End of training
Browse files
README.md
CHANGED
@@ -1,11 +1,23 @@
|
|
1 |
---
|
2 |
tags:
|
3 |
- generated_from_trainer
|
|
|
|
|
4 |
metrics:
|
5 |
- accuracy
|
6 |
model-index:
|
7 |
- name: smolm-autoreg-bpe-counterfactual-babylm-only_other_det_removal-1e-4
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -13,7 +25,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# smolm-autoreg-bpe-counterfactual-babylm-only_other_det_removal-1e-4
|
15 |
|
16 |
-
This model was trained from scratch on
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- Loss: 3.4247
|
19 |
- Accuracy: 0.4065
|
|
|
1 |
---
|
2 |
tags:
|
3 |
- generated_from_trainer
|
4 |
+
datasets:
|
5 |
+
- kanishka/counterfactual-babylm-only_other_det_removal
|
6 |
metrics:
|
7 |
- accuracy
|
8 |
model-index:
|
9 |
- name: smolm-autoreg-bpe-counterfactual-babylm-only_other_det_removal-1e-4
|
10 |
+
results:
|
11 |
+
- task:
|
12 |
+
name: Causal Language Modeling
|
13 |
+
type: text-generation
|
14 |
+
dataset:
|
15 |
+
name: kanishka/counterfactual-babylm-only_other_det_removal
|
16 |
+
type: kanishka/counterfactual-babylm-only_other_det_removal
|
17 |
+
metrics:
|
18 |
+
- name: Accuracy
|
19 |
+
type: accuracy
|
20 |
+
value: 0.40654968657553286
|
21 |
---
|
22 |
|
23 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
25 |
|
26 |
# smolm-autoreg-bpe-counterfactual-babylm-only_other_det_removal-1e-4
|
27 |
|
28 |
+
This model was trained from scratch on the kanishka/counterfactual-babylm-only_other_det_removal dataset.
|
29 |
It achieves the following results on the evaluation set:
|
30 |
- Loss: 3.4247
|
31 |
- Accuracy: 0.4065
|
all_results.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 20.0,
|
3 |
+
"eval_accuracy": 0.40654968657553286,
|
4 |
+
"eval_loss": 3.424689292907715,
|
5 |
+
"eval_runtime": 153.5128,
|
6 |
+
"eval_samples": 57918,
|
7 |
+
"eval_samples_per_second": 377.285,
|
8 |
+
"eval_steps_per_second": 5.895,
|
9 |
+
"perplexity": 30.71310059066534,
|
10 |
+
"train_loss": 3.1534978556531716,
|
11 |
+
"train_runtime": 80977.43,
|
12 |
+
"train_samples": 595074,
|
13 |
+
"train_samples_per_second": 146.973,
|
14 |
+
"train_steps_per_second": 4.593
|
15 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 20.0,
|
3 |
+
"eval_accuracy": 0.40654968657553286,
|
4 |
+
"eval_loss": 3.424689292907715,
|
5 |
+
"eval_runtime": 153.5128,
|
6 |
+
"eval_samples": 57918,
|
7 |
+
"eval_samples_per_second": 377.285,
|
8 |
+
"eval_steps_per_second": 5.895,
|
9 |
+
"perplexity": 30.71310059066534
|
10 |
+
}
|
runs/Feb06_09-44-59_phyl-ling-p01.la.utexas.edu/events.out.tfevents.1707315812.phyl-ling-p01.la.utexas.edu.3217545.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fa29ad8105f271291b12fa5066b4ce734ef3c2a093a3b3ee9a1d6306d14d4956
|
3 |
+
size 417
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 20.0,
|
3 |
+
"train_loss": 3.1534978556531716,
|
4 |
+
"train_runtime": 80977.43,
|
5 |
+
"train_samples": 595074,
|
6 |
+
"train_samples_per_second": 146.973,
|
7 |
+
"train_steps_per_second": 4.593
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 20.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 371940,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.05,
|
13 |
+
"learning_rate": 3.125e-06,
|
14 |
+
"loss": 7.5485,
|
15 |
+
"step": 1000
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.11,
|
19 |
+
"learning_rate": 6.25e-06,
|
20 |
+
"loss": 5.8151,
|
21 |
+
"step": 2000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.16,
|
25 |
+
"learning_rate": 9.375000000000001e-06,
|
26 |
+
"loss": 5.3768,
|
27 |
+
"step": 3000
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.22,
|
31 |
+
"learning_rate": 1.25e-05,
|
32 |
+
"loss": 5.1553,
|
33 |
+
"step": 4000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.27,
|
37 |
+
"learning_rate": 1.5625e-05,
|
38 |
+
"loss": 5.0099,
|
39 |
+
"step": 5000
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.32,
|
43 |
+
"learning_rate": 1.8750000000000002e-05,
|
44 |
+
"loss": 4.8767,
|
45 |
+
"step": 6000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.38,
|
49 |
+
"learning_rate": 2.1875e-05,
|
50 |
+
"loss": 4.7604,
|
51 |
+
"step": 7000
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.43,
|
55 |
+
"learning_rate": 2.5e-05,
|
56 |
+
"loss": 4.6562,
|
57 |
+
"step": 8000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.48,
|
61 |
+
"learning_rate": 2.8125000000000003e-05,
|
62 |
+
"loss": 4.5602,
|
63 |
+
"step": 9000
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.54,
|
67 |
+
"learning_rate": 3.125e-05,
|
68 |
+
"loss": 4.4821,
|
69 |
+
"step": 10000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.59,
|
73 |
+
"learning_rate": 3.4375e-05,
|
74 |
+
"loss": 4.4129,
|
75 |
+
"step": 11000
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.65,
|
79 |
+
"learning_rate": 3.7500000000000003e-05,
|
80 |
+
"loss": 4.3503,
|
81 |
+
"step": 12000
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.7,
|
85 |
+
"learning_rate": 4.061875e-05,
|
86 |
+
"loss": 4.2942,
|
87 |
+
"step": 13000
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.75,
|
91 |
+
"learning_rate": 4.374375e-05,
|
92 |
+
"loss": 4.2367,
|
93 |
+
"step": 14000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.81,
|
97 |
+
"learning_rate": 4.686875e-05,
|
98 |
+
"loss": 4.1833,
|
99 |
+
"step": 15000
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.86,
|
103 |
+
"learning_rate": 4.999375e-05,
|
104 |
+
"loss": 4.1412,
|
105 |
+
"step": 16000
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.91,
|
109 |
+
"learning_rate": 5.3115625000000005e-05,
|
110 |
+
"loss": 4.0988,
|
111 |
+
"step": 17000
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.97,
|
115 |
+
"learning_rate": 5.62375e-05,
|
116 |
+
"loss": 4.0532,
|
117 |
+
"step": 18000
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 1.0,
|
121 |
+
"eval_accuracy": 0.30846132523601394,
|
122 |
+
"eval_loss": 4.257880210876465,
|
123 |
+
"eval_runtime": 153.3032,
|
124 |
+
"eval_samples_per_second": 377.8,
|
125 |
+
"eval_steps_per_second": 5.903,
|
126 |
+
"step": 18597
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 1.02,
|
130 |
+
"learning_rate": 5.93625e-05,
|
131 |
+
"loss": 4.0065,
|
132 |
+
"step": 19000
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 1.08,
|
136 |
+
"learning_rate": 6.24875e-05,
|
137 |
+
"loss": 3.9678,
|
138 |
+
"step": 20000
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 1.13,
|
142 |
+
"learning_rate": 6.56125e-05,
|
143 |
+
"loss": 3.9301,
|
144 |
+
"step": 21000
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.18,
|
148 |
+
"learning_rate": 6.8734375e-05,
|
149 |
+
"loss": 3.8874,
|
150 |
+
"step": 22000
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 1.24,
|
154 |
+
"learning_rate": 7.185625e-05,
|
155 |
+
"loss": 3.8598,
|
156 |
+
"step": 23000
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 1.29,
|
160 |
+
"learning_rate": 7.498125e-05,
|
161 |
+
"loss": 3.8228,
|
162 |
+
"step": 24000
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 1.34,
|
166 |
+
"learning_rate": 7.8103125e-05,
|
167 |
+
"loss": 3.7974,
|
168 |
+
"step": 25000
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.4,
|
172 |
+
"learning_rate": 8.1228125e-05,
|
173 |
+
"loss": 3.7705,
|
174 |
+
"step": 26000
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 1.45,
|
178 |
+
"learning_rate": 8.435e-05,
|
179 |
+
"loss": 3.7446,
|
180 |
+
"step": 27000
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 1.51,
|
184 |
+
"learning_rate": 8.747500000000001e-05,
|
185 |
+
"loss": 3.7237,
|
186 |
+
"step": 28000
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 1.56,
|
190 |
+
"learning_rate": 9.06e-05,
|
191 |
+
"loss": 3.7083,
|
192 |
+
"step": 29000
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 1.61,
|
196 |
+
"learning_rate": 9.3721875e-05,
|
197 |
+
"loss": 3.6864,
|
198 |
+
"step": 30000
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 1.67,
|
202 |
+
"learning_rate": 9.6846875e-05,
|
203 |
+
"loss": 3.6687,
|
204 |
+
"step": 31000
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 1.72,
|
208 |
+
"learning_rate": 9.996875e-05,
|
209 |
+
"loss": 3.6513,
|
210 |
+
"step": 32000
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 1.77,
|
214 |
+
"learning_rate": 9.970877213625934e-05,
|
215 |
+
"loss": 3.6299,
|
216 |
+
"step": 33000
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 1.83,
|
220 |
+
"learning_rate": 9.941460257692534e-05,
|
221 |
+
"loss": 3.6153,
|
222 |
+
"step": 34000
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 1.88,
|
226 |
+
"learning_rate": 9.912072718715068e-05,
