End of training
Browse files- README.md +3 -3
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +10 -0
- trainer_state.json +1345 -0
README.md
CHANGED
|
@@ -18,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [EleutherAI/pythia-410m](https://huggingface.co/EleutherAI/pythia-410m) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 1.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
-
- Num Input Tokens Seen:
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [EleutherAI/pythia-410m](https://huggingface.co/EleutherAI/pythia-410m) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.6151
|
| 22 |
+
- Accuracy: 0.2123
|
| 23 |
+
- Num Input Tokens Seen: 5873664
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.21226415094339623,
|
| 4 |
+
"eval_loss": 1.615054726600647,
|
| 5 |
+
"eval_runtime": 1.7054,
|
| 6 |
+
"eval_samples": 212,
|
| 7 |
+
"eval_samples_per_second": 124.312,
|
| 8 |
+
"eval_steps_per_second": 15.832,
|
| 9 |
+
"num_input_tokens_seen": 5873664,
|
| 10 |
+
"total_flos": 1.0654234806583296e+16,
|
| 11 |
+
"train_loss": 1.7262512909318612,
|
| 12 |
+
"train_runtime": 402.0358,
|
| 13 |
+
"train_samples": 1905,
|
| 14 |
+
"train_samples_per_second": 14.215,
|
| 15 |
+
"train_steps_per_second": 1.783
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.21226415094339623,
|
| 4 |
+
"eval_loss": 1.615054726600647,
|
| 5 |
+
"eval_runtime": 1.7054,
|
| 6 |
+
"eval_samples": 212,
|
| 7 |
+
"eval_samples_per_second": 124.312,
|
| 8 |
+
"eval_steps_per_second": 15.832,
|
| 9 |
+
"num_input_tokens_seen": 5873664
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"num_input_tokens_seen": 5873664,
|
| 4 |
+
"total_flos": 1.0654234806583296e+16,
|
| 5 |
+
"train_loss": 1.7262512909318612,
|
| 6 |
+
"train_runtime": 402.0358,
|
| 7 |
+
"train_samples": 1905,
|
| 8 |
+
"train_samples_per_second": 14.215,
|
| 9 |
+
"train_steps_per_second": 1.783
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1345 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.615054726600647,
|
| 3 |
+
"best_model_checkpoint": "/data/user_data/gonilude/oop_and_text_pythia_410m/checkpoint-500",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 717,
|
| 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,
|
| 13 |
+
"eval_accuracy": 0.20754716981132076,
|
| 14 |
+
"eval_loss": 3.0230751037597656,
|
| 15 |
+
"eval_runtime": 1.9283,
|
| 16 |
+
"eval_samples_per_second": 109.944,
|
| 17 |
+
"eval_steps_per_second": 14.002,
|
| 18 |
+
"num_input_tokens_seen": 0,
|
| 19 |
+
"step": 0
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.0041841004184100415,
|
| 23 |
+
"grad_norm": NaN,
|
| 24 |
+
"learning_rate": 0.0,
|
| 25 |
+
"loss": 2.7406,
|
| 26 |
+
"num_input_tokens_seen": 8192,
|
| 27 |
+
"step": 1
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.02092050209205021,
|
| 31 |
+
"grad_norm": 246.1937255859375,
|
| 32 |
+
"learning_rate": 9.090909090909091e-07,
|
| 33 |
+
"loss": 2.8031,
|
| 34 |
+
"num_input_tokens_seen": 40960,
|
| 35 |
+
"step": 5
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.04184100418410042,
|
| 39 |
+
"grad_norm": Infinity,
|
| 40 |
+
"learning_rate": 3.6363636363636366e-06,
|
| 41 |
+
"loss": 3.2837,
|
| 42 |
+
"num_input_tokens_seen": 81920,
|
| 43 |
+
"step": 10
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.06276150627615062,
|
| 47 |
+
"grad_norm": 205.18572998046875,
|
| 48 |
+
"learning_rate": 8.181818181818183e-06,
|
| 49 |
+
"loss": 2.9171,
|
| 50 |
+
"num_input_tokens_seen": 122880,
|
| 51 |
+
"step": 15
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08368200836820083,
|
| 55 |
+
"grad_norm": 221.2069549560547,
|
| 56 |
+
"learning_rate": 1.2727272727272728e-05,
|
| 57 |
