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: cria-babylm2-subset-default-1e-3
|
| 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 |
# cria-babylm2-subset-default-1e-3
|
| 15 |
|
| 16 |
-
This model was trained from scratch on
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
- Loss: 2.6626
|
| 19 |
- Accuracy: 0.5204
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
+
datasets:
|
| 5 |
+
- kanishka/babylm2-subset
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
| 8 |
model-index:
|
| 9 |
- name: cria-babylm2-subset-default-1e-3
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: Causal Language Modeling
|
| 13 |
+
type: text-generation
|
| 14 |
+
dataset:
|
| 15 |
+
name: kanishka/babylm2-subset
|
| 16 |
+
type: kanishka/babylm2-subset
|
| 17 |
+
metrics:
|
| 18 |
+
- name: Accuracy
|
| 19 |
+
type: accuracy
|
| 20 |
+
value: 0.5203706477236009
|
| 21 |
---
|
| 22 |
|
| 23 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 25 |
|
| 26 |
# cria-babylm2-subset-default-1e-3
|
| 27 |
|
| 28 |
+
This model was trained from scratch on the kanishka/babylm2-subset dataset.
|
| 29 |
It achieves the following results on the evaluation set:
|
| 30 |
- Loss: 2.6626
|
| 31 |
- Accuracy: 0.5204
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.5203706477236009,
|
| 4 |
+
"eval_loss": 2.6626083850860596,
|
| 5 |
+
"eval_runtime": 101.9412,
|
| 6 |
+
"eval_samples": 46868,
|
| 7 |
+
"eval_samples_per_second": 459.755,
|
| 8 |
+
"eval_steps_per_second": 7.19,
|
| 9 |
+
"perplexity": 14.333628001897594,
|
| 10 |
+
"total_flos": 6.171008476428288e+17,
|
| 11 |
+
"train_loss": 2.0168114449748256,
|
| 12 |
+
"train_runtime": 24063.5613,
|
| 13 |
+
"train_samples": 452524,
|
| 14 |
+
"train_samples_per_second": 188.054,
|
| 15 |
+
"train_steps_per_second": 5.877
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.5203706477236009,
|
| 4 |
+
"eval_loss": 2.6626083850860596,
|
| 5 |
+
"eval_runtime": 101.9412,
|
| 6 |
+
"eval_samples": 46868,
|
| 7 |
+
"eval_samples_per_second": 459.755,
|
| 8 |
+
"eval_steps_per_second": 7.19,
|
| 9 |
+
"perplexity": 14.333628001897594
|
| 10 |
+
}
|
runs/Jul24_20-10-48_phyl-ling-p01.la.utexas.edu/events.out.tfevents.1721894287.phyl-ling-p01.la.utexas.edu.130506.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0fea37449358621fdf730edb3521f6f9e1b3d2a0b86759c96a3aa248d1ba938
|
| 3 |
+
size 417
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 6.171008476428288e+17,
|
| 4 |
+
"train_loss": 2.0168114449748256,
|
| 5 |
+
"train_runtime": 24063.5613,
|
| 6 |
+
"train_samples": 452524,
|
| 7 |
+
"train_samples_per_second": 188.054,
|
| 8 |
+
"train_steps_per_second": 5.877
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1119 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 141420,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07071135624381275,
|
| 13 |
+
"grad_norm": 1.5066859722137451,
|
| 14 |
+
"learning_rate": 3.125e-05,
|
| 15 |
+
"loss": 4.9729,
|
| 16 |
+
"step": 1000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1414227124876255,
|
| 20 |
+
"grad_norm": 0.9705550670623779,
|
| 21 |
+
"learning_rate": 6.25e-05,
|
| 22 |
+
"loss": 3.374,
|
| 23 |
+
"step": 2000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.21213406873143828,
|
| 27 |
+
"grad_norm": 0.8380181789398193,
|
| 28 |
+
"learning_rate": 9.375e-05,
|
| 29 |
+
"loss": 3.0791,
|
| 30 |
+
"step": 3000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.282845424975251,
|
| 34 |
+
"grad_norm": 0.7704618573188782,
|
| 35 |
+
"learning_rate": 0.000125,
|
| 36 |
+
"loss": 2.9038,
|
| 37 |
+
"step": 4000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.3535567812190638,
|
| 41 |
+
"grad_norm": 0.675819993019104,
|
| 42 |
+
"learning_rate": 0.00015625,
|
| 43 |
+
"loss": 2.7896,
|
| 44 |
+
"step": 5000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.42426813746287656,
|
| 48 |
+
"grad_norm": 0.6245794892311096,
|
| 49 |
+
"learning_rate": 0.0001875,
|
| 50 |
+
"loss": 2.6919,
|
| 51 |
+
"step": 6000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.4949794937066893,
|
| 55 |
+
"grad_norm": 0.5709772109985352,
|
| 56 |
+
"learning_rate": 0.00021875,
|
| 57 |
+
"loss": 2.6272,
|
| 58 |
+
"step": 7000
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.565690849950502,
|
| 62 |
+
"grad_norm": 0.5335475206375122,
|
| 63 |
+
"learning_rate": 0.00025,
|
| 64 |
+
"loss": 2.5715,
|
| 65 |
+
"step": 8000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.6364022061943148,
|
| 69 |
+
"grad_norm": 0.4924204349517822,
|
| 70 |
+
"learning_rate": 0.00028125000000000003,
|
| 71 |
+
"loss": 2.5384,
|
| 72 |
+
"step": 9000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.7071135624381276,
|
| 76 |
+
"grad_norm": 0.49077439308166504,
|
| 77 |
+