|
227 |
+
"loss": 3.5913,
|
228 |
+
"step": 35000
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 1.94,
|
232 |
+
"learning_rate": 9.882655762781668e-05,
|
233 |
+
"loss": 3.5765,
|
234 |
+
"step": 36000
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 1.99,
|
238 |
+
"learning_rate": 9.8532682238042e-05,
|
239 |
+
"loss": 3.566,
|
240 |
+
"step": 37000
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 2.0,
|
244 |
+
"eval_accuracy": 0.3620144970824847,
|
245 |
+
"eval_loss": 3.760495662689209,
|
246 |
+
"eval_runtime": 155.0908,
|
247 |
+
"eval_samples_per_second": 373.446,
|
248 |
+
"eval_steps_per_second": 5.835,
|
249 |
+
"step": 37194
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 2.04,
|
253 |
+
"learning_rate": 9.823851267870802e-05,
|
254 |
+
"loss": 3.5234,
|
255 |
+
"step": 38000
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 2.1,
|
259 |
+
"learning_rate": 9.794463728893335e-05,
|
260 |
+
"loss": 3.5119,
|
261 |
+
"step": 39000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 2.15,
|
265 |
+
"learning_rate": 9.765046772959934e-05,
|
266 |
+
"loss": 3.5002,
|
267 |
+
"step": 40000
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 2.2,
|
271 |
+
"learning_rate": 9.735629817026534e-05,
|
272 |
+
"loss": 3.4914,
|
273 |
+
"step": 41000
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 2.26,
|
277 |
+
"learning_rate": 9.706212861093135e-05,
|
278 |
+
"loss": 3.4834,
|
279 |
+
"step": 42000
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 2.31,
|
283 |
+
"learning_rate": 9.676825322115668e-05,
|
284 |
+
"loss": 3.4774,
|
285 |
+
"step": 43000
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 2.37,
|
289 |
+
"learning_rate": 9.647437783138201e-05,
|
290 |
+
"loss": 3.4655,
|
291 |
+
"step": 44000
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 2.42,
|
295 |
+
"learning_rate": 9.618020827204801e-05,
|
296 |
+
"loss": 3.4555,
|
297 |
+
"step": 45000
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 2.47,
|
301 |
+
"learning_rate": 9.588603871271401e-05,
|
302 |
+
"loss": 3.4512,
|
303 |
+
"step": 46000
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 2.53,
|
307 |
+
"learning_rate": 9.559186915338e-05,
|
308 |
+
"loss": 3.4414,
|
309 |
+
"step": 47000
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 2.58,
|
313 |
+
"learning_rate": 9.529799376360535e-05,
|
314 |
+
"loss": 3.4313,
|
315 |
+
"step": 48000
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 2.63,
|
319 |
+
"learning_rate": 9.500382420427135e-05,
|
320 |
+
"loss": 3.4242,
|
321 |
+
"step": 49000
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 2.69,
|
325 |
+
"learning_rate": 9.470965464493736e-05,
|
326 |
+
"loss": 3.4185,
|
327 |
+
"step": 50000
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 2.74,
|
331 |
+
"learning_rate": 9.441577925516269e-05,
|
332 |
+
"loss": 3.416,
|
333 |
+
"step": 51000
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 2.8,
|
337 |
+
"learning_rate": 9.412160969582868e-05,
|
338 |
+
"loss": 3.4054,
|
339 |
+
"step": 52000
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 2.85,
|
343 |
+
"learning_rate": 9.382744013649468e-05,
|
344 |
+
"loss": 3.3961,
|
345 |
+
"step": 53000
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 2.9,
|
349 |
+
"learning_rate": 9.353356474672001e-05,
|
350 |
+
"loss": 3.3925,
|
351 |
+
"step": 54000
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 2.96,
|
355 |
+
"learning_rate": 9.323939518738601e-05,
|
356 |
+
"loss": 3.3886,
|
357 |
+
"step": 55000
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 3.0,
|
361 |
+
"eval_accuracy": 0.3805629265473003,
|
362 |
+
"eval_loss": 3.596226453781128,
|
363 |
+
"eval_runtime": 155.0494,
|
364 |
+
"eval_samples_per_second": 373.545,
|
365 |
+
"eval_steps_per_second": 5.837,
|
366 |
+
"step": 55791
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 3.01,
|
370 |
+
"learning_rate": 9.294551979761135e-05,
|
371 |
+
"loss": 3.3707,
|
372 |
+
"step": 56000
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 3.07,
|
376 |
+
"learning_rate": 9.265164440783668e-05,
|
377 |
+
"loss": 3.3276,
|
378 |
+
"step": 57000
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 3.12,
|
382 |
+
"learning_rate": 9.235747484850268e-05,
|
383 |
+
"loss": 3.3305,
|
384 |
+
"step": 58000
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 3.17,
|
388 |
+
"learning_rate": 9.206330528916868e-05,
|
389 |
+
"loss": 3.3319,
|
390 |
+
"step": 59000
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 3.23,
|
394 |
+
"learning_rate": 9.176913572983467e-05,
|
395 |
+
"loss": 3.3226,
|
396 |
+
"step": 60000
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 3.28,
|
400 |
+
"learning_rate": 9.147526034006e-05,
|
401 |
+
"loss": 3.3256,
|
402 |
+
"step": 61000
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 3.33,
|
406 |
+
"learning_rate": 9.118109078072601e-05,
|
407 |
+
"loss": 3.3253,
|
408 |
+
"step": 62000
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 3.39,
|
412 |
+
"learning_rate": 9.088721539095136e-05,
|
413 |
+
"loss": 3.3165,
|
414 |
+
"step": 63000
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 3.44,
|
418 |
+
"learning_rate": 9.059304583161736e-05,
|
419 |
+
"loss": 3.3167,
|
420 |
+
"step": 64000
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 3.5,
|
424 |
+
"learning_rate": 9.029917044184268e-05,
|
425 |
+
"loss": 3.3141,
|
426 |
+
"step": 65000
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 3.55,
|
430 |
+
"learning_rate": 9.000529505206801e-05,
|
431 |
+
"loss": 3.3114,
|
432 |
+
"step": 66000
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 3.6,
|
436 |
+
"learning_rate": 8.971112549273401e-05,
|
437 |
+
"loss": 3.3068,
|
438 |
+
"step": 67000
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 3.66,
|
442 |
+
"learning_rate": 8.941695593340001e-05,
|
443 |
+
"loss": 3.3044,
|
444 |
+
"step": 68000
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 3.71,
|
448 |
+
"learning_rate": 8.912278637406601e-05,
|
449 |
+
"loss": 3.2989,
|
450 |
+
"step": 69000
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 3.76,
|
454 |
+
"learning_rate": 8.882891098429135e-05,
|
455 |
+
"loss": 3.3017,
|
456 |
+
"step": 70000
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 3.82,
|
460 |
+
"learning_rate": 8.853503559451668e-05,
|
461 |
+
"loss": 3.2947,
|
462 |
+
"step": 71000
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 3.87,
|
466 |
+
"learning_rate": 8.824086603518268e-05,
|
467 |
+
"loss": 3.2908,
|
468 |
+
"step": 72000
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 3.93,
|
472 |
+
"learning_rate": 8.794669647584869e-05,
|
473 |
+
"loss": 3.29,
|
474 |
+
"step": 73000
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 3.98,
|
478 |
+
"learning_rate": 8.765252691651469e-05,
|
479 |
+
"loss": 3.2899,
|
480 |
+
"step": 74000
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 4.0,
|
484 |
+
"eval_accuracy": 0.38936209692947077,
|
485 |
+
"eval_loss": 3.5175118446350098,
|
486 |
+
"eval_runtime": 154.67,
|
487 |
+
"eval_samples_per_second": 374.462,
|
488 |
+
"eval_steps_per_second": 5.851,
|
489 |
+
"step": 74388
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 4.03,
|
493 |
+
"learning_rate": 8.735865152674002e-05,
|
494 |
+
"loss": 3.2502,
|
495 |
+
"step": 75000
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 4.09,
|
499 |
+
"learning_rate": 8.706448196740601e-05,
|
500 |
+
"loss": 3.2342,
|
501 |
+
"step": 76000
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 4.14,
|
505 |
+
"learning_rate": 8.677031240807203e-05,
|
506 |
+
"loss": 3.2327,
|
507 |
+
"step": 77000
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 4.19,
|
511 |
+
"learning_rate": 8.647614284873802e-05,
|
512 |
+
"loss": 3.2342,
|
513 |
+
"step": 78000
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 4.25,
|
517 |
+
"learning_rate": 8.618226745896335e-05,
|
518 |
+
"loss": 3.2326,
|
519 |
+
"step": 79000
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 4.3,
|
523 |
+
"learning_rate": 8.588809789962935e-05,
|
524 |
+
"loss": 3.2345,
|
525 |
+
"step": 80000
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 4.36,
|
529 |
+
"learning_rate": 8.559392834029535e-05,
|
530 |
+
"loss": 3.2361,
|
531 |
+
"step": 81000
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 4.41,
|
535 |
+
"learning_rate": 8.529975878096135e-05,
|
536 |
+
"loss": 3.2337,
|
537 |
+
"step": 82000
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 4.46,
|
541 |
+
"learning_rate": 8.500588339118668e-05,
|
542 |
+
"loss": 3.2295,
|
543 |
+
"step": 83000