+
"loss": 2.6217,
|
| 58 |
+
"num_input_tokens_seen": 163840,
|
| 59 |
+
"step": 20
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.10460251046025104,
|
| 63 |
+
"grad_norm": 126.98030090332031,
|
| 64 |
+
"learning_rate": 1.7272727272727274e-05,
|
| 65 |
+
"loss": 2.2064,
|
| 66 |
+
"num_input_tokens_seen": 204800,
|
| 67 |
+
"step": 25
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.12552301255230125,
|
| 71 |
+
"grad_norm": 69.57221984863281,
|
| 72 |
+
"learning_rate": 1.9999591344664163e-05,
|
| 73 |
+
"loss": 2.1854,
|
| 74 |
+
"num_input_tokens_seen": 245760,
|
| 75 |
+
"step": 30
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.14644351464435146,
|
| 79 |
+
"grad_norm": 40.90021896362305,
|
| 80 |
+
"learning_rate": 1.999499435570359e-05,
|
| 81 |
+
"loss": 1.8907,
|
| 82 |
+
"num_input_tokens_seen": 286720,
|
| 83 |
+
"step": 35
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.16736401673640167,
|
| 87 |
+
"grad_norm": 40.66322326660156,
|
| 88 |
+
"learning_rate": 1.998529191458689e-05,
|
| 89 |
+
"loss": 2.0112,
|
| 90 |
+
"num_input_tokens_seen": 327680,
|
| 91 |
+
"step": 40
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.18828451882845187,
|
| 95 |
+
"grad_norm": 53.441314697265625,
|
| 96 |
+
"learning_rate": 1.9970488977329757e-05,
|
| 97 |
+
"loss": 2.1832,
|
| 98 |
+
"num_input_tokens_seen": 368640,
|
| 99 |
+
"step": 45
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.20920502092050208,
|
| 103 |
+
"grad_norm": 42.00421142578125,
|
| 104 |
+
"learning_rate": 1.9950593105285927e-05,
|
| 105 |
+
"loss": 1.8656,
|
| 106 |
+
"num_input_tokens_seen": 409600,
|
| 107 |
+
"step": 50
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.20920502092050208,
|
| 111 |
+
"eval_accuracy": 0.2358490566037736,
|
| 112 |
+
"eval_loss": 2.0404305458068848,
|
| 113 |
+
"eval_runtime": 1.7411,
|
| 114 |
+
"eval_samples_per_second": 121.76,
|
| 115 |
+
"eval_steps_per_second": 15.507,
|
| 116 |
+
"num_input_tokens_seen": 409600,
|
| 117 |
+
"step": 50
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.2301255230125523,
|
| 121 |
+
"grad_norm": 13.718694686889648,
|
| 122 |
+
"learning_rate": 1.992561446128484e-05,
|
| 123 |
+
"loss": 1.9829,
|
| 124 |
+
"num_input_tokens_seen": 450560,
|
| 125 |
+
"step": 55
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.2510460251046025,
|
| 129 |
+
"grad_norm": 53.16058349609375,
|
| 130 |
+
"learning_rate": 1.9895565804440435e-05,
|
| 131 |
+
"loss": 1.8415,
|
| 132 |
+
"num_input_tokens_seen": 491520,
|
| 133 |
+
"step": 60
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.2719665271966527,
|
| 137 |
+
"grad_norm": 39.195396423339844,
|
| 138 |
+
"learning_rate": 1.986046248363381e-05,
|
| 139 |
+
"loss": 2.4103,
|
| 140 |
+
"num_input_tokens_seen": 532480,
|
| 141 |
+
"step": 65
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.2928870292887029,
|
| 145 |
+
"grad_norm": 30.37375259399414,
|
| 146 |
+
"learning_rate": 1.9820322429672978e-05,
|
| 147 |
+
"loss": 2.0039,
|
| 148 |
+
"num_input_tokens_seen": 573440,
|
| 149 |
+
"step": 70
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3138075313807531,
|
| 153 |
+
"grad_norm": 34.187950134277344,
|
| 154 |
+
"learning_rate": 1.97751661461338e-05,
|
| 155 |
+
"loss": 1.901,
|
| 156 |
+
"num_input_tokens_seen": 614400,
|
| 157 |
+
"step": 75
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.33472803347280333,
|
| 161 |
+
"grad_norm": 26.08450698852539,
|
| 162 |
+
"learning_rate": 1.9725016698886748e-05,
|
| 163 |
+
"loss": 1.7461,
|
| 164 |
+
"num_input_tokens_seen": 655360,
|
| 165 |
+
"step": 80
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.35564853556485354,
|
| 169 |
+