"learning_rate": 0.0003125,
|
| 78 |
+
"loss": 2.5154,
|
| 79 |
+
"step": 10000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.7778249186819404,
|
| 83 |
+
"grad_norm": 0.4304497241973877,
|
| 84 |
+
"learning_rate": 0.00034371875,
|
| 85 |
+
"loss": 2.4784,
|
| 86 |
+
"step": 11000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.8485362749257531,
|
| 90 |
+
"grad_norm": 0.45384302735328674,
|
| 91 |
+
"learning_rate": 0.00037496875000000003,
|
| 92 |
+
"loss": 2.458,
|
| 93 |
+
"step": 12000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.9192476311695659,
|
| 97 |
+
"grad_norm": 0.3978016674518585,
|
| 98 |
+
"learning_rate": 0.0004061875,
|
| 99 |
+
"loss": 2.4536,
|
| 100 |
+
"step": 13000
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.9899589874133786,
|
| 104 |
+
"grad_norm": 0.3981296718120575,
|
| 105 |
+
"learning_rate": 0.0004374375,
|
| 106 |
+
"loss": 2.4397,
|
| 107 |
+
"step": 14000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.0,
|
| 111 |
+
"eval_accuracy": 0.49545158536152034,
|
| 112 |
+
"eval_loss": 2.6684181690216064,
|
| 113 |
+
"eval_runtime": 101.2526,
|
| 114 |
+
"eval_samples_per_second": 462.882,
|
| 115 |
+
"eval_steps_per_second": 7.239,
|
| 116 |
+
"step": 14142
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.0606703436571914,
|
| 120 |
+
"grad_norm": 0.36285269260406494,
|
| 121 |
+
"learning_rate": 0.00046865625,
|
| 122 |
+
"loss": 2.3887,
|
| 123 |
+
"step": 15000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.131381699901004,
|
| 127 |
+
"grad_norm": 0.3568965494632721,
|
| 128 |
+
"learning_rate": 0.00049990625,
|
| 129 |
+
"loss": 2.3848,
|
| 130 |
+
"step": 16000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.2020930561448169,
|
| 134 |
+
"grad_norm": 0.32918983697891235,
|
| 135 |
+
"learning_rate": 0.00053109375,
|
| 136 |
+
"loss": 2.3763,
|
| 137 |
+
"step": 17000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.2728044123886297,
|
| 141 |
+
"grad_norm": 0.3192691206932068,
|
| 142 |
+
"learning_rate": 0.00056234375,
|
| 143 |
+
"loss": 2.3757,
|
| 144 |
+
"step": 18000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.3435157686324424,
|
| 148 |
+
"grad_norm": 0.29217350482940674,
|
| 149 |
+
"learning_rate": 0.00059359375,
|
| 150 |
+
"loss": 2.3728,
|
| 151 |
+
"step": 19000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.414227124876255,
|
| 155 |
+
"grad_norm": 0.2726396918296814,
|
| 156 |
+
"learning_rate": 0.0006248437500000001,
|
| 157 |
+
"loss": 2.3482,
|
| 158 |
+
"step": 20000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.4849384811200679,
|
| 162 |
+
"grad_norm": 0.2647142708301544,
|
| 163 |
+
"learning_rate": 0.0006560625,
|
| 164 |
+
"loss": 2.361,
|
| 165 |
+
"step": 21000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.5556498373638807,
|
| 169 |
+
"grad_norm": 0.24640022218227386,
|
| 170 |
+
"learning_rate": 0.00068728125,
|
| 171 |
+
"loss": 2.3414,
|
| 172 |
+
"step": 22000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.6263611936076934,
|
| 176 |
+
"grad_norm": 0.2376652956008911,
|
| 177 |
+
"learning_rate": 0.00071853125,
|
| 178 |
+
"loss": 2.3469,
|
| 179 |
+
"step": 23000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.697072549851506,
|
| 183 |
+
"grad_norm": 0.20667687058448792,
|
| 184 |
+
"learning_rate": 0.00074978125,
|
| 185 |
+
"loss": 2.3334,
|
| 186 |
+
"step": 24000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.7677839060953189,
|
| 190 |
+
"grad_norm": 0.21862906217575073,
|
| 191 |
+
"learning_rate": 0.0007810312499999999,
|
| 192 |
+
"loss": 2.325,
|
| 193 |
+
"step": 25000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.8384952623391317,
|
| 197 |
+
"grad_norm": 0.19700638949871063,
|
| 198 |
+
"learning_rate": 0.00081225,
|
| 199 |
+
"loss": 2.3169,
|
| 200 |
+
"step": 26000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.9092066185829444,
|
| 204 |
+
"grad_norm": 0.19530941545963287,
|
| 205 |
+
"learning_rate": 0.00084346875,
|
| 206 |
+
"loss": 2.3085,
|
| 207 |
+
"step": 27000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.979917974826757,
|
| 211 |
+
"grad_norm": 0.18496540188789368,
|
| 212 |
+
"learning_rate": 0.00087471875,
|
| 213 |
+
"loss": 2.3085,
|
| 214 |
+
"step": 28000
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.0,
|
| 218 |
+
"eval_accuracy": 0.5092843747934841,
|
| 219 |
+
"eval_loss": 2.5420279502868652,
|
| 220 |
+
"eval_runtime": 102.0041,
|
| 221 |
+
"eval_samples_per_second": 459.472,
|
| 222 |
+
"eval_steps_per_second": 7.186,
|
| 223 |
+
"step": 28284
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.05062933107057,
|
| 227 |
+
"grad_norm": 0.18070034682750702,
|
| 228 |
+
"learning_rate": 0.00090596875,
|
| 229 |
+
"loss": 2.2557,
|
| 230 |
+