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 4.52,
|
547 |
+
"learning_rate": 8.471171383185267e-05,
|
548 |
+
"loss": 3.2325,
|
549 |
+
"step": 84000
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 4.57,
|
553 |
+
"learning_rate": 8.441754427251868e-05,
|
554 |
+
"loss": 3.2275,
|
555 |
+
"step": 85000
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 4.62,
|
559 |
+
"learning_rate": 8.412337471318468e-05,
|
560 |
+
"loss": 3.2247,
|
561 |
+
"step": 86000
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 4.68,
|
565 |
+
"learning_rate": 8.382949932341003e-05,
|
566 |
+
"loss": 3.2267,
|
567 |
+
"step": 87000
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 4.73,
|
571 |
+
"learning_rate": 8.353532976407602e-05,
|
572 |
+
"loss": 3.2255,
|
573 |
+
"step": 88000
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 4.79,
|
577 |
+
"learning_rate": 8.324145437430135e-05,
|
578 |
+
"loss": 3.2242,
|
579 |
+
"step": 89000
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 4.84,
|
583 |
+
"learning_rate": 8.294728481496735e-05,
|
584 |
+
"loss": 3.2247,
|
585 |
+
"step": 90000
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 4.89,
|
589 |
+
"learning_rate": 8.265340942519268e-05,
|
590 |
+
"loss": 3.2213,
|
591 |
+
"step": 91000
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 4.95,
|
595 |
+
"learning_rate": 8.235923986585868e-05,
|
596 |
+
"loss": 3.2214,
|
597 |
+
"step": 92000
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 5.0,
|
601 |
+
"eval_accuracy": 0.3939114645245801,
|
602 |
+
"eval_loss": 3.4618098735809326,
|
603 |
+
"eval_runtime": 153.3377,
|
604 |
+
"eval_samples_per_second": 377.715,
|
605 |
+
"eval_steps_per_second": 5.902,
|
606 |
+
"step": 92985
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 5.0,
|
610 |
+
"learning_rate": 8.206536447608402e-05,
|
611 |
+
"loss": 3.2176,
|
612 |
+
"step": 93000
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 5.05,
|
616 |
+
"learning_rate": 8.177119491675002e-05,
|
617 |
+
"loss": 3.1605,
|
618 |
+
"step": 94000
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 5.11,
|
622 |
+
"learning_rate": 8.147702535741602e-05,
|
623 |
+
"loss": 3.1621,
|
624 |
+
"step": 95000
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 5.16,
|
628 |
+
"learning_rate": 8.118314996764134e-05,
|
629 |
+
"loss": 3.1684,
|
630 |
+
"step": 96000
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 5.22,
|
634 |
+
"learning_rate": 8.088927457786669e-05,
|
635 |
+
"loss": 3.1668,
|
636 |
+
"step": 97000
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 5.27,
|
640 |
+
"learning_rate": 8.059510501853269e-05,
|
641 |
+
"loss": 3.1728,
|
642 |
+
"step": 98000
|
643 |
+
},
|
644 |
+
{
|
645 |
+
"epoch": 5.32,
|
646 |
+
"learning_rate": 8.030122962875803e-05,
|
647 |
+
"loss": 3.1745,
|
648 |
+
"step": 99000
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 5.38,
|
652 |
+
"learning_rate": 8.000706006942403e-05,
|
653 |
+
"loss": 3.1689,
|
654 |
+
"step": 100000
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 5.43,
|
658 |
+
"learning_rate": 7.971289051009002e-05,
|
659 |
+
"loss": 3.1721,
|
660 |
+
"step": 101000
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 5.48,
|
664 |
+
"learning_rate": 7.941872095075602e-05,
|
665 |
+
"loss": 3.168,
|
666 |
+
"step": 102000
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 5.54,
|
670 |
+
"learning_rate": 7.912484556098135e-05,
|
671 |
+
"loss": 3.172,
|
672 |
+
"step": 103000
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 5.59,
|
676 |
+
"learning_rate": 7.883067600164735e-05,
|
677 |
+
"loss": 3.1735,
|
678 |
+
"step": 104000
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 5.65,
|
682 |
+
"learning_rate": 7.853650644231335e-05,
|
683 |
+
"loss": 3.1725,
|
684 |
+
"step": 105000
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 5.7,
|
688 |
+
"learning_rate": 7.824263105253869e-05,
|
689 |
+
"loss": 3.1731,
|
690 |
+
"step": 106000
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 5.75,
|
694 |
+
"learning_rate": 7.794846149320469e-05,
|
695 |
+
"loss": 3.1738,
|
696 |
+
"step": 107000
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 5.81,
|
700 |
+
"learning_rate": 7.765458610343002e-05,
|
701 |
+
"loss": 3.1708,
|
702 |
+
"step": 108000
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 5.86,
|
706 |
+
"learning_rate": 7.736041654409603e-05,
|
707 |
+
"loss": 3.1697,
|
708 |
+
"step": 109000
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 5.91,
|
712 |
+
"learning_rate": 7.706624698476203e-05,
|
713 |
+
"loss": 3.169,
|
714 |
+
"step": 110000
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 5.97,
|
718 |
+
"learning_rate": 7.677207742542802e-05,
|
719 |
+
"loss": 3.1702,
|
720 |
+
"step": 111000
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 6.0,
|
724 |
+
"eval_accuracy": 0.3978669427043171,
|
725 |
+
"eval_loss": 3.425184726715088,
|
726 |
+
"eval_runtime": 153.9864,
|
727 |
+
"eval_samples_per_second": 376.124,
|
728 |
+
"eval_steps_per_second": 5.877,
|
729 |
+
"step": 111582
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 6.02,
|
733 |
+
"learning_rate": 7.647820203565335e-05,
|
734 |
+
"loss": 3.1396,
|
735 |
+
"step": 112000
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 6.08,
|
739 |
+
"learning_rate": 7.618403247631935e-05,
|
740 |
+
"loss": 3.1105,
|
741 |
+
"step": 113000
|
742 |
+
},
|
743 |
+
{
|
744 |
+
"epoch": 6.13,
|
745 |
+
"learning_rate": 7.589015708654469e-05,
|
746 |
+
"loss": 3.115,
|
747 |
+
"step": 114000
|
748 |
+
},
|
749 |
+
{
|
750 |
+
"epoch": 6.18,
|
751 |
+
"learning_rate": 7.559598752721069e-05,
|
752 |
+
"loss": 3.1222,
|
753 |
+
"step": 115000
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 6.24,
|
757 |
+
"learning_rate": 7.530181796787669e-05,
|
758 |
+
"loss": 3.1212,
|
759 |
+
"step": 116000
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 6.29,
|
763 |
+
"learning_rate": 7.500794257810202e-05,
|
764 |
+
"loss": 3.1252,
|
765 |
+
"step": 117000
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 6.35,
|
769 |
+
"learning_rate": 7.471377301876802e-05,
|
770 |
+
"loss": 3.1216,
|
771 |
+
"step": 118000
|
772 |
+
},
|
773 |
+
{
|
774 |
+
"epoch": 6.4,
|
775 |
+
"learning_rate": 7.441960345943401e-05,
|
776 |
+
"loss": 3.1252,
|
777 |
+
"step": 119000
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 6.45,
|
781 |
+
"learning_rate": 7.412543390010001e-05,
|
782 |
+
"loss": 3.1233,
|
783 |
+
"step": 120000
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 6.51,
|
787 |
+
"learning_rate": 7.383126434076602e-05,
|
788 |
+
"loss": 3.1243,
|
789 |
+
"step": 121000
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 6.56,
|
793 |
+
"learning_rate": 7.353738895099135e-05,
|
794 |
+
"loss": 3.1264,
|
795 |
+
"step": 122000
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 6.61,
|
799 |
+
"learning_rate": 7.324321939165736e-05,
|
800 |
+
"loss": 3.1294,
|
801 |
+
"step": 123000
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 6.67,
|
805 |
+
"learning_rate": 7.29493440018827e-05,
|
806 |
+
"loss": 3.1263,
|
807 |
+
"step": 124000
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 6.72,
|
811 |
+
"learning_rate": 7.265517444254869e-05,
|
812 |
+
"loss": 3.1298,
|
813 |
+
"step": 125000
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 6.78,
|
817 |
+
"learning_rate": 7.236100488321469e-05,
|
818 |
+
"loss": 3.1298,
|
819 |
+
"step": 126000
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 6.83,
|
823 |
+
"learning_rate": 7.206712949344002e-05,
|
824 |
+
"loss": 3.1292,
|
825 |
+
"step": 127000
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 6.88,
|
829 |
+
"learning_rate": 7.177325410366536e-05,
|
830 |
+
"loss": 3.1258,
|
831 |
+
"step": 128000
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 6.94,
|
835 |
+
"learning_rate": 7.147908454433136e-05,
|
836 |
+
"loss": 3.1263,
|
837 |
+
"step": 129000
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 6.99,
|
841 |
+
"learning_rate": 7.118491498499736e-05,
|
842 |
+
"loss": 3.1294,
|
843 |
+
"step": 130000
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 7.0,
|
847 |
+
"eval_accuracy": 0.39953184852917645,
|
848 |
+
"eval_loss": 3.425506114959717,
|
849 |
+
"eval_runtime": 153.9146,
|
850 |
+
"eval_samples_per_second": 376.3,
|
851 |
+
"eval_steps_per_second": 5.88,
|
852 |
+
"step": 130179
|
853 |
+
},
|
854 |
+
{
|
855 |
+
"epoch": 7.04,
|
856 |
+
"learning_rate": 7.089074542566336e-05,
|
857 |
+
"loss": 3.0759,
|
858 |
+
"step": 131000
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 7.1,
|
862 |
+