"grad_norm": 22.850889205932617,
|
| 170 |
+
"learning_rate": 1.9669899704314828e-05,
|
| 171 |
+
"loss": 1.8595,
|
| 172 |
+
"num_input_tokens_seen": 696320,
|
| 173 |
+
"step": 85
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.37656903765690375,
|
| 177 |
+
"grad_norm": 30.59270668029785,
|
| 178 |
+
"learning_rate": 1.960984331622872e-05,
|
| 179 |
+
"loss": 1.9209,
|
| 180 |
+
"num_input_tokens_seen": 737280,
|
| 181 |
+
"step": 90
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.39748953974895396,
|
| 185 |
+
"grad_norm": 38.62232971191406,
|
| 186 |
+
"learning_rate": 1.9544878211485763e-05,
|
| 187 |
+
"loss": 1.7399,
|
| 188 |
+
"num_input_tokens_seen": 778240,
|
| 189 |
+
"step": 95
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.41841004184100417,
|
| 193 |
+
"grad_norm": 11.531844139099121,
|
| 194 |
+
"learning_rate": 1.9475037574320217e-05,
|
| 195 |
+
"loss": 1.8788,
|
| 196 |
+
"num_input_tokens_seen": 819200,
|
| 197 |
+
"step": 100
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.41841004184100417,
|
| 201 |
+
"eval_accuracy": 0.2169811320754717,
|
| 202 |
+
"eval_loss": 1.8193060159683228,
|
| 203 |
+
"eval_runtime": 1.7516,
|
| 204 |
+
"eval_samples_per_second": 121.029,
|
| 205 |
+
"eval_steps_per_second": 15.414,
|
| 206 |
+
"num_input_tokens_seen": 819200,
|
| 207 |
+
"step": 100
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.4393305439330544,
|
| 211 |
+
"grad_norm": 20.226055145263672,
|
| 212 |
+
"learning_rate": 1.9400357079392714e-05,
|
| 213 |
+
"loss": 1.8651,
|
| 214 |
+
"num_input_tokens_seen": 860160,
|
| 215 |
+
"step": 105
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.4602510460251046,
|
| 219 |
+
"grad_norm": 19.466136932373047,
|
| 220 |
+
"learning_rate": 1.9320874873567598e-05,
|
| 221 |
+
"loss": 1.8928,
|
| 222 |
+
"num_input_tokens_seen": 901120,
|
| 223 |
+
"step": 110
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.4811715481171548,
|
| 227 |
+
"grad_norm": 9.3947114944458,
|
| 228 |
+
"learning_rate": 1.923663155642748e-05,
|
| 229 |
+
"loss": 1.6679,
|
| 230 |
+
"num_input_tokens_seen": 942080,
|
| 231 |
+
"step": 115
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.502092050209205,
|
| 235 |
+
"grad_norm": 24.72301483154297,
|
| 236 |
+
"learning_rate": 1.9147670159534953e-05,
|
| 237 |
+
"loss": 1.9847,
|
| 238 |
+
"num_input_tokens_seen": 983040,
|
| 239 |
+
"step": 120
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.5230125523012552,
|
| 243 |
+
"grad_norm": 30.969099044799805,
|
| 244 |
+
"learning_rate": 1.9054036124452e-05,
|
| 245 |
+
"loss": 1.683,
|
| 246 |
+
"num_input_tokens_seen": 1024000,
|
| 247 |
+
"step": 125
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5439330543933054,
|
| 251 |
+
"grad_norm": 16.989919662475586,
|
| 252 |
+
"learning_rate": 1.8955777279528414e-05,
|
| 253 |
+
"loss": 1.9693,
|
| 254 |
+
"num_input_tokens_seen": 1064960,
|
| 255 |
+
"step": 130
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.5648535564853556,
|
| 259 |
+
"grad_norm": 31.935850143432617,
|
| 260 |
+
"learning_rate": 1.8852943815471058e-05,
|
| 261 |
+
"loss": 1.7894,
|
| 262 |
+
"num_input_tokens_seen": 1105920,
|
| 263 |
+
"step": 135
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.5857740585774058,
|
| 267 |
+
"grad_norm": 21.976633071899414,
|
| 268 |
+
"learning_rate": 1.8745588259706366e-05,
|
| 269 |
+
"loss": 1.7507,
|
| 270 |
+
"num_input_tokens_seen": 1146880,
|
| 271 |
+
"step": 140
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.606694560669456,
|
| 275 |
+
"grad_norm": 24.684856414794922,
|
| 276 |
+
"learning_rate": 1.863376544954931e-05,
|
| 277 |
+
"loss": 1.768,
|
| 278 |
+
"num_input_tokens_seen": 1187840,