"step": 29000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.1213406873143827,
|
| 234 |
+
"grad_norm": 0.17614798247814178,
|
| 235 |
+
"learning_rate": 0.0009371875,
|
| 236 |
+
"loss": 2.248,
|
| 237 |
+
"step": 30000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.1920520435581956,
|
| 241 |
+
"grad_norm": 0.18162938952445984,
|
| 242 |
+
"learning_rate": 0.0009684375,
|
| 243 |
+
"loss": 2.246,
|
| 244 |
+
"step": 31000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.262763399802008,
|
| 248 |
+
"grad_norm": 0.16680462658405304,
|
| 249 |
+
"learning_rate": 0.0009996875,
|
| 250 |
+
"loss": 2.2398,
|
| 251 |
+
"step": 32000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.333474756045821,
|
| 255 |
+
"grad_norm": 0.17343448102474213,
|
| 256 |
+
"learning_rate": 0.0009909614330104186,
|
| 257 |
+
"loss": 2.2332,
|
| 258 |
+
"step": 33000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.4041861122896337,
|
| 262 |
+
"grad_norm": 0.15368333458900452,
|
| 263 |
+
"learning_rate": 0.0009818223359532078,
|
| 264 |
+
"loss": 2.2367,
|
| 265 |
+
"step": 34000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.4748974685334466,
|
| 269 |
+
"grad_norm": 0.14444677531719208,
|
| 270 |
+
"learning_rate": 0.0009726832388959971,
|
| 271 |
+
"loss": 2.2277,
|
| 272 |
+
"step": 35000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.5456088247772595,
|
| 276 |
+
"grad_norm": 0.16958372294902802,
|
| 277 |
+
"learning_rate": 0.0009635441418387864,
|
| 278 |
+
"loss": 2.2136,
|
| 279 |
+
"step": 36000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.616320181021072,
|
| 283 |
+
"grad_norm": 0.15171754360198975,
|
| 284 |
+
"learning_rate": 0.0009544141838786328,
|
| 285 |
+
"loss": 2.2105,
|
| 286 |
+
"step": 37000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 2.6870315372648848,
|
| 290 |
+
"grad_norm": 0.13588131964206696,
|
| 291 |
+
"learning_rate": 0.0009452750868214221,
|
| 292 |
+
"loss": 2.2056,
|
| 293 |
+
"step": 38000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.7577428935086976,
|
| 297 |
+
"grad_norm": 0.13553854823112488,
|
| 298 |
+
"learning_rate": 0.0009361359897642113,
|
| 299 |
+
"loss": 2.1988,
|
| 300 |
+
"step": 39000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.82845424975251,
|
| 304 |
+
"grad_norm": 0.15744280815124512,
|
| 305 |
+
"learning_rate": 0.0009269968927070006,
|
| 306 |
+
"loss": 2.1949,
|
| 307 |
+
"step": 40000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.899165605996323,
|
| 311 |
+
"grad_norm": 0.1427813470363617,
|
| 312 |
+
"learning_rate": 0.000917866934746847,
|
| 313 |
+
"loss": 2.1875,
|
| 314 |
+
"step": 41000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.9698769622401358,
|
| 318 |
+
"grad_norm": 0.14179003238677979,
|
| 319 |
+
"learning_rate": 0.0009087278376896363,
|
| 320 |
+
"loss": 2.19,
|
| 321 |
+
"step": 42000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.0,
|
| 325 |
+
"eval_accuracy": 0.5214661161125094,
|
| 326 |
+
"eval_loss": 2.439739942550659,
|
| 327 |
+
"eval_runtime": 102.0725,
|
| 328 |
+
"eval_samples_per_second": 459.164,
|
| 329 |
+
"eval_steps_per_second": 7.181,
|
| 330 |
+
"step": 42426
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.0405883184839486,
|
| 334 |
+
"grad_norm": 0.1600356101989746,
|
| 335 |
+
"learning_rate": 0.0008995978797294828,
|
| 336 |
+
"loss": 2.13,
|
| 337 |
+
"step": 43000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.1112996747277615,
|
| 341 |
+
"grad_norm": 0.16733036935329437,
|
| 342 |
+
"learning_rate": 0.0008904587826722719,
|
| 343 |
+
"loss": 2.0964,
|
| 344 |
+
"step": 44000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.182011030971574,
|
| 348 |
+
"grad_norm": 0.15149937570095062,
|
| 349 |
+
"learning_rate": 0.0008813379638091756,
|
| 350 |
+
"loss": 2.0964,
|
| 351 |
+
"step": 45000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.2527223872153868,
|
| 355 |
+
"grad_norm": 0.1375265121459961,
|
| 356 |
+
"learning_rate": 0.0008721988667519649,
|
| 357 |
+
"loss": 2.1021,
|
| 358 |
+
"step": 46000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.3234337434591996,
|
| 362 |
+
"grad_norm": 0.13642068207263947,
|
| 363 |
+
"learning_rate": 0.0008630597696947542,
|
| 364 |
+
"loss": 2.1062,
|
| 365 |
+
"step": 47000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.3941450997030125,
|
| 369 |
+
"grad_norm": 0.15942348539829254,
|
| 370 |
+
"learning_rate": 0.0008539206726375435,
|
| 371 |
+
"loss": 2.0943,
|
| 372 |
+
"step": 48000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.464856455946825,
|
| 376 |
+
"grad_norm": 0.14231225848197937,
|
| 377 |
+
"learning_rate": 0.0008447815755803326,
|
| 378 |
+
"loss": 2.0968,
|
| 379 |
+
"step": 49000
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 3.5355678121906378,
