"step": 4040
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 2.02,
|
| 1404 |
+
"learning_rate": 3.252750916972324e-05,
|
| 1405 |
+
"loss": 5.0333,
|
| 1406 |
+
"step": 4060
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.03,
|
| 1410 |
+
"learning_rate": 3.2194064688229413e-05,
|
| 1411 |
+
"loss": 4.981,
|
| 1412 |
+
"step": 4080
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.04,
|
| 1416 |
+
"learning_rate": 3.186062020673558e-05,
|
| 1417 |
+
"loss": 5.0187,
|
| 1418 |
+
"step": 4100
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.05,
|
| 1422 |
+
"learning_rate": 3.152717572524175e-05,
|
| 1423 |
+
"loss": 4.9909,
|
| 1424 |
+
"step": 4120
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.06,
|
| 1428 |
+
"learning_rate": 3.119373124374792e-05,
|
| 1429 |
+
"loss": 4.9913,
|
| 1430 |
+
"step": 4140
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.07,
|
| 1434 |
+
"learning_rate": 3.0860286762254085e-05,
|
| 1435 |
+
"loss": 4.9999,
|
| 1436 |
+
"step": 4160
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 2.08,
|
| 1440 |
+
"learning_rate": 3.052684228076025e-05,
|
| 1441 |
+
"loss": 5.0122,
|
| 1442 |
+
"step": 4180
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.09,
|
| 1446 |
+
"learning_rate": 3.0193397799266425e-05,
|
| 1447 |
+
"loss": 5.0152,
|
| 1448 |
+
"step": 4200
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 2.09,
|
| 1452 |
+
"eval_accuracy": 0.375609093672024,
|
| 1453 |
+
"eval_loss": 5.06640625,
|
| 1454 |
+
"eval_runtime": 893.7828,
|
| 1455 |
+
"eval_samples_per_second": 11.548,
|
| 1456 |
+
"eval_steps_per_second": 0.181,
|
| 1457 |
+
"step": 4200
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.1,
|
| 1461 |
+
"learning_rate": 2.9859953317772593e-05,
|
| 1462 |
+
"loss": 4.9893,
|
| 1463 |
+
"step": 4220
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 2.11,
|
| 1467 |
+
"learning_rate": 2.952650883627876e-05,
|
| 1468 |
+
"loss": 4.9974,
|
| 1469 |
+
"step": 4240
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 2.12,
|
| 1473 |
+
"learning_rate": 2.919306435478493e-05,
|
| 1474 |
+
"loss": 5.0072,
|
| 1475 |
+
"step": 4260
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 2.13,
|
| 1479 |
+
"learning_rate": 2.88596198732911e-05,
|
| 1480 |
+
"loss": 4.9678,
|
| 1481 |
+
"step": 4280
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 2.14,
|
| 1485 |
+
"learning_rate": 2.8526175391797265e-05,
|
| 1486 |
+
"loss": 4.9532,
|
| 1487 |
+
"step": 4300
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.15,
|
| 1491 |
+
"learning_rate": 2.8192730910303433e-05,
|
| 1492 |
+
"loss": 4.9994,
|
| 1493 |
+
"step": 4320
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 2.16,
|
| 1497 |
+
"learning_rate": 2.7859286428809604e-05,
|
| 1498 |
+
"loss": 4.9984,
|
| 1499 |
+
"step": 4340
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 2.17,
|
| 1503 |
+
"learning_rate": 2.7525841947315772e-05,
|
| 1504 |
+
"loss": 4.9659,
|
| 1505 |
+
"step": 4360
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 2.18,
|
| 1509 |
+
"learning_rate": 2.719239746582194e-05,
|
| 1510 |
+
"loss": 4.9355,
|
| 1511 |
+
"step": 4380
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 2.19,
|
| 1515 |
+
"learning_rate": 2.685895298432811e-05,
|
| 1516 |
+
"loss": 5.0251,
|
| 1517 |
+
"step": 4400