"learning_rate": 7.059687003588868e-05,
|
863 |
+
"loss": 3.0733,
|
864 |
+
"step": 132000
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 7.15,
|
868 |
+
"learning_rate": 7.03027004765547e-05,
|
869 |
+
"loss": 3.0779,
|
870 |
+
"step": 133000
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 7.21,
|
874 |
+
"learning_rate": 7.000911925633936e-05,
|
875 |
+
"loss": 3.0819,
|
876 |
+
"step": 134000
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 7.26,
|
880 |
+
"learning_rate": 6.971494969700537e-05,
|
881 |
+
"loss": 3.0831,
|
882 |
+
"step": 135000
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 7.31,
|
886 |
+
"learning_rate": 6.942078013767136e-05,
|
887 |
+
"loss": 3.0837,
|
888 |
+
"step": 136000
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 7.37,
|
892 |
+
"learning_rate": 6.912661057833736e-05,
|
893 |
+
"loss": 3.0831,
|
894 |
+
"step": 137000
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 7.42,
|
898 |
+
"learning_rate": 6.883244101900336e-05,
|
899 |
+
"loss": 3.0849,
|
900 |
+
"step": 138000
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 7.47,
|
904 |
+
"learning_rate": 6.853856562922869e-05,
|
905 |
+
"loss": 3.0915,
|
906 |
+
"step": 139000
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 7.53,
|
910 |
+
"learning_rate": 6.824439606989469e-05,
|
911 |
+
"loss": 3.0902,
|
912 |
+
"step": 140000
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 7.58,
|
916 |
+
"learning_rate": 6.795022651056069e-05,
|
917 |
+
"loss": 3.0879,
|
918 |
+
"step": 141000
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 7.64,
|
922 |
+
"learning_rate": 6.765635112078601e-05,
|
923 |
+
"loss": 3.0852,
|
924 |
+
"step": 142000
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 7.69,
|
928 |
+
"learning_rate": 6.736247573101136e-05,
|
929 |
+
"loss": 3.0939,
|
930 |
+
"step": 143000
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 7.74,
|
934 |
+
"learning_rate": 6.706830617167736e-05,
|
935 |
+
"loss": 3.0892,
|
936 |
+
"step": 144000
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 7.8,
|
940 |
+
"learning_rate": 6.67744307819027e-05,
|
941 |
+
"loss": 3.0892,
|
942 |
+
"step": 145000
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 7.85,
|
946 |
+
"learning_rate": 6.64802612225687e-05,
|
947 |
+
"loss": 3.0942,
|
948 |
+
"step": 146000
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 7.9,
|
952 |
+
"learning_rate": 6.61860916632347e-05,
|
953 |
+
"loss": 3.0951,
|
954 |
+
"step": 147000
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 7.96,
|
958 |
+
"learning_rate": 6.589192210390069e-05,
|
959 |
+
"loss": 3.0899,
|
960 |
+
"step": 148000
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 8.0,
|
964 |
+
"eval_accuracy": 0.40102234972481554,
|
965 |
+
"eval_loss": 3.418954372406006,
|
966 |
+
"eval_runtime": 153.2262,
|
967 |
+
"eval_samples_per_second": 377.99,
|
968 |
+
"eval_steps_per_second": 5.906,
|
969 |
+
"step": 148776
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 8.01,
|
973 |
+
"learning_rate": 6.559775254456669e-05,
|
974 |
+
"loss": 3.0764,
|
975 |
+
"step": 149000
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 8.07,
|
979 |
+
"learning_rate": 6.530387715479203e-05,
|
980 |
+
"loss": 3.0342,
|
981 |
+
"step": 150000
|
982 |
+
},
|
983 |
+
{
|
984 |
+
"epoch": 8.12,
|
985 |
+
"learning_rate": 6.500970759545803e-05,
|
986 |
+
"loss": 3.0423,
|
987 |
+
"step": 151000
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 8.17,
|
991 |
+
"learning_rate": 6.471583220568336e-05,
|
992 |
+
"loss": 3.0441,
|
993 |
+
"step": 152000
|
994 |
+
},
|
995 |
+
{
|
996 |
+
"epoch": 8.23,
|
997 |
+
"learning_rate": 6.442195681590869e-05,
|
998 |
+
"loss": 3.0435,
|
999 |
+
"step": 153000
|
1000 |
+
},
|
1001 |
+
{
|
1002 |
+
"epoch": 8.28,
|
1003 |
+
"learning_rate": 6.412778725657469e-05,
|
1004 |
+
"loss": 3.0519,
|
1005 |
+
"step": 154000
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 8.33,
|
1009 |
+
"learning_rate": 6.383391186680003e-05,
|
1010 |
+
"loss": 3.05,
|
1011 |
+
"step": 155000
|
1012 |
+
},
|
1013 |
+
{
|
1014 |
+
"epoch": 8.39,
|
1015 |
+
"learning_rate": 6.353974230746603e-05,
|
1016 |
+
"loss": 3.0542,
|
1017 |
+
"step": 156000
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 8.44,
|
1021 |
+
"learning_rate": 6.324557274813202e-05,
|
1022 |
+
"loss": 3.0513,
|
1023 |
+
"step": 157000
|
1024 |
+
},
|
1025 |
+
{
|
1026 |
+
"epoch": 8.5,
|
1027 |
+
"learning_rate": 6.295169735835737e-05,
|
1028 |
+
"loss": 3.0561,
|
1029 |
+
"step": 158000
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 8.55,
|
1033 |
+
"learning_rate": 6.265752779902337e-05,
|
1034 |
+
"loss": 3.0625,
|
1035 |
+
"step": 159000
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"epoch": 8.6,
|
1039 |
+
"learning_rate": 6.236335823968936e-05,
|
1040 |
+
"loss": 3.057,
|
1041 |
+
"step": 160000
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 8.66,
|
1045 |
+
"learning_rate": 6.206918868035536e-05,
|
1046 |
+
"loss": 3.0619,
|
1047 |
+
"step": 161000
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 8.71,
|
1051 |
+
"learning_rate": 6.177531329058069e-05,
|
1052 |
+
"loss": 3.0584,
|
1053 |
+
"step": 162000
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 8.76,
|
1057 |
+
"learning_rate": 6.148114373124669e-05,
|
1058 |
+
"loss": 3.0562,
|
1059 |
+
"step": 163000
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 8.82,
|
1063 |
+
"learning_rate": 6.118726834147203e-05,
|
1064 |
+
"loss": 3.0594,
|
1065 |
+
"step": 164000
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 8.87,
|
1069 |
+
"learning_rate": 6.089309878213802e-05,
|
1070 |
+
"loss": 3.0593,
|
1071 |
+
"step": 165000
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"epoch": 8.93,
|
1075 |
+
"learning_rate": 6.059922339236336e-05,
|
1076 |
+
"loss": 3.0619,
|
1077 |
+
"step": 166000
|
1078 |
+
},
|
1079 |
+
{
|
1080 |
+
"epoch": 8.98,
|
1081 |
+
"learning_rate": 6.030534800258869e-05,
|
1082 |
+
"loss": 3.0639,
|
1083 |
+
"step": 167000
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 9.0,
|
1087 |
+
"eval_accuracy": 0.40267046003453427,
|
1088 |
+
"eval_loss": 3.40413498878479,
|
1089 |
+
"eval_runtime": 154.0554,
|
1090 |
+
"eval_samples_per_second": 375.956,
|
1091 |
+
"eval_steps_per_second": 5.875,
|
1092 |
+
"step": 167373
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 9.03,
|
1096 |
+
"learning_rate": 6.001117844325469e-05,
|
1097 |
+
"loss": 3.0252,
|
1098 |
+
"step": 168000
|
1099 |
+
},
|
1100 |
+
{
|
1101 |
+
"epoch": 9.09,
|
1102 |
+
"learning_rate": 5.97170088839207e-05,
|
1103 |
+
"loss": 3.0072,
|
1104 |
+
"step": 169000
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 9.14,
|
1108 |
+
"learning_rate": 5.94228393245867e-05,
|
1109 |
+
"loss": 3.0135,
|
1110 |
+
"step": 170000
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 9.2,
|
1114 |
+
"learning_rate": 5.912896393481203e-05,
|
1115 |
+
"loss": 3.0186,
|
1116 |
+
"step": 171000
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 9.25,
|
1120 |
+
"learning_rate": 5.8835088545037366e-05,
|
1121 |
+
"loss": 3.0199,
|
1122 |
+
"step": 172000
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 9.3,
|
1126 |
+
"learning_rate": 5.8540918985703364e-05,
|
1127 |
+
"loss": 3.0205,
|
1128 |
+
"step": 173000
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 9.36,
|
1132 |
+
"learning_rate": 5.824674942636936e-05,
|
1133 |
+
"loss": 3.0232,
|
1134 |
+
"step": 174000
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 9.41,
|
1138 |
+
"learning_rate": 5.795257986703536e-05,
|
1139 |
+
"loss": 3.0294,
|
1140 |
+
"step": 175000
|
1141 |
+
},
|
1142 |
+
{
|
1143 |
+
"epoch": 9.46,
|
1144 |
+
"learning_rate": 5.7658704477260696e-05,
|
1145 |
+
"loss": 3.0248,
|
1146 |
+
"step": 176000
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 9.52,
|
1150 |
+
"learning_rate": 5.7364829087486024e-05,
|
1151 |
+
"loss": 3.0275,
|
1152 |
+
"step": 177000
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 9.57,
|
1156 |
+
"learning_rate": 5.707065952815203e-05,
|
1157 |
+
"loss": 3.0257,
|
1158 |
+
"step": 178000
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 9.63,
|
1162 |
+
"learning_rate": 5.677678413837736e-05,
|
1163 |
+
"loss": 3.0285,
|
1164 |
+
"step": 179000
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 9.68,
|
1168 |
+
"learning_rate": 5.6482614579043356e-05,
|
1169 |
+
"loss": 3.0275,
|
1170 |
+
"step": 180000
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 9.73,
|
1174 |
+
"learning_rate": 5.618844501970937e-05,
|
1175 |
+
"loss": 3.0305,
|
1176 |
+
"step": 181000