|
| 383 |
+
"grad_norm": 0.13483628630638123,
|
| 384 |
+
"learning_rate": 0.0008356516176201791,
|
| 385 |
+
"loss": 2.0923,
|
| 386 |
+
"step": 50000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.6062791684344506,
|
| 390 |
+
"grad_norm": 0.15377779304981232,
|
| 391 |
+
"learning_rate": 0.0008265125205629684,
|
| 392 |
+
"loss": 2.0929,
|
| 393 |
+
"step": 51000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.6769905246782635,
|
| 397 |
+
"grad_norm": 0.13733841478824615,
|
| 398 |
+
"learning_rate": 0.0008173825626028149,
|
| 399 |
+
"loss": 2.0929,
|
| 400 |
+
"step": 52000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 3.747701880922076,
|
| 404 |
+
"grad_norm": 0.13640180230140686,
|
| 405 |
+
"learning_rate": 0.0008082434655456042,
|
| 406 |
+
"loss": 2.0938,
|
| 407 |
+
"step": 53000
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 3.8184132371658888,
|
| 411 |
+
"grad_norm": 0.13909070193767548,
|
| 412 |
+
"learning_rate": 0.0007991135075854505,
|
| 413 |
+
"loss": 2.0907,
|
| 414 |
+
"step": 54000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 3.8891245934097016,
|
| 418 |
+
"grad_norm": 0.1521981954574585,
|
| 419 |
+
"learning_rate": 0.0007899744105282398,
|
| 420 |
+
"loss": 2.0816,
|
| 421 |
+
"step": 55000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 3.9598359496535145,
|
| 425 |
+
"grad_norm": 0.12255113571882248,
|
| 426 |
+
"learning_rate": 0.0007808444525680864,
|
| 427 |
+
"loss": 2.0865,
|
| 428 |
+
"step": 56000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.0,
|
| 432 |
+
"eval_accuracy": 0.5276129432475146,
|
| 433 |
+
"eval_loss": 2.3943161964416504,
|
| 434 |
+
"eval_runtime": 104.7687,
|
| 435 |
+
"eval_samples_per_second": 447.347,
|
| 436 |
+
"eval_steps_per_second": 6.996,
|
| 437 |
+
"step": 56568
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.030547305897327,
|
| 441 |
+
"grad_norm": 0.1423817127943039,
|
| 442 |
+
"learning_rate": 0.0007717053555108755,
|
| 443 |
+
"loss": 2.0304,
|
| 444 |
+
"step": 57000
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.10125866214114,
|
| 448 |
+
"grad_norm": 0.13736553490161896,
|
| 449 |
+
"learning_rate": 0.0007625662584536648,
|
| 450 |
+
"loss": 1.9815,
|
| 451 |
+
"step": 58000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 4.171970018384952,
|
| 455 |
+
"grad_norm": 0.1411396712064743,
|
| 456 |
+
"learning_rate": 0.0007534363004935113,
|
| 457 |
+
"loss": 1.9919,
|
| 458 |
+
"step": 59000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 4.2426813746287655,
|
| 462 |
+
"grad_norm": 0.14484618604183197,
|
| 463 |
+
"learning_rate": 0.0007442972034363005,
|
| 464 |
+
"loss": 1.9915,
|
| 465 |
+
"step": 60000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.313392730872578,
|
| 469 |
+
"grad_norm": 0.1606305092573166,
|
| 470 |
+
"learning_rate": 0.000735167245476147,
|
| 471 |
+
"loss": 1.9925,
|
| 472 |
+
"step": 61000
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 4.384104087116391,
|
| 476 |
+
"grad_norm": 0.15816234052181244,
|
| 477 |
+
"learning_rate": 0.0007260281484189363,
|
| 478 |
+
"loss": 1.9963,
|
| 479 |
+
"step": 62000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 4.454815443360204,
|
| 483 |
+
"grad_norm": 0.14397823810577393,
|
| 484 |
+
"learning_rate": 0.0007168981904587826,
|
| 485 |
+
"loss": 1.9989,
|
| 486 |
+
"step": 63000
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 4.525526799604016,
|
| 490 |
+
"grad_norm": 0.15473702549934387,
|
| 491 |
+
"learning_rate": 0.0007077590934015719,
|
| 492 |
+
"loss": 1.9965,
|
| 493 |
+
"step": 64000
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 4.596238155847829,
|
| 497 |
+
"grad_norm": 0.14191265404224396,
|
| 498 |
+
"learning_rate": 0.0006986291354414184,
|
| 499 |
+
"loss": 2.0005,
|
| 500 |
+
"step": 65000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 4.666949512091642,
|
| 504 |
+
"grad_norm": 0.15206751227378845,
|
| 505 |
+
"learning_rate": 0.0006894900383842077,
|
| 506 |
+
"loss": 2.0114,
|
| 507 |
+
"step": 66000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 4.737660868335455,
|
| 511 |
+
"grad_norm": 0.18548937141895294,
|
| 512 |
+
"learning_rate": 0.0006803600804240542,
|
| 513 |
+
"loss": 2.0021,
|
| 514 |
+
"step": 67000
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 4.8083722245792675,
|
| 518 |
+
"grad_norm": 0.16364724934101105,
|
| 519 |
+
"learning_rate": 0.0006712209833668433,
|
| 520 |
+
"loss": 2.0093,
|
| 521 |
+
"step": 68000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.87908358082308,
|
| 525 |
+
"grad_norm": 0.1373205929994583,
|
| 526 |
+
"learning_rate": 0.0006620818863096326,
|
| 527 |
+
"loss": 2.0073,
|
| 528 |
+
"step": 69000
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 4.949794937066893,