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 2.19,
|
| 1521 |
+
"eval_accuracy": 0.376910266900496,
|
| 1522 |
+
"eval_loss": 5.046875,
|
| 1523 |
+
"eval_runtime": 856.9216,
|
| 1524 |
+
"eval_samples_per_second": 12.044,
|
| 1525 |
+
"eval_steps_per_second": 0.189,
|
| 1526 |
+
"step": 4400
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 2.2,
|
| 1530 |
+
"learning_rate": 2.652550850283428e-05,
|
| 1531 |
+
"loss": 5.0034,
|
| 1532 |
+
"step": 4420
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 2.21,
|
| 1536 |
+
"learning_rate": 2.6192064021340444e-05,
|
| 1537 |
+
"loss": 4.9666,
|
| 1538 |
+
"step": 4440
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.22,
|
| 1542 |
+
"learning_rate": 2.585861953984662e-05,
|
| 1543 |
+
"loss": 4.9759,
|
| 1544 |
+
"step": 4460
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.23,
|
| 1548 |
+
"learning_rate": 2.5525175058352783e-05,
|
| 1549 |
+
"loss": 5.0003,
|
| 1550 |
+
"step": 4480
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.24,
|
| 1554 |
+
"learning_rate": 2.519173057685895e-05,
|
| 1555 |
+
"loss": 4.9566,
|
| 1556 |
+
"step": 4500
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.25,
|
| 1560 |
+
"learning_rate": 2.4858286095365123e-05,
|
| 1561 |
+
"loss": 4.9922,
|
| 1562 |
+
"step": 4520
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 2.26,
|
| 1566 |
+
"learning_rate": 2.452484161387129e-05,
|
| 1567 |
+
"loss": 4.9977,
|
| 1568 |
+
"step": 4540
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.27,
|
| 1572 |
+
"learning_rate": 2.4191397132377462e-05,
|
| 1573 |
+
"loss": 4.978,
|
| 1574 |
+
"step": 4560
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.28,
|
| 1578 |
+
"learning_rate": 2.3857952650883627e-05,
|
| 1579 |
+
"loss": 4.9841,
|
| 1580 |
+
"step": 4580
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.29,
|
| 1584 |
+
"learning_rate": 2.3524508169389798e-05,
|
| 1585 |
+
"loss": 5.022,
|
| 1586 |
+
"step": 4600
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.29,
|
| 1590 |
+
"eval_accuracy": 0.37901850832535505,
|
| 1591 |
+
"eval_loss": 5.02734375,
|
| 1592 |
+
"eval_runtime": 895.8024,
|
| 1593 |
+
"eval_samples_per_second": 11.522,
|
| 1594 |
+
"eval_steps_per_second": 0.181,
|
| 1595 |
+
"step": 4600
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 2.3,
|
| 1599 |
+
"learning_rate": 2.3191063687895966e-05,
|
| 1600 |
+
"loss": 4.965,
|
| 1601 |
+
"step": 4620
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 2.31,
|
| 1605 |
+
"learning_rate": 2.2857619206402134e-05,
|
| 1606 |
+
"loss": 4.9553,
|
| 1607 |
+
"step": 4640
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.32,
|
| 1611 |
+
"learning_rate": 2.2524174724908302e-05,
|
| 1612 |
+
"loss": 4.9534,
|
| 1613 |
+
"step": 4660
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.33,
|
| 1617 |
+
"learning_rate": 2.2190730243414473e-05,
|
| 1618 |
+
"loss": 4.9518,
|
| 1619 |
+
"step": 4680
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.34,
|
| 1623 |
+
"learning_rate": 2.185728576192064e-05,
|
| 1624 |
+
"loss": 4.9339,
|
| 1625 |
+
"step": 4700
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 2.35,
|
| 1629 |
+
"learning_rate": 2.152384128042681e-05,
|
| 1630 |
+
"loss": 4.9811,
|
| 1631 |
+
"step": 4720
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 2.36,
|