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 9.79,
|
1180 |
+
"learning_rate": 5.5894275460375366e-05,
|
1181 |
+
"loss": 3.0357,
|
1182 |
+
"step": 182000
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 9.84,
|
1186 |
+
"learning_rate": 5.56004000706007e-05,
|
1187 |
+
"loss": 3.0356,
|
1188 |
+
"step": 183000
|
1189 |
+
},
|
1190 |
+
{
|
1191 |
+
"epoch": 9.89,
|
1192 |
+
"learning_rate": 5.53062305112667e-05,
|
1193 |
+
"loss": 3.0363,
|
1194 |
+
"step": 184000
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 9.95,
|
1198 |
+
"learning_rate": 5.5012355121492035e-05,
|
1199 |
+
"loss": 3.0329,
|
1200 |
+
"step": 185000
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 10.0,
|
1204 |
+
"eval_accuracy": 0.4029401960076926,
|
1205 |
+
"eval_loss": 3.4231081008911133,
|
1206 |
+
"eval_runtime": 153.3432,
|
1207 |
+
"eval_samples_per_second": 377.702,
|
1208 |
+
"eval_steps_per_second": 5.902,
|
1209 |
+
"step": 185970
|
1210 |
+
},
|
1211 |
+
{
|
1212 |
+
"epoch": 10.0,
|
1213 |
+
"learning_rate": 5.471818556215803e-05,
|
1214 |
+
"loss": 3.0316,
|
1215 |
+
"step": 186000
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 10.06,
|
1219 |
+
"learning_rate": 5.442431017238336e-05,
|
1220 |
+
"loss": 2.9811,
|
1221 |
+
"step": 187000
|
1222 |
+
},
|
1223 |
+
{
|
1224 |
+
"epoch": 10.11,
|
1225 |
+
"learning_rate": 5.413014061304936e-05,
|
1226 |
+
"loss": 2.9815,
|
1227 |
+
"step": 188000
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 10.16,
|
1231 |
+
"learning_rate": 5.383597105371536e-05,
|
1232 |
+
"loss": 2.9895,
|
1233 |
+
"step": 189000
|
1234 |
+
},
|
1235 |
+
{
|
1236 |
+
"epoch": 10.22,
|
1237 |
+
"learning_rate": 5.3542095663940694e-05,
|
1238 |
+
"loss": 2.9893,
|
1239 |
+
"step": 190000
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 10.27,
|
1243 |
+
"learning_rate": 5.324792610460669e-05,
|
1244 |
+
"loss": 2.9913,
|
1245 |
+
"step": 191000
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 10.32,
|
1249 |
+
"learning_rate": 5.295375654527269e-05,
|
1250 |
+
"loss": 2.9966,
|
1251 |
+
"step": 192000
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 10.38,
|
1255 |
+
"learning_rate": 5.26595869859387e-05,
|
1256 |
+
"loss": 2.9958,
|
1257 |
+
"step": 193000
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 10.43,
|
1261 |
+
"learning_rate": 5.236571159616404e-05,
|
1262 |
+
"loss": 3.0016,
|
1263 |
+
"step": 194000
|
1264 |
+
},
|
1265 |
+
{
|
1266 |
+
"epoch": 10.49,
|
1267 |
+
"learning_rate": 5.2071542036830035e-05,
|
1268 |
+
"loss": 3.0028,
|
1269 |
+
"step": 195000
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 10.54,
|
1273 |
+
"learning_rate": 5.1777372477496033e-05,
|
1274 |
+
"loss": 3.005,
|
1275 |
+
"step": 196000
|
1276 |
+
},
|
1277 |
+
{
|
1278 |
+
"epoch": 10.59,
|
1279 |
+
"learning_rate": 5.14837912572807e-05,
|
1280 |
+
"loss": 2.9997,
|
1281 |
+
"step": 197000
|
1282 |
+
},
|
1283 |
+
{
|
1284 |
+
"epoch": 10.65,
|
1285 |
+
"learning_rate": 5.11896216979467e-05,
|
1286 |
+
"loss": 3.0042,
|
1287 |
+
"step": 198000
|
1288 |
+
},
|
1289 |
+
{
|
1290 |
+
"epoch": 10.7,
|
1291 |
+
"learning_rate": 5.0895452138612696e-05,
|
1292 |
+
"loss": 3.01,
|
1293 |
+
"step": 199000
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 10.75,
|
1297 |
+
"learning_rate": 5.060157674883803e-05,
|
1298 |
+
"loss": 3.0098,
|
1299 |
+
"step": 200000
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 10.81,
|
1303 |
+
"learning_rate": 5.030740718950403e-05,
|
1304 |
+
"loss": 3.0122,
|
1305 |
+
"step": 201000
|
1306 |
+
},
|
1307 |
+
{
|
1308 |
+
"epoch": 10.86,
|
1309 |
+
"learning_rate": 5.001323763017003e-05,
|
1310 |
+
"loss": 3.008,
|
1311 |
+
"step": 202000
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 10.92,
|
1315 |
+
"learning_rate": 4.971936224039537e-05,
|
1316 |
+
"loss": 3.0088,
|
1317 |
+
"step": 203000
|
1318 |
+
},
|
1319 |
+
{
|
1320 |
+
"epoch": 10.97,
|
1321 |
+
"learning_rate": 4.942519268106137e-05,
|
1322 |
+
"loss": 3.0093,
|
1323 |
+
"step": 204000
|
1324 |
+
},
|
1325 |
+
{
|
1326 |
+
"epoch": 11.0,
|
1327 |
+
"eval_accuracy": 0.40446972998051856,
|
1328 |
+
"eval_loss": 3.410041093826294,
|
1329 |
+
"eval_runtime": 153.1573,
|
1330 |
+
"eval_samples_per_second": 378.16,
|
1331 |
+
"eval_steps_per_second": 5.909,
|
1332 |
+
"step": 204567
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 11.02,
|
1336 |
+
"learning_rate": 4.9131023121727366e-05,
|
1337 |
+
"loss": 2.9805,
|
1338 |
+
"step": 205000
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 11.08,
|
1342 |
+
"learning_rate": 4.8836853562393364e-05,
|
1343 |
+
"loss": 2.9595,
|
1344 |
+
"step": 206000
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 11.13,
|
1348 |
+
"learning_rate": 4.854268400305936e-05,
|
1349 |
+
"loss": 2.9631,
|
1350 |
+
"step": 207000
|
1351 |
+
},
|
1352 |
+
{
|
1353 |
+
"epoch": 11.18,
|
1354 |
+
"learning_rate": 4.8248808613284705e-05,
|
1355 |
+
"loss": 2.9678,
|
1356 |
+
"step": 208000
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 11.24,
|
1360 |
+
"learning_rate": 4.79546390539507e-05,
|
1361 |
+
"loss": 2.9658,
|
1362 |
+
"step": 209000
|
1363 |
+
},
|
1364 |
+
{
|
1365 |
+
"epoch": 11.29,
|
1366 |
+
"learning_rate": 4.766076366417603e-05,
|
1367 |
+
"loss": 2.9768,
|
1368 |
+
"step": 210000
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 11.35,
|
1372 |
+
"learning_rate": 4.736659410484203e-05,
|
1373 |
+
"loss": 2.9732,
|
1374 |
+
"step": 211000
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 11.4,
|
1378 |
+
"learning_rate": 4.7072424545508035e-05,
|
1379 |
+
"loss": 2.9822,
|
1380 |
+
"step": 212000
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 11.45,
|
1384 |
+
"learning_rate": 4.6778549155733364e-05,
|
1385 |
+
"loss": 2.9789,
|
1386 |
+
"step": 213000
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 11.51,
|
1390 |
+
"learning_rate": 4.648437959639937e-05,
|
1391 |
+
"loss": 2.9786,
|
1392 |
+
"step": 214000
|
1393 |
+
},
|
1394 |
+
{
|
1395 |
+
"epoch": 11.56,
|
1396 |
+
"learning_rate": 4.6190210037065367e-05,
|
1397 |
+
"loss": 2.9792,
|
1398 |
+
"step": 215000
|
1399 |
+
},
|
1400 |
+
{
|
1401 |
+
"epoch": 11.61,
|
1402 |
+
"learning_rate": 4.5896040477731365e-05,
|
1403 |
+
"loss": 2.983,
|
1404 |
+
"step": 216000
|
1405 |
+
},
|
1406 |
+
{
|
1407 |
+
"epoch": 11.67,
|
1408 |
+
"learning_rate": 4.560187091839736e-05,
|
1409 |
+
"loss": 2.9819,
|
1410 |
+
"step": 217000
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 11.72,
|
1414 |
+
"learning_rate": 4.53079955286227e-05,
|
1415 |
+
"loss": 2.9833,
|
1416 |
+
"step": 218000
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 11.78,
|
1420 |
+
"learning_rate": 4.5013825969288696e-05,
|
1421 |
+
"loss": 2.9847,
|
1422 |
+
"step": 219000
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 11.83,
|
1426 |
+
"learning_rate": 4.471995057951404e-05,
|
1427 |
+
"loss": 2.9858,
|
1428 |
+
"step": 220000
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 11.88,
|
1432 |
+
"learning_rate": 4.442607518973937e-05,
|
1433 |
+
"loss": 2.9812,
|
1434 |
+
"step": 221000
|
1435 |
+
},
|
1436 |
+
{
|
1437 |
+
"epoch": 11.94,
|
1438 |
+
"learning_rate": 4.4131905630405366e-05,
|
1439 |
+
"loss": 2.9871,
|
1440 |
+
"step": 222000
|
1441 |
+
},
|
1442 |
+
{
|
1443 |
+
"epoch": 11.99,
|
1444 |
+
"learning_rate": 4.3837736071071364e-05,
|
1445 |
+
"loss": 2.9859,
|
1446 |
+
"step": 223000
|
1447 |
+
},
|
1448 |
+
{
|
1449 |
+
"epoch": 12.0,
|
1450 |
+
"eval_accuracy": 0.40492038724277163,
|
1451 |
+
"eval_loss": 3.409681797027588,
|
1452 |
+
"eval_runtime": 153.4512,
|
1453 |
+
"eval_samples_per_second": 377.436,
|
1454 |
+
"eval_steps_per_second": 5.898,
|
1455 |
+
"step": 223164
|
1456 |
+
},
|
1457 |
+
{
|
1458 |
+
"epoch": 12.04,
|
1459 |
+
"learning_rate": 4.354356651173737e-05,
|
1460 |
+
"loss": 2.943,
|
1461 |
+
"step": 224000
|
1462 |
+
},
|
1463 |
+
{
|
1464 |
+
"epoch": 12.1,
|
1465 |
+
"learning_rate": 4.32496911219627e-05,
|
1466 |
+
"loss": 2.9399,
|
1467 |
+
"step": 225000
|
1468 |
+
},
|
1469 |
+
{
|
1470 |
+
"epoch": 12.15,
|
1471 |
+
"learning_rate": 4.29555215626287e-05,
|
1472 |
+
"loss": 2.9462,
|
1473 |
+
"step": 226000
|
1474 |
+
},
|
1475 |
+
{
|
1476 |
+
"epoch": 12.21,
|
1477 |
+
"learning_rate": 4.26613520032947e-05,
|
1478 |
+
"loss": 2.9433,
|
1479 |
+
"step": 227000
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 12.26,
|
1483 |
+
"learning_rate": 4.23671824439607e-05,
|
1484 |
+
"loss": 2.9478,
|
1485 |
+
"step": 228000
|
1486 |
+
},
|
1487 |
+
{
|
1488 |
+
"epoch": 12.31,
|
1489 |
+
"learning_rate": 4.2073012884626704e-05,
|
1490 |
+
"loss": 2.9522,