|
| 532 |
+
"grad_norm": 0.15305304527282715,
|
| 533 |
+
"learning_rate": 0.000652951928349479,
|
| 534 |
+
"loss": 1.9957,
|
| 535 |
+
"step": 70000
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 5.0,
|
| 539 |
+
"eval_accuracy": 0.5305311481637808,
|
| 540 |
+
"eval_loss": 2.3786160945892334,
|
| 541 |
+
"eval_runtime": 102.0738,
|
| 542 |
+
"eval_samples_per_second": 459.158,
|
| 543 |
+
"eval_steps_per_second": 7.181,
|
| 544 |
+
"step": 70710
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 5.020506293310706,
|
| 548 |
+
"grad_norm": 0.17954622209072113,
|
| 549 |
+
"learning_rate": 0.0006438128312922683,
|
| 550 |
+
"loss": 1.963,
|
| 551 |
+
"step": 71000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 5.091217649554518,
|
| 555 |
+
"grad_norm": 0.17249706387519836,
|
| 556 |
+
"learning_rate": 0.0006346828733321149,
|
| 557 |
+
"loss": 1.8814,
|
| 558 |
+
"step": 72000
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 5.161929005798331,
|
| 562 |
+
"grad_norm": 0.16035763919353485,
|
| 563 |
+
"learning_rate": 0.000625543776274904,
|
| 564 |
+
"loss": 1.8888,
|
| 565 |
+
"step": 73000
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 5.232640362042144,
|
| 569 |
+
"grad_norm": 0.16601450741291046,
|
| 570 |
+
"learning_rate": 0.0006164046792176932,
|
| 571 |
+
"loss": 1.8945,
|
| 572 |
+
"step": 74000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 5.303351718285957,
|
| 576 |
+
"grad_norm": 0.1559607982635498,
|
| 577 |
+
"learning_rate": 0.0006072747212575398,
|
| 578 |
+
"loss": 1.9005,
|
| 579 |
+
"step": 75000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.3740630745297695,
|
| 583 |
+
"grad_norm": 0.1599714308977127,
|
| 584 |
+
"learning_rate": 0.000598135624200329,
|
| 585 |
+
"loss": 1.9056,
|
| 586 |
+
"step": 76000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 5.444774430773582,
|
| 590 |
+
"grad_norm": 0.15538254380226135,
|
| 591 |
+
"learning_rate": 0.0005890056662401755,
|
| 592 |
+
"loss": 1.9091,
|
| 593 |
+
"step": 77000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 5.515485787017395,
|
| 597 |
+
"grad_norm": 0.1645193099975586,
|
| 598 |
+
"learning_rate": 0.0005798665691829647,
|
| 599 |
+
"loss": 1.9138,
|
| 600 |
+
"step": 78000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 5.586197143261208,
|
| 604 |
+
"grad_norm": 0.1560288369655609,
|
| 605 |
+
"learning_rate": 0.0005707366112228112,
|
| 606 |
+
"loss": 1.9276,
|
| 607 |
+
"step": 79000
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 5.65690849950502,
|
| 611 |
+
"grad_norm": 0.169467955827713,
|
| 612 |
+
"learning_rate": 0.0005615975141656004,
|
| 613 |
+
"loss": 1.9167,
|
| 614 |
+
"step": 80000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 5.727619855748833,
|
| 618 |
+
"grad_norm": 0.18090558052062988,
|
| 619 |
+
"learning_rate": 0.0005524675562054469,
|
| 620 |
+
"loss": 1.9289,
|
| 621 |
+
"step": 81000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 5.798331211992646,
|
| 625 |
+
"grad_norm": 0.16788819432258606,
|
| 626 |
+
"learning_rate": 0.0005433284591482362,
|
| 627 |
+
"loss": 1.9228,
|
| 628 |
+
"step": 82000
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 5.869042568236459,
|
| 632 |
+
"grad_norm": 0.15961690247058868,
|
| 633 |
+
"learning_rate": 0.0005341893620910255,
|
| 634 |
+
"loss": 1.9178,
|
| 635 |
+
"step": 83000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 5.9397539244802715,
|
| 639 |
+
"grad_norm": 0.15657977759838104,
|
| 640 |
+
"learning_rate": 0.0005250594041308718,
|
| 641 |
+
"loss": 1.9161,
|
| 642 |
+
"step": 84000
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 6.0,
|
| 646 |
+
"eval_accuracy": 0.5312578351518911,
|
| 647 |
+
"eval_loss": 2.3910350799560547,
|
| 648 |
+
"eval_runtime": 102.0407,
|
| 649 |
+
"eval_samples_per_second": 459.307,
|
| 650 |
+
"eval_steps_per_second": 7.183,
|
| 651 |
+
"step": 84852
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.010465280724084,
|
| 655 |
+
"grad_norm": 0.15551112592220306,
|
| 656 |
+
"learning_rate": 0.0005159203070736611,
|
| 657 |
+
"loss": 1.9123,
|
| 658 |
+
"step": 85000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 6.081176636967897,
|
| 662 |
+
"grad_norm": 0.18589554727077484,
|
| 663 |
+
"learning_rate": 0.0005067812100164504,
|
| 664 |
+
"loss": 1.7906,
|
| 665 |
+
"step": 86000
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 6.15188799321171,
|
| 669 |
+
"grad_norm": 0.16240116953849792,
|
| 670 |
+
"learning_rate": 0.0004976512520562968,
|
| 671 |
+
"loss": 1.805,
|
| 672 |
+
"step": 87000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.222599349455523,
|
| 676 |
+
"grad_norm": 0.1752467155456543,
|
| 677 |
+
"learning_rate": 0.0004885121549990861,
|
| 678 |
+