| 1635 |
+
"learning_rate": 2.119039679893298e-05,
|
| 1636 |
+
"loss": 4.9615,
|
| 1637 |
+
"step": 4740
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 2.37,
|
| 1641 |
+
"learning_rate": 2.0856952317439145e-05,
|
| 1642 |
+
"loss": 4.9562,
|
| 1643 |
+
"step": 4760
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 2.38,
|
| 1647 |
+
"learning_rate": 2.0523507835945317e-05,
|
| 1648 |
+
"loss": 4.9349,
|
| 1649 |
+
"step": 4780
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 2.39,
|
| 1653 |
+
"learning_rate": 2.0190063354451485e-05,
|
| 1654 |
+
"loss": 4.9939,
|
| 1655 |
+
"step": 4800
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.39,
|
| 1659 |
+
"eval_accuracy": 0.37982254723521514,
|
| 1660 |
+
"eval_loss": 5.015625,
|
| 1661 |
+
"eval_runtime": 530.0593,
|
| 1662 |
+
"eval_samples_per_second": 19.471,
|
| 1663 |
+
"eval_steps_per_second": 0.306,
|
| 1664 |
+
"step": 4800
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.4,
|
| 1668 |
+
"learning_rate": 1.9856618872957653e-05,
|
| 1669 |
+
"loss": 4.9611,
|
| 1670 |
+
"step": 4820
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.41,
|
| 1674 |
+
"learning_rate": 1.952317439146382e-05,
|
| 1675 |
+
"loss": 4.914,
|
| 1676 |
+
"step": 4840
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.42,
|
| 1680 |
+
"learning_rate": 1.9189729909969992e-05,
|
| 1681 |
+
"loss": 4.9286,
|
| 1682 |
+
"step": 4860
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.43,
|
| 1686 |
+
"learning_rate": 1.885628542847616e-05,
|
| 1687 |
+
"loss": 4.9763,
|
| 1688 |
+
"step": 4880
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 2.44,
|
| 1692 |
+
"learning_rate": 1.8522840946982328e-05,
|
| 1693 |
+
"loss": 4.947,
|
| 1694 |
+
"step": 4900
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 2.45,
|
| 1698 |
+
"learning_rate": 1.81893964654885e-05,
|
| 1699 |
+
"loss": 4.8647,
|
| 1700 |
+
"step": 4920
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 2.46,
|
| 1704 |
+
"learning_rate": 1.7855951983994664e-05,
|
| 1705 |
+
"loss": 4.9485,
|
| 1706 |
+
"step": 4940
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 2.47,
|
| 1710 |
+
"learning_rate": 1.7522507502500835e-05,
|
| 1711 |
+
"loss": 4.914,
|
| 1712 |
+
"step": 4960
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 2.48,
|
| 1716 |
+
"learning_rate": 1.7189063021007003e-05,
|
| 1717 |
+
"loss": 4.9502,
|
| 1718 |
+
"step": 4980
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.49,
|
| 1722 |
+
"learning_rate": 1.685561853951317e-05,
|
| 1723 |
+
"loss": 4.924,
|
| 1724 |
+
"step": 5000
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.49,
|
| 1728 |
+
"eval_accuracy": 0.381116573240788,
|
| 1729 |
+
"eval_loss": 5.0,
|
| 1730 |
+
"eval_runtime": 626.5478,
|
| 1731 |
+
"eval_samples_per_second": 16.473,
|
| 1732 |
+
"eval_steps_per_second": 0.259,
|
| 1733 |
+
"step": 5000
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 2.5,
|
| 1737 |
+
"learning_rate": 1.6555518506168722e-05,
|
| 1738 |
+
"loss": 4.9324,
|
| 1739 |
+
"step": 5020
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.51,
|
| 1743 |
+
"learning_rate": 1.6222074024674894e-05,
|
| 1744 |
+
"loss": 4.9441,
|
| 1745 |
+
"step": 5040
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 2.52,
|
| 1749 |
+
"learning_rate": 1.588862954318106e-05,
|
| 1750 |