|
1491 |
+
"step": 229000
|
1492 |
+
},
|
1493 |
+
{
|
1494 |
+
"epoch": 12.37,
|
1495 |
+
"learning_rate": 4.177913749485203e-05,
|
1496 |
+
"loss": 2.9532,
|
1497 |
+
"step": 230000
|
1498 |
+
},
|
1499 |
+
{
|
1500 |
+
"epoch": 12.42,
|
1501 |
+
"learning_rate": 4.148526210507737e-05,
|
1502 |
+
"loss": 2.9551,
|
1503 |
+
"step": 231000
|
1504 |
+
},
|
1505 |
+
{
|
1506 |
+
"epoch": 12.48,
|
1507 |
+
"learning_rate": 4.119109254574337e-05,
|
1508 |
+
"loss": 2.961,
|
1509 |
+
"step": 232000
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 12.53,
|
1513 |
+
"learning_rate": 4.089692298640937e-05,
|
1514 |
+
"loss": 2.9528,
|
1515 |
+
"step": 233000
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 12.58,
|
1519 |
+
"learning_rate": 4.060275342707537e-05,
|
1520 |
+
"loss": 2.9639,
|
1521 |
+
"step": 234000
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 12.64,
|
1525 |
+
"learning_rate": 4.0308878037300705e-05,
|
1526 |
+
"loss": 2.9596,
|
1527 |
+
"step": 235000
|
1528 |
+
},
|
1529 |
+
{
|
1530 |
+
"epoch": 12.69,
|
1531 |
+
"learning_rate": 4.00147084779667e-05,
|
1532 |
+
"loss": 2.9624,
|
1533 |
+
"step": 236000
|
1534 |
+
},
|
1535 |
+
{
|
1536 |
+
"epoch": 12.74,
|
1537 |
+
"learning_rate": 3.972083308819203e-05,
|
1538 |
+
"loss": 2.9609,
|
1539 |
+
"step": 237000
|
1540 |
+
},
|
1541 |
+
{
|
1542 |
+
"epoch": 12.8,
|
1543 |
+
"learning_rate": 3.9426663528858036e-05,
|
1544 |
+
"loss": 2.9616,
|
1545 |
+
"step": 238000
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 12.85,
|
1549 |
+
"learning_rate": 3.9132493969524034e-05,
|
1550 |
+
"loss": 2.9694,
|
1551 |
+
"step": 239000
|
1552 |
+
},
|
1553 |
+
{
|
1554 |
+
"epoch": 12.91,
|
1555 |
+
"learning_rate": 3.88389127493087e-05,
|
1556 |
+
"loss": 2.9629,
|
1557 |
+
"step": 240000
|
1558 |
+
},
|
1559 |
+
{
|
1560 |
+
"epoch": 12.96,
|
1561 |
+
"learning_rate": 3.8544743189974706e-05,
|
1562 |
+
"loss": 2.9662,
|
1563 |
+
"step": 241000
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 13.0,
|
1567 |
+
"eval_accuracy": 0.4052818267285978,
|
1568 |
+
"eval_loss": 3.4042649269104004,
|
1569 |
+
"eval_runtime": 153.6035,
|
1570 |
+
"eval_samples_per_second": 377.062,
|
1571 |
+
"eval_steps_per_second": 5.892,
|
1572 |
+
"step": 241761
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 13.01,
|
1576 |
+
"learning_rate": 3.8250573630640704e-05,
|
1577 |
+
"loss": 2.9532,
|
1578 |
+
"step": 242000
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 13.07,
|
1582 |
+
"learning_rate": 3.79564040713067e-05,
|
1583 |
+
"loss": 2.9159,
|
1584 |
+
"step": 243000
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 13.12,
|
1588 |
+
"learning_rate": 3.766252868153204e-05,
|
1589 |
+
"loss": 2.9206,
|
1590 |
+
"step": 244000
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 13.17,
|
1594 |
+
"learning_rate": 3.7368359122198036e-05,
|
1595 |
+
"loss": 2.9259,
|
1596 |
+
"step": 245000
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 13.23,
|
1600 |
+
"learning_rate": 3.707448373242337e-05,
|
1601 |
+
"loss": 2.9251,
|
1602 |
+
"step": 246000
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 13.28,
|
1606 |
+
"learning_rate": 3.678060834264871e-05,
|
1607 |
+
"loss": 2.9312,
|
1608 |
+
"step": 247000
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 13.34,
|
1612 |
+
"learning_rate": 3.6486438783314705e-05,
|
1613 |
+
"loss": 2.9362,
|
1614 |
+
"step": 248000
|
1615 |
+
},
|
1616 |
+
{
|
1617 |
+
"epoch": 13.39,
|
1618 |
+
"learning_rate": 3.61922692239807e-05,
|
1619 |
+
"loss": 2.932,
|
1620 |
+
"step": 249000
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 13.44,
|
1624 |
+
"learning_rate": 3.589809966464671e-05,
|
1625 |
+
"loss": 2.9368,
|
1626 |
+
"step": 250000
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 13.5,
|
1630 |
+
"learning_rate": 3.5603930105312706e-05,
|
1631 |
+
"loss": 2.9376,
|
1632 |
+
"step": 251000
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 13.55,
|
1636 |
+
"learning_rate": 3.531034888509737e-05,
|
1637 |
+
"loss": 2.938,
|
1638 |
+
"step": 252000
|
1639 |
+
},
|
1640 |
+
{
|
1641 |
+
"epoch": 13.6,
|
1642 |
+
"learning_rate": 3.501617932576337e-05,
|
1643 |
+
"loss": 2.942,
|
1644 |
+
"step": 253000
|
1645 |
+
},
|
1646 |
+
{
|
1647 |
+
"epoch": 13.66,
|
1648 |
+
"learning_rate": 3.472200976642937e-05,
|
1649 |
+
"loss": 2.9405,
|
1650 |
+
"step": 254000
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 13.71,
|
1654 |
+
"learning_rate": 3.4427840207095367e-05,
|
1655 |
+
"loss": 2.9403,
|
1656 |
+
"step": 255000
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 13.77,
|
1660 |
+
"learning_rate": 3.413396481732071e-05,
|
1661 |
+
"loss": 2.9459,
|
1662 |
+
"step": 256000
|
1663 |
+
},
|
1664 |
+
{
|
1665 |
+
"epoch": 13.82,
|
1666 |
+
"learning_rate": 3.383979525798671e-05,
|
1667 |
+
"loss": 2.9432,
|
1668 |
+
"step": 257000
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 13.87,
|
1672 |
+
"learning_rate": 3.3545625698652705e-05,
|
1673 |
+
"loss": 2.9483,
|
1674 |
+
"step": 258000
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 13.93,
|
1678 |
+
"learning_rate": 3.325175030887804e-05,
|
1679 |
+
"loss": 2.9437,
|
1680 |
+
"step": 259000
|
1681 |
+
},
|
1682 |
+
{
|
1683 |
+
"epoch": 13.98,
|
1684 |
+
"learning_rate": 3.295758074954404e-05,
|
1685 |
+
"loss": 2.9424,
|
1686 |
+
"step": 260000
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 14.0,
|
1690 |
+
"eval_accuracy": 0.4056935855777852,
|
1691 |
+
"eval_loss": 3.4045979976654053,
|
1692 |
+
"eval_runtime": 153.3807,
|
1693 |
+
"eval_samples_per_second": 377.61,
|
1694 |
+
"eval_steps_per_second": 5.9,
|
1695 |
+
"step": 260358
|
1696 |
+
},
|
1697 |
+
{
|
1698 |
+
"epoch": 14.03,
|
1699 |
+
"learning_rate": 3.2663705359769374e-05,
|
1700 |
+
"loss": 2.9161,
|
1701 |
+
"step": 261000
|
1702 |
+
},
|
1703 |
+
{
|
1704 |
+
"epoch": 14.09,
|
1705 |
+
"learning_rate": 3.236953580043537e-05,
|
1706 |
+
"loss": 2.9053,
|
1707 |
+
"step": 262000
|
1708 |
+
},
|
1709 |
+
{
|
1710 |
+
"epoch": 14.14,
|
1711 |
+
"learning_rate": 3.207566041066071e-05,
|
1712 |
+
"loss": 2.9064,
|
1713 |
+
"step": 263000
|
1714 |
+
},
|
1715 |
+
{
|
1716 |
+
"epoch": 14.2,
|
1717 |
+
"learning_rate": 3.1781490851326706e-05,
|
1718 |
+
"loss": 2.9123,
|
1719 |
+
"step": 264000
|
1720 |
+
},
|
1721 |
+
{
|
1722 |
+
"epoch": 14.25,
|
1723 |
+
"learning_rate": 3.148761546155204e-05,
|
1724 |
+
"loss": 2.9084,
|
1725 |
+
"step": 265000
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 14.3,
|
1729 |
+
"learning_rate": 3.119344590221804e-05,
|
1730 |
+
"loss": 2.9113,
|
1731 |
+
"step": 266000
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 14.36,
|
1735 |
+
"learning_rate": 3.0899570512443376e-05,
|
1736 |
+
"loss": 2.913,
|
1737 |
+
"step": 267000
|
1738 |
+
},
|
1739 |
+
{
|
1740 |
+
"epoch": 14.41,
|
1741 |
+
"learning_rate": 3.060569512266871e-05,
|
1742 |
+
"loss": 2.921,
|
1743 |
+
"step": 268000
|
1744 |
+
},
|
1745 |
+
{
|
1746 |
+
"epoch": 14.46,
|
1747 |
+
"learning_rate": 3.031152556333471e-05,
|
1748 |
+
"loss": 2.9186,
|
1749 |
+
"step": 269000
|
1750 |
+
},
|
1751 |
+
{
|
1752 |
+
"epoch": 14.52,
|
1753 |
+
"learning_rate": 3.0017356004000707e-05,
|
1754 |
+
"loss": 2.9163,
|
1755 |
+
"step": 270000
|
1756 |
+
},
|
1757 |
+
{
|
1758 |
+
"epoch": 14.57,
|
1759 |
+
"learning_rate": 2.9723186444666705e-05,
|
1760 |
+
"loss": 2.9189,
|
1761 |
+
"step": 271000
|
1762 |
+
},
|
1763 |
+
{
|
1764 |
+
"epoch": 14.63,
|
1765 |
+
"learning_rate": 2.9429311054892038e-05,
|
1766 |
+
"loss": 2.9215,
|
1767 |
+
"step": 272000
|
1768 |
+
},
|
1769 |
+
{
|
1770 |
+
"epoch": 14.68,
|
1771 |
+
"learning_rate": 2.9135141495558042e-05,
|
1772 |
+
"loss": 2.9208,
|
1773 |
+
"step": 273000
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 14.73,
|
1777 |
+
"learning_rate": 2.8841266105783378e-05,
|
1778 |
+
"loss": 2.9262,
|
1779 |
+
"step": 274000
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 14.79,
|
1783 |
+
"learning_rate": 2.8547096546449376e-05,
|
1784 |
+
"loss": 2.9249,
|
1785 |
+
"step": 275000
|
1786 |
+
},
|
1787 |
+
{
|
1788 |
+
"epoch": 14.84,
|
1789 |
+
"learning_rate": 2.8252926987115374e-05,
|
1790 |
+
"loss": 2.9224,
|
1791 |
+
"step": 276000
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 14.89,
|
1795 |
+
"learning_rate": 2.7959051597340706e-05,
|
1796 |
+
"loss": 2.9233,
|
1797 |
+
"step": 277000
|
1798 |
+
},
|
1799 |
+
{
|
1800 |
+
"epoch": 14.95,
|
1801 |
+
"learning_rate": 2.7664882038006708e-05,
|
1802 |
+
"loss": 2.928,
|
1803 |
+
"step": 278000
|
1804 |
+
},
|
1805 |
+
{
|
1806 |
+
"epoch": 15.0,
|
1807 |
+