"loss": 1.8147,
|
| 679 |
+
"step": 88000
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 6.293310705699335,
|
| 683 |
+
"grad_norm": 0.15973269939422607,
|
| 684 |
+
"learning_rate": 0.00047937305794187537,
|
| 685 |
+
"loss": 1.8063,
|
| 686 |
+
"step": 89000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 6.364022061943148,
|
| 690 |
+
"grad_norm": 0.18358197808265686,
|
| 691 |
+
"learning_rate": 0.0004702430999817218,
|
| 692 |
+
"loss": 1.8182,
|
| 693 |
+
"step": 90000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 6.434733418186961,
|
| 697 |
+
"grad_norm": 0.20550867915153503,
|
| 698 |
+
"learning_rate": 0.00046110400292451105,
|
| 699 |
+
"loss": 1.8251,
|
| 700 |
+
"step": 91000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 6.5054447744307735,
|
| 704 |
+
"grad_norm": 0.18148034811019897,
|
| 705 |
+
"learning_rate": 0.0004519740449643575,
|
| 706 |
+
"loss": 1.8283,
|
| 707 |
+
"step": 92000
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 6.576156130674587,
|
| 711 |
+
"grad_norm": 0.1863207072019577,
|
| 712 |
+
"learning_rate": 0.0004428349479071468,
|
| 713 |
+
"loss": 1.834,
|
| 714 |
+
"step": 93000
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 6.646867486918399,
|
| 718 |
+
"grad_norm": 0.1836949586868286,
|
| 719 |
+
"learning_rate": 0.000433695850849936,
|
| 720 |
+
"loss": 1.8257,
|
| 721 |
+
"step": 94000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 6.717578843162212,
|
| 725 |
+
"grad_norm": 0.18851223587989807,
|
| 726 |
+
"learning_rate": 0.00042456589288978247,
|
| 727 |
+
"loss": 1.8291,
|
| 728 |
+
"step": 95000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 6.788290199406025,
|
| 732 |
+
"grad_norm": 0.16575908660888672,
|
| 733 |
+
"learning_rate": 0.00041542679583257176,
|
| 734 |
+
"loss": 1.8412,
|
| 735 |
+
"step": 96000
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 6.859001555649837,
|
| 739 |
+
"grad_norm": 0.1861979216337204,
|
| 740 |
+
"learning_rate": 0.000406287698775361,
|
| 741 |
+
"loss": 1.848,
|
| 742 |
+
"step": 97000
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 6.92971291189365,
|
| 746 |
+
"grad_norm": 0.1783532202243805,
|
| 747 |
+
"learning_rate": 0.00039714860171815024,
|
| 748 |
+
"loss": 1.8361,
|
| 749 |
+
"step": 98000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 7.0,
|
| 753 |
+
"eval_accuracy": 0.5303533991815411,
|
| 754 |
+
"eval_loss": 2.4205334186553955,
|
| 755 |
+
"eval_runtime": 102.4141,
|
| 756 |
+
"eval_samples_per_second": 457.632,
|
| 757 |
+
"eval_steps_per_second": 7.157,
|
| 758 |
+
"step": 98994
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 7.000424268137463,
|
| 762 |
+
"grad_norm": 0.1907605677843094,
|
| 763 |
+
"learning_rate": 0.00038801864375799674,
|
| 764 |
+
"loss": 1.8413,
|
| 765 |
+
"step": 99000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 7.0711356243812755,
|
| 769 |
+
"grad_norm": 0.21442489326000214,
|
| 770 |
+
"learning_rate": 0.0003788795467007859,
|
| 771 |
+
"loss": 1.6956,
|
| 772 |
+
"step": 100000
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 7.141846980625088,
|
| 776 |
+
"grad_norm": 0.19562986493110657,
|
| 777 |
+
"learning_rate": 0.0003697404496435752,
|
| 778 |
+
"loss": 1.7053,
|
| 779 |
+
"step": 101000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 7.212558336868901,
|
| 783 |
+
"grad_norm": 0.23670311272144318,
|
| 784 |
+
"learning_rate": 0.00036060135258636445,
|
| 785 |
+
"loss": 1.7196,
|
| 786 |
+
"step": 102000
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 7.283269693112714,
|
| 790 |
+
"grad_norm": 0.19641369581222534,
|
| 791 |
+
"learning_rate": 0.00035148053372326815,
|
| 792 |
+
"loss": 1.719,
|
| 793 |
+
"step": 103000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 7.353981049356527,
|
| 797 |
+
"grad_norm": 0.2086309790611267,
|
| 798 |
+
"learning_rate": 0.0003423414366660574,
|
| 799 |
+
"loss": 1.7279,
|
| 800 |
+
"step": 104000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 7.424692405600339,
|
| 804 |
+
"grad_norm": 0.1947568953037262,
|
| 805 |
+
"learning_rate": 0.0003332023396088467,
|
| 806 |
+
"loss": 1.7389,
|
| 807 |
+
"step": 105000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 7.495403761844152,
|
| 811 |
+
"grad_norm": 0.19536983966827393,
|
| 812 |
+
"learning_rate": 0.00032407238164869313,
|
| 813 |
+
"loss": 1.7428,
|
| 814 |
+
"step": 106000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 7.566115118087965,
|
| 818 |
+
"grad_norm": 0.1872589886188507,
|
| 819 |
+
"learning_rate": 0.00031493328459148237,
|
| 820 |
+
"loss": 1.7463,
|
| 821 |
+
"step": 107000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 7.6368264743317775,
|
| 825 |
+
"grad_norm": 0.22906361520290375,
|
| 826 |
+
"learning_rate": 0.0003057941875342716,