|
| 279 |
+
"step": 145
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.6276150627615062,
|
| 283 |
+
"grad_norm": 34.6214599609375,
|
| 284 |
+
"learning_rate": 1.8517532504192456e-05,
|
| 285 |
+
"loss": 1.7635,
|
| 286 |
+
"num_input_tokens_seen": 1228800,
|
| 287 |
+
"step": 150
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.6276150627615062,
|
| 291 |
+
"eval_accuracy": 0.18867924528301888,
|
| 292 |
+
"eval_loss": 1.6324992179870605,
|
| 293 |
+
"eval_runtime": 1.7594,
|
| 294 |
+
"eval_samples_per_second": 120.495,
|
| 295 |
+
"eval_steps_per_second": 15.346,
|
| 296 |
+
"num_input_tokens_seen": 1228800,
|
| 297 |
+
"step": 150
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.6485355648535565,
|
| 301 |
+
"grad_norm": 33.269325256347656,
|
| 302 |
+
"learning_rate": 1.8396948795529405e-05,
|
| 303 |
+
"loss": 1.6621,
|
| 304 |
+
"num_input_tokens_seen": 1269760,
|
| 305 |
+
"step": 155
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.6694560669456067,
|
| 309 |
+
"grad_norm": 18.24401092529297,
|
| 310 |
+
"learning_rate": 1.8272075917827597e-05,
|
| 311 |
+
"loss": 1.7442,
|
| 312 |
+
"num_input_tokens_seen": 1310720,
|
| 313 |
+
"step": 160
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.6903765690376569,
|
| 317 |
+
"grad_norm": 22.88331413269043,
|
| 318 |
+
"learning_rate": 1.814297765626589e-05,
|
| 319 |
+
"loss": 1.8839,
|
| 320 |
+
"num_input_tokens_seen": 1351680,
|
| 321 |
+
"step": 165
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.7112970711297071,
|
| 325 |
+
"grad_norm": 42.35792922973633,
|
| 326 |
+
"learning_rate": 1.800971995435305e-05,
|
| 327 |
+
"loss": 1.8275,
|
| 328 |
+
"num_input_tokens_seen": 1392640,
|
| 329 |
+
"step": 170
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.7322175732217573,
|
| 333 |
+
"grad_norm": 16.296783447265625,
|
| 334 |
+
"learning_rate": 1.787237088024372e-05,
|
| 335 |
+
"loss": 1.6375,
|
| 336 |
+
"num_input_tokens_seen": 1433600,
|
| 337 |
+
"step": 175
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.7531380753138075,
|
| 341 |
+
"grad_norm": 32.96072769165039,
|
| 342 |
+
"learning_rate": 1.7731000591969182e-05,
|
| 343 |
+
"loss": 1.771,
|
| 344 |
+
"num_input_tokens_seen": 1474560,
|
| 345 |
+
"step": 180
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7740585774058577,
|
| 349 |
+
"grad_norm": 25.586259841918945,
|
| 350 |
+
"learning_rate": 1.758568130160053e-05,
|
| 351 |
+
"loss": 1.7087,
|
| 352 |
+
"num_input_tokens_seen": 1515520,
|
| 353 |
+
"step": 185
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.7949790794979079,
|
| 357 |
+
"grad_norm": 14.068694114685059,
|
| 358 |
+
"learning_rate": 1.743648723836271e-05,
|
| 359 |
+
"loss": 1.5889,
|
| 360 |
+
"num_input_tokens_seen": 1556480,
|
| 361 |
+
"step": 190
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.8158995815899581,
|
| 365 |
+
"grad_norm": 18.822080612182617,
|
| 366 |
+
"learning_rate": 1.7283494610718153e-05,
|
| 367 |
+
"loss": 1.7154,
|
| 368 |
+
"num_input_tokens_seen": 1597440,
|
| 369 |
+
"step": 195
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.8368200836820083,
|
| 373 |
+
"grad_norm": 8.740741729736328,
|
| 374 |
+
"learning_rate": 1.7126781567439418e-05,
|
| 375 |
+
"loss": 1.6773,
|
| 376 |
+
"num_input_tokens_seen": 1638400,
|
| 377 |
+
"step": 200
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.8368200836820083,
|
| 381 |
+
"eval_accuracy": 0.18867924528301888,
|
| 382 |
+
"eval_loss": 1.6925463676452637,
|
| 383 |
+
"eval_runtime": 1.7108,
|
| 384 |
+
"eval_samples_per_second": 123.919,
|
| 385 |
+
"eval_steps_per_second": 15.782,
|
| 386 |
+
"num_input_tokens_seen": 1638400,
|
| 387 |
+
"step": 200