"eval_accuracy": 0.40587793315196863,
|
1808 |
+
"eval_loss": 3.4078826904296875,
|
1809 |
+
"eval_runtime": 153.8126,
|
1810 |
+
"eval_samples_per_second": 376.549,
|
1811 |
+
"eval_steps_per_second": 5.884,
|
1812 |
+
"step": 278955
|
1813 |
+
},
|
1814 |
+
{
|
1815 |
+
"epoch": 15.0,
|
1816 |
+
"learning_rate": 2.7371006648232044e-05,
|
1817 |
+
"loss": 2.9273,
|
1818 |
+
"step": 279000
|
1819 |
+
},
|
1820 |
+
{
|
1821 |
+
"epoch": 15.06,
|
1822 |
+
"learning_rate": 2.7076837088898045e-05,
|
1823 |
+
"loss": 2.8835,
|
1824 |
+
"step": 280000
|
1825 |
+
},
|
1826 |
+
{
|
1827 |
+
"epoch": 15.11,
|
1828 |
+
"learning_rate": 2.678325586868271e-05,
|
1829 |
+
"loss": 2.8879,
|
1830 |
+
"step": 281000
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 15.16,
|
1834 |
+
"learning_rate": 2.648908630934871e-05,
|
1835 |
+
"loss": 2.8894,
|
1836 |
+
"step": 282000
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 15.22,
|
1840 |
+
"learning_rate": 2.6194916750014708e-05,
|
1841 |
+
"loss": 2.8945,
|
1842 |
+
"step": 283000
|
1843 |
+
},
|
1844 |
+
{
|
1845 |
+
"epoch": 15.27,
|
1846 |
+
"learning_rate": 2.590074719068071e-05,
|
1847 |
+
"loss": 2.8966,
|
1848 |
+
"step": 284000
|
1849 |
+
},
|
1850 |
+
{
|
1851 |
+
"epoch": 15.33,
|
1852 |
+
"learning_rate": 2.5606871800906045e-05,
|
1853 |
+
"loss": 2.8984,
|
1854 |
+
"step": 285000
|
1855 |
+
},
|
1856 |
+
{
|
1857 |
+
"epoch": 15.38,
|
1858 |
+
"learning_rate": 2.531299641113138e-05,
|
1859 |
+
"loss": 2.8989,
|
1860 |
+
"step": 286000
|
1861 |
+
},
|
1862 |
+
{
|
1863 |
+
"epoch": 15.43,
|
1864 |
+
"learning_rate": 2.501882685179738e-05,
|
1865 |
+
"loss": 2.9001,
|
1866 |
+
"step": 287000
|
1867 |
+
},
|
1868 |
+
{
|
1869 |
+
"epoch": 15.49,
|
1870 |
+
"learning_rate": 2.472495146202271e-05,
|
1871 |
+
"loss": 2.9013,
|
1872 |
+
"step": 288000
|
1873 |
+
},
|
1874 |
+
{
|
1875 |
+
"epoch": 15.54,
|
1876 |
+
"learning_rate": 2.443078190268871e-05,
|
1877 |
+
"loss": 2.9033,
|
1878 |
+
"step": 289000
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 15.59,
|
1882 |
+
"learning_rate": 2.413661234335471e-05,
|
1883 |
+
"loss": 2.9021,
|
1884 |
+
"step": 290000
|
1885 |
+
},
|
1886 |
+
{
|
1887 |
+
"epoch": 15.65,
|
1888 |
+
"learning_rate": 2.384244278402071e-05,
|
1889 |
+
"loss": 2.9055,
|
1890 |
+
"step": 291000
|
1891 |
+
},
|
1892 |
+
{
|
1893 |
+
"epoch": 15.7,
|
1894 |
+
"learning_rate": 2.354827322468671e-05,
|
1895 |
+
"loss": 2.904,
|
1896 |
+
"step": 292000
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 15.76,
|
1900 |
+
"learning_rate": 2.3254397834912045e-05,
|
1901 |
+
"loss": 2.9048,
|
1902 |
+
"step": 293000
|
1903 |
+
},
|
1904 |
+
{
|
1905 |
+
"epoch": 15.81,
|
1906 |
+
"learning_rate": 2.2960522445137378e-05,
|
1907 |
+
"loss": 2.9072,
|
1908 |
+
"step": 294000
|
1909 |
+
},
|
1910 |
+
{
|
1911 |
+
"epoch": 15.86,
|
1912 |
+
"learning_rate": 2.2666352885803376e-05,
|
1913 |
+
"loss": 2.9067,
|
1914 |
+
"step": 295000
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 15.92,
|
1918 |
+
"learning_rate": 2.2372183326469377e-05,
|
1919 |
+
"loss": 2.9076,
|
1920 |
+
"step": 296000
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 15.97,
|
1924 |
+
"learning_rate": 2.2078307936694713e-05,
|
1925 |
+
"loss": 2.908,
|
1926 |
+
"step": 297000
|
1927 |
+
},
|
1928 |
+
{
|
1929 |
+
"epoch": 16.0,
|
1930 |
+
"eval_accuracy": 0.4060550922456719,
|
1931 |
+
"eval_loss": 3.411921262741089,
|
1932 |
+
"eval_runtime": 153.4871,
|
1933 |
+
"eval_samples_per_second": 377.348,
|
1934 |
+
"eval_steps_per_second": 5.896,
|
1935 |
+
"step": 297552
|
1936 |
+
},
|
1937 |
+
{
|
1938 |
+
"epoch": 16.02,
|
1939 |
+
"learning_rate": 2.178413837736071e-05,
|
1940 |
+
"loss": 2.893,
|
1941 |
+
"step": 298000
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 16.08,
|
1945 |
+
"learning_rate": 2.1490262987586046e-05,
|
1946 |
+
"loss": 2.8702,
|
1947 |
+
"step": 299000
|
1948 |
+
},
|
1949 |
+
{
|
1950 |
+
"epoch": 16.13,
|
1951 |
+
"learning_rate": 2.1196093428252045e-05,
|
1952 |
+
"loss": 2.8782,
|
1953 |
+
"step": 300000
|
1954 |
+
},
|
1955 |
+
{
|
1956 |
+
"epoch": 16.19,
|
1957 |
+
"learning_rate": 2.0901923868918043e-05,
|
1958 |
+
"loss": 2.8769,
|
1959 |
+
"step": 301000
|
1960 |
+
},
|
1961 |
+
{
|
1962 |
+
"epoch": 16.24,
|
1963 |
+
"learning_rate": 2.0607754309584047e-05,
|
1964 |
+
"loss": 2.88,
|
1965 |
+
"step": 302000
|
1966 |
+
},
|
1967 |
+
{
|
1968 |
+
"epoch": 16.29,
|
1969 |
+
"learning_rate": 2.0313584750250046e-05,
|
1970 |
+
"loss": 2.8792,
|
1971 |
+
"step": 303000
|
1972 |
+
},
|
1973 |
+
{
|
1974 |
+
"epoch": 16.35,
|
1975 |
+
"learning_rate": 2.0019709360475378e-05,
|
1976 |
+
"loss": 2.8793,
|
1977 |
+
"step": 304000
|
1978 |
+
},
|
1979 |
+
{
|
1980 |
+
"epoch": 16.4,
|
1981 |
+
"learning_rate": 1.9725833970700713e-05,
|
1982 |
+
"loss": 2.8812,
|
1983 |
+
"step": 305000
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 16.45,
|
1987 |
+
"learning_rate": 1.943166441136671e-05,
|
1988 |
+
"loss": 2.8852,
|
1989 |
+
"step": 306000
|
1990 |
+
},
|
1991 |
+
{
|
1992 |
+
"epoch": 16.51,
|
1993 |
+
"learning_rate": 1.9137494852032713e-05,
|
1994 |
+
"loss": 2.8845,
|
1995 |
+
"step": 307000
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 16.56,
|
1999 |
+
"learning_rate": 1.8843325292698714e-05,
|
2000 |
+
"loss": 2.8871,
|
2001 |
+
"step": 308000
|
2002 |
+
},
|
2003 |
+
{
|
2004 |
+
"epoch": 16.62,
|
2005 |
+
"learning_rate": 1.8549155733364713e-05,
|
2006 |
+
"loss": 2.8845,
|
2007 |
+
"step": 309000
|
2008 |
+
},
|
2009 |
+
{
|
2010 |
+
"epoch": 16.67,
|
2011 |
+
"learning_rate": 1.8255280343590045e-05,
|
2012 |
+
"loss": 2.8865,
|
2013 |
+
"step": 310000
|
2014 |
+
},
|
2015 |
+
{
|
2016 |
+
"epoch": 16.72,
|
2017 |
+
"learning_rate": 1.7961110784256046e-05,
|
2018 |
+
"loss": 2.8871,
|
2019 |
+
"step": 311000
|
2020 |
+
},
|
2021 |
+
{
|
2022 |
+
"epoch": 16.78,
|
2023 |
+
"learning_rate": 1.7667529564040713e-05,
|
2024 |
+
"loss": 2.8893,
|
2025 |
+
"step": 312000
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 16.83,
|
2029 |
+
"learning_rate": 1.7373360004706714e-05,
|
2030 |
+
"loss": 2.8926,
|
2031 |
+
"step": 313000
|
2032 |
+
},
|
2033 |
+
{
|
2034 |
+
"epoch": 16.88,
|
2035 |
+
"learning_rate": 1.7079190445372716e-05,
|
2036 |
+
"loss": 2.8909,
|
2037 |
+
"step": 314000
|
2038 |
+
},
|
2039 |
+
{
|
2040 |
+
"epoch": 16.94,
|
2041 |
+
"learning_rate": 1.6785315055598048e-05,
|
2042 |
+
"loss": 2.8898,
|
2043 |
+
"step": 315000
|
2044 |
+
},
|
2045 |
+
{
|
2046 |
+
"epoch": 16.99,
|
2047 |
+
"learning_rate": 1.6491145496264046e-05,
|
2048 |
+
"loss": 2.8912,
|
2049 |
+
"step": 316000
|
2050 |
+
},
|
2051 |
+
{
|
2052 |
+
"epoch": 17.0,
|
2053 |
+
"eval_accuracy": 0.40615673870330293,
|
2054 |
+
"eval_loss": 3.4119112491607666,
|
2055 |
+
"eval_runtime": 153.7299,
|
2056 |
+
"eval_samples_per_second": 376.752,
|
2057 |
+
"eval_steps_per_second": 5.887,
|
2058 |
+
"step": 316149
|
2059 |
+
},
|
2060 |
+
{
|
2061 |
+
"epoch": 17.05,
|
2062 |
+
"learning_rate": 1.6196975936930047e-05,
|
2063 |
+
"loss": 2.8649,
|
2064 |
+
"step": 317000
|
2065 |
+
},
|
2066 |
+
{
|
2067 |
+
"epoch": 17.1,
|
2068 |
+
"learning_rate": 1.5902806377596045e-05,
|
2069 |
+
"loss": 2.8592,
|
2070 |
+
"step": 318000
|
2071 |
+
},
|
2072 |
+
{
|
2073 |
+
"epoch": 17.15,
|
2074 |
+
"learning_rate": 1.5608636818262047e-05,
|
2075 |
+
"loss": 2.8635,
|
2076 |
+
"step": 319000
|
2077 |
+
},
|
2078 |
+
{
|
2079 |
+
"epoch": 17.21,
|
2080 |
+
"learning_rate": 1.531446725892805e-05,
|
2081 |
+
"loss": 2.8652,
|
2082 |
+
"step": 320000
|
2083 |
+
},
|
2084 |
+
{
|
2085 |
+
"epoch": 17.26,
|
2086 |
+
"learning_rate": 1.502059186915338e-05,
|
2087 |
+
"loss": 2.8603,
|
2088 |
+
"step": 321000
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 17.31,
|
2092 |
+
"learning_rate": 1.4726422309819379e-05,
|
2093 |
+
"loss": 2.8673,
|
2094 |
+
"step": 322000
|
2095 |
+
},
|
2096 |
+
{
|
2097 |
+
"epoch": 17.37,
|
2098 |
+
"learning_rate": 1.4432546920044714e-05,
|
2099 |
+
"loss": 2.8641,
|
2100 |
+
"step": 323000
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 17.42,
|
2104 |
+
"learning_rate": 1.4138377360710714e-05,
|
2105 |
+
"loss": 2.8724,
|
2106 |
+
"step": 324000
|
2107 |
+
},
|
2108 |
+
{
|
2109 |
+
"epoch": 17.48,
|
2110 |
+
"learning_rate": 1.3844207801376712e-05,
|
2111 |
+
"loss": 2.8704,
|
2112 |
+
"step": 325000
|
2113 |
+
},
|
2114 |
+
{
|
2115 |
+
"epoch": 17.53,
|
2116 |
+
"learning_rate": 1.3550626581161382e-05,
|
2117 |
+
"loss": 2.8711,
|
2118 |
+
"step": 326000
|
2119 |
+
},
|
2120 |
+
{
|
2121 |
+
"epoch": 17.58,
|
2122 |
+
"learning_rate": 1.3256457021827382e-05,