|
| 827 |
+
"loss": 1.7479,
|
| 828 |
+
"step": 108000
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 7.707537830575591,
|
| 832 |
+
"grad_norm": 0.19299902021884918,
|
| 833 |
+
"learning_rate": 0.0002966642295741181,
|
| 834 |
+
"loss": 1.7514,
|
| 835 |
+
"step": 109000
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 7.778249186819403,
|
| 839 |
+
"grad_norm": 0.19876809418201447,
|
| 840 |
+
"learning_rate": 0.00028752513251690734,
|
| 841 |
+
"loss": 1.7467,
|
| 842 |
+
"step": 110000
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 7.848960543063216,
|
| 846 |
+
"grad_norm": 0.22273430228233337,
|
| 847 |
+
"learning_rate": 0.0002783860354596966,
|
| 848 |
+
"loss": 1.76,
|
| 849 |
+
"step": 111000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 7.919671899307029,
|
| 853 |
+
"grad_norm": 0.1979241669178009,
|
| 854 |
+
"learning_rate": 0.0002692560774995431,
|
| 855 |
+
"loss": 1.7547,
|
| 856 |
+
"step": 112000
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 7.990383255550841,
|
| 860 |
+
"grad_norm": 0.2099294811487198,
|
| 861 |
+
"learning_rate": 0.00026011698044233226,
|
| 862 |
+
"loss": 1.7477,
|
| 863 |
+
"step": 113000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 8.0,
|
| 867 |
+
"eval_accuracy": 0.5282502161049046,
|
| 868 |
+
"eval_loss": 2.474827289581299,
|
| 869 |
+
"eval_runtime": 102.4954,
|
| 870 |
+
"eval_samples_per_second": 457.269,
|
| 871 |
+
"eval_steps_per_second": 7.152,
|
| 872 |
+
"step": 113136
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 8.061094611794655,
|
| 876 |
+
"grad_norm": 0.24672599136829376,
|
| 877 |
+
"learning_rate": 0.00025097788338512156,
|
| 878 |
+
"loss": 1.6197,
|
| 879 |
+
"step": 114000
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 8.131805968038467,
|
| 883 |
+
"grad_norm": 0.21202607452869415,
|
| 884 |
+
"learning_rate": 0.00024183878632791082,
|
| 885 |
+
"loss": 1.6192,
|
| 886 |
+
"step": 115000
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 8.20251732428228,
|
| 890 |
+
"grad_norm": 0.24981403350830078,
|
| 891 |
+
"learning_rate": 0.00023271796746481447,
|
| 892 |
+
"loss": 1.6329,
|
| 893 |
+
"step": 116000
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 8.273228680526092,
|
| 897 |
+
"grad_norm": 0.25290995836257935,
|
| 898 |
+
"learning_rate": 0.00022357887040760373,
|
| 899 |
+
"loss": 1.6386,
|
| 900 |
+
"step": 117000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 8.343940036769904,
|
| 904 |
+
"grad_norm": 0.2473640739917755,
|
| 905 |
+
"learning_rate": 0.000214439773350393,
|
| 906 |
+
"loss": 1.6414,
|
| 907 |
+
"step": 118000
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 8.414651393013719,
|
| 911 |
+
"grad_norm": 0.20307676494121552,
|
| 912 |
+
"learning_rate": 0.00020530981539023944,
|
| 913 |
+
"loss": 1.6458,
|
| 914 |
+
"step": 119000
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 8.485362749257531,
|
| 918 |
+
"grad_norm": 0.21696613729000092,
|
| 919 |
+
"learning_rate": 0.0001961707183330287,
|
| 920 |
+
"loss": 1.6473,
|
| 921 |
+
"step": 120000
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 8.556074105501343,
|
| 925 |
+
"grad_norm": 0.23408186435699463,
|
| 926 |
+
"learning_rate": 0.00018703162127581797,
|
| 927 |
+
"loss": 1.656,
|
| 928 |
+
"step": 121000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 8.626785461745156,
|
| 932 |
+
"grad_norm": 0.23058977723121643,
|
| 933 |
+
"learning_rate": 0.0001778925242186072,
|
| 934 |
+
"loss": 1.6578,
|
| 935 |
+
"step": 122000
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 8.697496817988968,
|
| 939 |
+
"grad_norm": 0.23317036032676697,
|
| 940 |
+
"learning_rate": 0.00016877170535551086,
|
| 941 |
+
"loss": 1.6516,
|
| 942 |
+
"step": 123000
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 8.768208174232782,
|
| 946 |
+
"grad_norm": 0.2361781746149063,
|
| 947 |
+
"learning_rate": 0.00015963260829830012,
|
| 948 |
+
"loss": 1.6525,
|
| 949 |
+
"step": 124000
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 8.838919530476595,
|
| 953 |
+
"grad_norm": 0.260776549577713,
|
| 954 |
+
"learning_rate": 0.00015049351124108936,
|
| 955 |
+
"loss": 1.6547,
|
| 956 |
+
"step": 125000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 8.909630886720407,
|
| 960 |
+
"grad_norm": 0.2507932186126709,
|
| 961 |
+
"learning_rate": 0.00014136355328093583,
|
| 962 |
+
"loss": 1.6556,
|
| 963 |
+
"step": 126000
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 8.98034224296422,
|
| 967 |
+
"grad_norm": 0.2422228902578354,
|
| 968 |
+
"learning_rate": 0.0001322244562237251,
|
| 969 |
+
"loss": 1.6549,
|
| 970 |
+
"step": 127000
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 9.0,
|
| 974 |
+
"eval_accuracy": 0.5249380742803117,
|