+
"loss": 4.9695,
|
| 1751 |
+
"step": 5060
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 2.53,
|
| 1755 |
+
"learning_rate": 1.555518506168723e-05,
|
| 1756 |
+
"loss": 4.9103,
|
| 1757 |
+
"step": 5080
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 2.54,
|
| 1761 |
+
"learning_rate": 1.52217405801934e-05,
|
| 1762 |
+
"loss": 4.9284,
|
| 1763 |
+
"step": 5100
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 2.55,
|
| 1767 |
+
"learning_rate": 1.4888296098699567e-05,
|
| 1768 |
+
"loss": 4.9198,
|
| 1769 |
+
"step": 5120
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 2.56,
|
| 1773 |
+
"learning_rate": 1.4554851617205737e-05,
|
| 1774 |
+
"loss": 4.9303,
|
| 1775 |
+
"step": 5140
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 2.57,
|
| 1779 |
+
"learning_rate": 1.4221407135711903e-05,
|
| 1780 |
+
"loss": 4.8952,
|
| 1781 |
+
"step": 5160
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.58,
|
| 1785 |
+
"learning_rate": 1.3887962654218073e-05,
|
| 1786 |
+
"loss": 4.9393,
|
| 1787 |
+
"step": 5180
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 2.59,
|
| 1791 |
+
"learning_rate": 1.3554518172724243e-05,
|
| 1792 |
+
"loss": 4.9335,
|
| 1793 |
+
"step": 5200
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 2.59,
|
| 1797 |
+
"eval_accuracy": 0.38211472315137435,
|
| 1798 |
+
"eval_loss": 4.98828125,
|
| 1799 |
+
"eval_runtime": 1028.0221,
|
| 1800 |
+
"eval_samples_per_second": 10.04,
|
| 1801 |
+
"eval_steps_per_second": 0.158,
|
| 1802 |
+
"step": 5200
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.6,
|
| 1806 |
+
"learning_rate": 1.3221073691230409e-05,
|
| 1807 |
+
"loss": 4.919,
|
| 1808 |
+
"step": 5220
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 2.61,
|
| 1812 |
+
"learning_rate": 1.2887629209736579e-05,
|
| 1813 |
+
"loss": 4.9249,
|
| 1814 |
+
"step": 5240
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 2.62,
|
| 1818 |
+
"learning_rate": 1.2554184728242748e-05,
|
| 1819 |
+
"loss": 4.8953,
|
| 1820 |
+
"step": 5260
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 2.63,
|
| 1824 |
+
"learning_rate": 1.2220740246748916e-05,
|
| 1825 |
+
"loss": 4.9371,
|
| 1826 |
+
"step": 5280
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 2.64,
|
| 1830 |
+
"learning_rate": 1.1887295765255086e-05,
|
| 1831 |
+
"loss": 4.9752,
|
| 1832 |
+
"step": 5300
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 2.65,
|
| 1836 |
+
"learning_rate": 1.1553851283761254e-05,
|
| 1837 |
+
"loss": 4.9155,
|
| 1838 |
+
"step": 5320
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.66,
|
| 1842 |
+
"learning_rate": 1.1220406802267424e-05,
|
| 1843 |
+
"loss": 4.9085,
|
| 1844 |
+
"step": 5340
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.67,
|
| 1848 |
+
"learning_rate": 1.0886962320773592e-05,
|
| 1849 |
+
"loss": 4.9508,
|
| 1850 |
+
"step": 5360
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.68,
|
| 1854 |
+
"learning_rate": 1.055351783927976e-05,
|
| 1855 |
+
"loss": 4.8926,
|
| 1856 |
+
"step": 5380
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 2.69,
|
| 1860 |
+
"learning_rate": 1.022007335778593e-05,
|
| 1861 |
+
"loss": 4.9231,
|
| 1862 |
+
"step": 5400
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 2.69,
|
| 1866 |
+
"eval_accuracy": 0.38292519929510394,
|
| 1867 |
+