|
2123 |
+
"loss": 2.8714,
|
2124 |
+
"step": 327000
|
2125 |
+
},
|
2126 |
+
{
|
2127 |
+
"epoch": 17.64,
|
2128 |
+
"learning_rate": 1.296228746249338e-05,
|
2129 |
+
"loss": 2.8715,
|
2130 |
+
"step": 328000
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 17.69,
|
2134 |
+
"learning_rate": 1.2668117903159383e-05,
|
2135 |
+
"loss": 2.8702,
|
2136 |
+
"step": 329000
|
2137 |
+
},
|
2138 |
+
{
|
2139 |
+
"epoch": 17.74,
|
2140 |
+
"learning_rate": 1.2373948343825381e-05,
|
2141 |
+
"loss": 2.8737,
|
2142 |
+
"step": 330000
|
2143 |
+
},
|
2144 |
+
{
|
2145 |
+
"epoch": 17.8,
|
2146 |
+
"learning_rate": 1.2080072954050715e-05,
|
2147 |
+
"loss": 2.8738,
|
2148 |
+
"step": 331000
|
2149 |
+
},
|
2150 |
+
{
|
2151 |
+
"epoch": 17.85,
|
2152 |
+
"learning_rate": 1.1785903394716716e-05,
|
2153 |
+
"loss": 2.8758,
|
2154 |
+
"step": 332000
|
2155 |
+
},
|
2156 |
+
{
|
2157 |
+
"epoch": 17.91,
|
2158 |
+
"learning_rate": 1.1492028004942049e-05,
|
2159 |
+
"loss": 2.8721,
|
2160 |
+
"step": 333000
|
2161 |
+
},
|
2162 |
+
{
|
2163 |
+
"epoch": 17.96,
|
2164 |
+
"learning_rate": 1.1197858445608048e-05,
|
2165 |
+
"loss": 2.8716,
|
2166 |
+
"step": 334000
|
2167 |
+
},
|
2168 |
+
{
|
2169 |
+
"epoch": 18.0,
|
2170 |
+
"eval_accuracy": 0.4063988628495701,
|
2171 |
+
"eval_loss": 3.415924549102783,
|
2172 |
+
"eval_runtime": 153.4402,
|
2173 |
+
"eval_samples_per_second": 377.463,
|
2174 |
+
"eval_steps_per_second": 5.898,
|
2175 |
+
"step": 334746
|
2176 |
+
},
|
2177 |
+
{
|
2178 |
+
"epoch": 18.01,
|
2179 |
+
"learning_rate": 1.0904277225392716e-05,
|
2180 |
+
"loss": 2.8667,
|
2181 |
+
"step": 335000
|
2182 |
+
},
|
2183 |
+
{
|
2184 |
+
"epoch": 18.07,
|
2185 |
+
"learning_rate": 1.0610107666058716e-05,
|
2186 |
+
"loss": 2.8473,
|
2187 |
+
"step": 336000
|
2188 |
+
},
|
2189 |
+
{
|
2190 |
+
"epoch": 18.12,
|
2191 |
+
"learning_rate": 1.0315938106724716e-05,
|
2192 |
+
"loss": 2.8493,
|
2193 |
+
"step": 337000
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 18.17,
|
2197 |
+
"learning_rate": 1.0021768547390718e-05,
|
2198 |
+
"loss": 2.8574,
|
2199 |
+
"step": 338000
|
2200 |
+
},
|
2201 |
+
{
|
2202 |
+
"epoch": 18.23,
|
2203 |
+
"learning_rate": 9.727598988056716e-06,
|
2204 |
+
"loss": 2.8522,
|
2205 |
+
"step": 339000
|
2206 |
+
},
|
2207 |
+
{
|
2208 |
+
"epoch": 18.28,
|
2209 |
+
"learning_rate": 9.43372359828205e-06,
|
2210 |
+
"loss": 2.8504,
|
2211 |
+
"step": 340000
|
2212 |
+
},
|
2213 |
+
{
|
2214 |
+
"epoch": 18.34,
|
2215 |
+
"learning_rate": 9.139554038948051e-06,
|
2216 |
+
"loss": 2.854,
|
2217 |
+
"step": 341000
|
2218 |
+
},
|
2219 |
+
{
|
2220 |
+
"epoch": 18.39,
|
2221 |
+
"learning_rate": 8.845384479614049e-06,
|
2222 |
+
"loss": 2.8497,
|
2223 |
+
"step": 342000
|
2224 |
+
},
|
2225 |
+
{
|
2226 |
+
"epoch": 18.44,
|
2227 |
+
"learning_rate": 8.551509089839383e-06,
|
2228 |
+
"loss": 2.8555,
|
2229 |
+
"step": 343000
|
2230 |
+
},
|
2231 |
+
{
|
2232 |
+
"epoch": 18.5,
|
2233 |
+
"learning_rate": 8.257339530505384e-06,
|
2234 |
+
"loss": 2.8554,
|
2235 |
+
"step": 344000
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 18.55,
|
2239 |
+
"learning_rate": 7.963464140730717e-06,
|
2240 |
+
"loss": 2.8551,
|
2241 |
+
"step": 345000
|
2242 |
+
},
|
2243 |
+
{
|
2244 |
+
"epoch": 18.61,
|
2245 |
+
"learning_rate": 7.669294581396718e-06,
|
2246 |
+
"loss": 2.8556,
|
2247 |
+
"step": 346000
|
2248 |
+
},
|
2249 |
+
{
|
2250 |
+
"epoch": 18.66,
|
2251 |
+
"learning_rate": 7.375125022062718e-06,
|
2252 |
+
"loss": 2.8563,
|
2253 |
+
"step": 347000
|
2254 |
+
},
|
2255 |
+
{
|
2256 |
+
"epoch": 18.71,
|
2257 |
+
"learning_rate": 7.080955462728717e-06,
|
2258 |
+
"loss": 2.8591,
|
2259 |
+
"step": 348000
|
2260 |
+
},
|
2261 |
+
{
|
2262 |
+
"epoch": 18.77,
|
2263 |
+
"learning_rate": 6.786785903394717e-06,
|
2264 |
+
"loss": 2.8552,
|
2265 |
+
"step": 349000
|
2266 |
+
},
|
2267 |
+
{
|
2268 |
+
"epoch": 18.82,
|
2269 |
+
"learning_rate": 6.4929105136200514e-06,
|
2270 |
+
"loss": 2.8633,
|
2271 |
+
"step": 350000
|
2272 |
+
},
|
2273 |
+
{
|
2274 |
+
"epoch": 18.87,
|
2275 |
+
"learning_rate": 6.198740954286051e-06,
|
2276 |
+
"loss": 2.8542,
|
2277 |
+
"step": 351000
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 18.93,
|
2281 |
+
"learning_rate": 5.904865564511384e-06,
|
2282 |
+
"loss": 2.8612,
|
2283 |
+
"step": 352000
|
2284 |
+
},
|
2285 |
+
{
|
2286 |
+
"epoch": 18.98,
|
2287 |
+
"learning_rate": 5.610990174736719e-06,
|
2288 |
+
"loss": 2.8589,
|
2289 |
+
"step": 353000
|
2290 |
+
},
|
2291 |
+
{
|
2292 |
+
"epoch": 19.0,
|
2293 |
+
"eval_accuracy": 0.406526172854336,
|
2294 |
+
"eval_loss": 3.422281503677368,
|
2295 |
+
"eval_runtime": 153.5913,
|
2296 |
+
"eval_samples_per_second": 377.092,
|
2297 |
+
"eval_steps_per_second": 5.892,
|
2298 |
+
"step": 353343
|
2299 |
+
},
|
2300 |
+
{
|
2301 |
+
"epoch": 19.04,
|
2302 |
+
"learning_rate": 5.316820615402719e-06,
|
2303 |
+
"loss": 2.8448,
|
2304 |
+
"step": 354000
|
2305 |
+
},
|
2306 |
+
{
|
2307 |
+
"epoch": 19.09,
|
2308 |
+
"learning_rate": 5.022945225628052e-06,
|
2309 |
+
"loss": 2.8385,
|
2310 |
+
"step": 355000
|
2311 |
+
},
|
2312 |
+
{
|
2313 |
+
"epoch": 19.14,
|
2314 |
+
"learning_rate": 4.7287756662940525e-06,
|
2315 |
+
"loss": 2.8445,
|
2316 |
+
"step": 356000
|
2317 |
+
},
|
2318 |
+
{
|
2319 |
+
"epoch": 19.2,
|
2320 |
+
"learning_rate": 4.434606106960052e-06,
|
2321 |
+
"loss": 2.8427,
|
2322 |
+
"step": 357000
|
2323 |
+
},
|
2324 |
+
{
|
2325 |
+
"epoch": 19.25,
|
2326 |
+
"learning_rate": 4.140436547626052e-06,
|
2327 |
+
"loss": 2.8386,
|
2328 |
+
"step": 358000
|
2329 |
+
},
|
2330 |
+
{
|
2331 |
+
"epoch": 19.3,
|
2332 |
+
"learning_rate": 3.846266988292051e-06,
|
2333 |
+
"loss": 2.8364,
|
2334 |
+
"step": 359000
|
2335 |
+
},
|
2336 |
+
{
|
2337 |
+
"epoch": 19.36,
|
2338 |
+
"learning_rate": 3.552391598517386e-06,
|
2339 |
+
"loss": 2.8417,
|
2340 |
+
"step": 360000
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 19.41,
|
2344 |
+
"learning_rate": 3.258222039183385e-06,
|
2345 |
+
"loss": 2.8461,
|
2346 |
+
"step": 361000
|
2347 |
+
},
|
2348 |
+
{
|
2349 |
+
"epoch": 19.47,
|
2350 |
+
"learning_rate": 2.9640524798493854e-06,
|
2351 |
+
"loss": 2.8412,
|
2352 |
+
"step": 362000
|
2353 |
+
},
|
2354 |
+
{
|
2355 |
+
"epoch": 19.52,
|
2356 |
+
"learning_rate": 2.6704712596340534e-06,
|
2357 |
+
"loss": 2.8461,
|
2358 |
+
"step": 363000
|
2359 |
+
},
|
2360 |
+
{
|
2361 |
+
"epoch": 19.57,
|
2362 |
+
"learning_rate": 2.3763017003000528e-06,
|
2363 |
+
"loss": 2.8483,
|
2364 |
+
"step": 364000
|
2365 |
+
},
|
2366 |
+
{
|
2367 |
+
"epoch": 19.63,
|
2368 |
+
"learning_rate": 2.082132140966053e-06,
|
2369 |
+
"loss": 2.8418,
|
2370 |
+
"step": 365000
|
2371 |
+
},
|
2372 |
+
{
|
2373 |
+
"epoch": 19.68,
|
2374 |
+
"learning_rate": 1.787962581632053e-06,
|
2375 |
+
"loss": 2.842,
|
2376 |
+
"step": 366000
|
2377 |
+
},
|
2378 |
+
{
|
2379 |
+
"epoch": 19.73,
|
2380 |
+
"learning_rate": 1.4937930222980528e-06,
|
2381 |
+
"loss": 2.8432,
|
2382 |
+
"step": 367000
|
2383 |
+
},
|
2384 |
+
{
|
2385 |
+
"epoch": 19.79,
|
2386 |
+
"learning_rate": 1.1996234629640526e-06,
|
2387 |
+
"loss": 2.8444,
|
2388 |
+
"step": 368000
|
2389 |
+
},
|
2390 |
+
{
|
2391 |
+
"epoch": 19.84,
|
2392 |
+
"learning_rate": 9.060422427487204e-07,
|
2393 |
+
"loss": 2.8428,
|
2394 |
+
"step": 369000
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 19.9,
|
2398 |
+
"learning_rate": 6.118726834147203e-07,
|
2399 |
+
"loss": 2.8443,
|
2400 |
+
"step": 370000
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 19.95,
|
2404 |
+
"learning_rate": 3.1770312408072017e-07,
|
2405 |
+
"loss": 2.8424,
|
2406 |
+
"step": 371000
|
2407 |
+
},
|
2408 |
+
{
|
2409 |
+
"epoch": 20.0,
|
2410 |
+
"eval_accuracy": 0.40654968657553286,
|
2411 |
+
"eval_loss": 3.424689292907715,
|
2412 |
+
"eval_runtime": 153.6195,
|
2413 |
+
"eval_samples_per_second": 377.022,
|
2414 |
+
"eval_steps_per_second": 5.891,
|
2415 |
+
"step": 371940
|
2416 |
+
},
|
2417 |
+
{
|
2418 |
+
"epoch": 20.0,
|
2419 |
+
"step": 371940,
|
2420 |
+
"total_flos": 1.56702845389824e+18,
|
2421 |
+
"train_loss": 3.1534978556531716,
|
2422 |
+
"train_runtime": 80977.43,
|
2423 |
+
"train_samples_per_second": 146.973,
|
2424 |
+
"train_steps_per_second": 4.593
|
2425 |
+
}
|
2426 |
+
],
|
2427 |
+
"logging_steps": 1000,
|
2428 |
+
"max_steps": 371940,
|
2429 |
+
"num_input_tokens_seen": 0,
|
2430 |
+
"num_train_epochs": 20,
|
2431 |
+
"save_steps": 5000,
|
2432 |
+
"total_flos": 1.56702845389824e+18,
|
2433 |
+
"train_batch_size": 32,
|
2434 |
+
"trial_name": null,
|
2435 |
+
"trial_params": null
|
2436 |
+
}
|