| 975 |
+
"eval_loss": 2.5581541061401367,
|
| 976 |
+
"eval_runtime": 102.2383,
|
| 977 |
+
"eval_samples_per_second": 458.419,
|
| 978 |
+
"eval_steps_per_second": 7.17,
|
| 979 |
+
"step": 127278
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 9.051053599208032,
|
| 983 |
+
"grad_norm": 0.2604562044143677,
|
| 984 |
+
"learning_rate": 0.00012308535916651437,
|
| 985 |
+
"loss": 1.5675,
|
| 986 |
+
"step": 128000
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 9.121764955451846,
|
| 990 |
+
"grad_norm": 0.22102615237236023,
|
| 991 |
+
"learning_rate": 0.0001139462621093036,
|
| 992 |
+
"loss": 1.5337,
|
| 993 |
+
"step": 129000
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 9.192476311695659,
|
| 997 |
+
"grad_norm": 0.2960878014564514,
|
| 998 |
+
"learning_rate": 0.00010481630414915007,
|
| 999 |
+
"loss": 1.5556,
|
| 1000 |
+
"step": 130000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 9.263187667939471,
|
| 1004 |
+
"grad_norm": 0.22400617599487305,
|
| 1005 |
+
"learning_rate": 9.567720709193931e-05,
|
| 1006 |
+
"loss": 1.5491,
|
| 1007 |
+
"step": 131000
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 9.333899024183284,
|
| 1011 |
+
"grad_norm": 0.24257275462150574,
|
| 1012 |
+
"learning_rate": 8.655638822884298e-05,
|
| 1013 |
+
"loss": 1.5502,
|
| 1014 |
+
"step": 132000
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 9.404610380427096,
|
| 1018 |
+
"grad_norm": 0.24599485099315643,
|
| 1019 |
+
"learning_rate": 7.741729117163225e-05,
|
| 1020 |
+
"loss": 1.552,
|
| 1021 |
+
"step": 133000
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 9.47532173667091,
|
| 1025 |
+
"grad_norm": 0.25757452845573425,
|
| 1026 |
+
"learning_rate": 6.82781941144215e-05,
|
| 1027 |
+
"loss": 1.5576,
|
| 1028 |
+
"step": 134000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 9.546033092914723,
|
| 1032 |
+
"grad_norm": 0.28276532888412476,
|
| 1033 |
+
"learning_rate": 5.914823615426796e-05,
|
| 1034 |
+
"loss": 1.5529,
|
| 1035 |
+
"step": 135000
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 9.616744449158535,
|
| 1039 |
+
"grad_norm": 0.2369563728570938,
|
| 1040 |
+
"learning_rate": 5.000913909705721e-05,
|
| 1041 |
+
"loss": 1.5548,
|
| 1042 |
+
"step": 136000
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 9.687455805402347,
|
| 1046 |
+
"grad_norm": 0.25778231024742126,
|
| 1047 |
+
"learning_rate": 4.0870042039846464e-05,
|
| 1048 |
+
"loss": 1.55,
|
| 1049 |
+
"step": 137000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 9.75816716164616,
|
| 1053 |
+
"grad_norm": 0.2770988941192627,
|
| 1054 |
+
"learning_rate": 3.173094498263571e-05,
|
| 1055 |
+
"loss": 1.559,
|
| 1056 |
+
"step": 138000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 9.828878517889972,
|
| 1060 |
+
"grad_norm": 0.2701665163040161,
|
| 1061 |
+
"learning_rate": 2.261012611953939e-05,
|
| 1062 |
+
"loss": 1.5579,
|
| 1063 |
+
"step": 139000
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 9.899589874133786,
|
| 1067 |
+
"grad_norm": 0.2540683448314667,
|
| 1068 |
+
"learning_rate": 1.3471029062328641e-05,
|
| 1069 |
+
"loss": 1.5563,
|
| 1070 |
+
"step": 140000
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 9.970301230377599,
|
| 1074 |
+
"grad_norm": 0.26811909675598145,
|
| 1075 |
+
"learning_rate": 4.341071102175106e-06,
|
| 1076 |
+
"loss": 1.5611,
|
| 1077 |
+
"step": 141000
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 10.0,
|
| 1081 |
+
"eval_accuracy": 0.5203706477236009,
|
| 1082 |
+
"eval_loss": 2.6626083850860596,
|
| 1083 |
+
"eval_runtime": 102.3167,
|
| 1084 |
+
"eval_samples_per_second": 458.068,
|
| 1085 |
+
"eval_steps_per_second": 7.164,
|
| 1086 |
+
"step": 141420
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 10.0,
|
| 1090 |
+
"step": 141420,
|
| 1091 |
+
"total_flos": 6.171008476428288e+17,
|
| 1092 |
+
"train_loss": 2.0168114449748256,
|
| 1093 |
+
"train_runtime": 24063.5613,
|
| 1094 |
+
"train_samples_per_second": 188.054,
|
| 1095 |
+
"train_steps_per_second": 5.877
|
| 1096 |
+
}
|
| 1097 |
+
],
|
| 1098 |
+
"logging_steps": 1000,
|
| 1099 |
+
"max_steps": 141420,
|
| 1100 |
+
"num_input_tokens_seen": 0,
|
| 1101 |
+
"num_train_epochs": 10,
|
| 1102 |
+
"save_steps": 5000,
|
| 1103 |
+
"stateful_callbacks": {
|
| 1104 |
+
"TrainerControl": {
|
| 1105 |
+
"args": {
|
| 1106 |
+
"should_epoch_stop": false,
|
| 1107 |
+
"should_evaluate": false,
|
| 1108 |
+
"should_log": false,
|
| 1109 |
+
"should_save": true,
|
| 1110 |
+
"should_training_stop": true
|
| 1111 |
+
},
|
| 1112 |
+
"attributes": {}
|
| 1113 |
+
}
|
| 1114 |
+
},
|
| 1115 |
+
"total_flos": 6.171008476428288e+17,
|
| 1116 |
+
"train_batch_size": 32,
|
| 1117 |
+
"trial_name": null,
|
| 1118 |
+
"trial_params": null
|
| 1119 |
+
}
|