"eval_loss": 4.98046875,
|
| 1868 |
+
"eval_runtime": 786.1183,
|
| 1869 |
+
"eval_samples_per_second": 13.129,
|
| 1870 |
+
"eval_steps_per_second": 0.206,
|
| 1871 |
+
"step": 5400
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.7,
|
| 1875 |
+
"learning_rate": 9.886628876292097e-06,
|
| 1876 |
+
"loss": 4.9256,
|
| 1877 |
+
"step": 5420
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 2.71,
|
| 1881 |
+
"learning_rate": 9.553184394798265e-06,
|
| 1882 |
+
"loss": 4.9355,
|
| 1883 |
+
"step": 5440
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 2.72,
|
| 1887 |
+
"learning_rate": 9.219739913304435e-06,
|
| 1888 |
+
"loss": 4.9676,
|
| 1889 |
+
"step": 5460
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 2.73,
|
| 1893 |
+
"learning_rate": 8.886295431810605e-06,
|
| 1894 |
+
"loss": 4.8958,
|
| 1895 |
+
"step": 5480
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 2.74,
|
| 1899 |
+
"learning_rate": 8.552850950316773e-06,
|
| 1900 |
+
"loss": 4.8705,
|
| 1901 |
+
"step": 5500
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.75,
|
| 1905 |
+
"learning_rate": 8.219406468822942e-06,
|
| 1906 |
+
"loss": 4.9561,
|
| 1907 |
+
"step": 5520
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.76,
|
| 1911 |
+
"learning_rate": 7.88596198732911e-06,
|
| 1912 |
+
"loss": 4.9005,
|
| 1913 |
+
"step": 5540
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.77,
|
| 1917 |
+
"learning_rate": 7.552517505835278e-06,
|
| 1918 |
+
"loss": 4.8861,
|
| 1919 |
+
"step": 5560
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 2.78,
|
| 1923 |
+
"learning_rate": 7.219073024341448e-06,
|
| 1924 |
+
"loss": 4.8598,
|
| 1925 |
+
"step": 5580
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 2.79,
|
| 1929 |
+
"learning_rate": 6.885628542847617e-06,
|
| 1930 |
+
"loss": 4.8886,
|
| 1931 |
+
"step": 5600
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 2.79,
|
| 1935 |
+
"eval_accuracy": 0.3835265126706772,
|
| 1936 |
+
"eval_loss": 4.97265625,
|
| 1937 |
+
"eval_runtime": 539.9358,
|
| 1938 |
+
"eval_samples_per_second": 19.115,
|
| 1939 |
+
"eval_steps_per_second": 0.3,
|
| 1940 |
+
"step": 5600
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 2.8,
|
| 1944 |
+
"learning_rate": 6.552184061353785e-06,
|
| 1945 |
+
"loss": 4.9213,
|
| 1946 |
+
"step": 5620
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 2.81,
|
| 1950 |
+
"learning_rate": 6.2187395798599535e-06,
|
| 1951 |
+
"loss": 4.8934,
|
| 1952 |
+
"step": 5640
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 2.82,
|
| 1956 |
+
"learning_rate": 5.885295098366122e-06,
|
| 1957 |
+
"loss": 4.8822,
|
| 1958 |
+
"step": 5660
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 2.83,
|
| 1962 |
+
"learning_rate": 5.551850616872291e-06,
|
| 1963 |
+
"loss": 4.9389,
|
| 1964 |
+
"step": 5680
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 2.84,
|
| 1968 |
+
"learning_rate": 5.21840613537846e-06,
|
| 1969 |
+
"loss": 4.9103,
|
| 1970 |
+
"step": 5700
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.85,
|
| 1974 |
+
"learning_rate": 4.884961653884629e-06,
|
| 1975 |
+
"loss": 4.8878,
|
| 1976 |
+
"step": 5720
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.86,
|
| 1980 |
+
"learning_rate": 4.551517172390797e-06,
|
| 1981 |
+
"loss": 4.9294,
|
| 1982 |
+
"step": 5740
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 2.87,
|
| 1986 |
+
"learning_rate": 4.218072690896966e-06,
|
| 1987 |
+
"loss": 4.9309,
|
| 1988 |
+
"step": 5760
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 2.88,
|
| 1992 |
+
"learning_rate": 3.8846282094031345e-06,
|
| 1993 |
+
"loss": 4.8913,
|
| 1994 |
+
"step": 5780
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 2.89,
|
| 1998 |
+
"learning_rate": 3.5511837279093033e-06,
|
| 1999 |
+
"loss": 4.9419,
|
| 2000 |
+
"step": 5800
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 2.89,
|
| 2004 |
+
"eval_accuracy": 0.38428875689298764,
|
| 2005 |
+
"eval_loss": 4.96484375,
|
| 2006 |
+
"eval_runtime": 548.4975,
|
| 2007 |
+
"eval_samples_per_second": 18.817,
|
| 2008 |
+
"eval_steps_per_second": 0.295,
|
| 2009 |
+
"step": 5800
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 2.9,
|
| 2013 |
+
"learning_rate": 3.217739246415472e-06,
|
| 2014 |
+
"loss": 4.9406,
|
| 2015 |
+
"step": 5820
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 2.91,
|
| 2019 |
+
"learning_rate": 2.884294764921641e-06,
|
| 2020 |
+
"loss": 4.8887,
|
| 2021 |
+
"step": 5840
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 2.92,
|
| 2025 |
+
"learning_rate": 2.5508502834278094e-06,
|
| 2026 |
+
"loss": 4.9541,
|
| 2027 |
+
"step": 5860
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 2.93,
|
| 2031 |
+
"learning_rate": 2.2174058019339782e-06,
|
| 2032 |
+
"loss": 4.8878,
|
| 2033 |
+
"step": 5880
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.94,
|
| 2037 |
+
"learning_rate": 1.8839613204401467e-06,
|
| 2038 |
+
"loss": 4.9403,
|
| 2039 |
+
"step": 5900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.95,
|
| 2043 |
+
"learning_rate": 1.5505168389463155e-06,
|
| 2044 |
+
"loss": 4.8785,
|
| 2045 |
+
"step": 5920
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 2.96,
|
| 2049 |
+
"learning_rate": 1.2170723574524843e-06,
|
| 2050 |
+
"loss": 4.9111,
|
| 2051 |
+
"step": 5940
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 2.97,
|
| 2055 |
+
"learning_rate": 8.836278759586529e-07,
|
| 2056 |
+
"loss": 4.8861,
|
| 2057 |
+
"step": 5960
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 2.98,
|
| 2061 |
+
"learning_rate": 5.501833944648216e-07,
|
| 2062 |
+
"loss": 4.8789,
|
| 2063 |
+
"step": 5980
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 2.99,
|
| 2067 |
+
"learning_rate": 2.1673891297099036e-07,
|
| 2068 |
+
"loss": 4.9227,
|
| 2069 |
+
"step": 6000
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.99,
|
| 2073 |
+
"eval_accuracy": 0.3842425129408517,
|
| 2074 |
+
"eval_loss": 4.96484375,
|
| 2075 |
+
"eval_runtime": 533.8356,
|
| 2076 |
+
"eval_samples_per_second": 19.334,
|
| 2077 |
+
"eval_steps_per_second": 0.303,
|
| 2078 |
+
"step": 6000
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 2.99,
|
| 2082 |
+
"step": 6001,
|
| 2083 |
+
"total_flos": 6.431553336633197e+18,
|
| 2084 |
+
"train_loss": 0.0007667992834527579,
|
| 2085 |
+
"train_runtime": 4.5998,
|
| 2086 |
+
"train_samples_per_second": 125223.897,
|
| 2087 |
+
"train_steps_per_second": 1304.416
|
| 2088 |
+
}
|
| 2089 |
+
],
|
| 2090 |
+
"max_steps": 6000,
|
| 2091 |
+
"num_train_epochs": 3,
|
| 2092 |
+
"total_flos": 6.431553336633197e+18,
|
| 2093 |
+
"trial_name": null,
|
| 2094 |
+
"trial_params": null
|
| 2095 |
+
}
|