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8577405857740585,
|
| 391 |
+
"grad_norm": 13.948569297790527,
|
| 392 |
+
"learning_rate": 1.696642815769075e-05,
|
| 393 |
+
"loss": 1.6262,
|
| 394 |
+
"num_input_tokens_seen": 1679360,
|
| 395 |
+
"step": 205
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.8786610878661087,
|
| 399 |
+
"grad_norm": 28.07710075378418,
|
| 400 |
+
"learning_rate": 1.680251629013885e-05,
|
| 401 |
+
"loss": 1.8119,
|
| 402 |
+
"num_input_tokens_seen": 1720320,
|
| 403 |
+
"step": 210
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.899581589958159,
|
| 407 |
+
"grad_norm": 26.009183883666992,
|
| 408 |
+
"learning_rate": 1.663512969111392e-05,
|
| 409 |
+
"loss": 1.723,
|
| 410 |
+
"num_input_tokens_seen": 1761280,
|
| 411 |
+
"step": 215
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.9205020920502092,
|
| 415 |
+
"grad_norm": 30.979652404785156,
|
| 416 |
+
"learning_rate": 1.6464353861842115e-05,
|
| 417 |
+
"loss": 1.7091,
|
| 418 |
+
"num_input_tokens_seen": 1802240,
|
| 419 |
+
"step": 220
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.9414225941422594,
|
| 423 |
+
"grad_norm": 29.44756317138672,
|
| 424 |
+
"learning_rate": 1.629027603477147e-05,
|
| 425 |
+
"loss": 1.6767,
|
| 426 |
+
"num_input_tokens_seen": 1843200,
|
| 427 |
+
"step": 225
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.9623430962343096,
|
| 431 |
+
"grad_norm": 27.677940368652344,
|
| 432 |
+
"learning_rate": 1.611298512901349e-05,
|
| 433 |
+
"loss": 1.7475,
|
| 434 |
+
"num_input_tokens_seen": 1884160,
|
| 435 |
+
"step": 230
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.9832635983263598,
|
| 439 |
+
"grad_norm": 23.408004760742188,
|
| 440 |
+
"learning_rate": 1.5932571704923168e-05,
|
| 441 |
+
"loss": 1.6768,
|
| 442 |
+
"num_input_tokens_seen": 1925120,
|
| 443 |
+
"step": 235
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.00418410041841,
|
| 447 |
+
"grad_norm": 36.993404388427734,
|
| 448 |
+
"learning_rate": 1.57491279178407e-05,
|
| 449 |
+
"loss": 1.6803,
|
| 450 |
+
"num_input_tokens_seen": 1966080,
|
| 451 |
+
"step": 240
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.0251046025104602,
|
| 455 |
+
"grad_norm": 40.108795166015625,
|
| 456 |
+
"learning_rate": 1.5562747471018415e-05,
|
| 457 |
+
"loss": 1.7311,
|
| 458 |
+
"num_input_tokens_seen": 2007040,
|
| 459 |
+
"step": 245
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.0460251046025104,
|
| 463 |
+
"grad_norm": 25.31317901611328,
|
| 464 |
+
"learning_rate": 1.5373525567757124e-05,
|
| 465 |
+
"loss": 1.6309,
|
| 466 |
+
"num_input_tokens_seen": 2048000,
|
| 467 |
+
"step": 250
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.0460251046025104,
|
| 471 |
+
"eval_accuracy": 0.19339622641509435,
|
| 472 |
+
"eval_loss": 1.6848835945129395,
|
| 473 |
+
"eval_runtime": 1.763,
|
| 474 |
+
"eval_samples_per_second": 120.249,
|
| 475 |
+
"eval_steps_per_second": 15.315,
|
| 476 |
+
"num_input_tokens_seen": 2048000,
|
| 477 |
+
"step": 250
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.0669456066945606,
|
| 481 |
+
"grad_norm": 31.694902420043945,
|
| 482 |
+
"learning_rate": 1.518155886277613e-05,
|
| 483 |
+
"loss": 1.655,
|
| 484 |
+
"num_input_tokens_seen": 2088960,
|
| 485 |
+
"step": 255
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.0878661087866108,
|
| 489 |
+
"grad_norm": 24.42173194885254,
|
| 490 |
+
"learning_rate": 1.498694541284195e-05,
|
| 491 |
+
"loss": 1.7565,
|
| 492 |
+
"num_input_tokens_seen": 2129920,
|
| 493 |
+
"step": 260
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.108786610878661,
|
| 497 |
+
"grad_norm": 30.931346893310547,
|
| 498 |
+
"learning_rate": 1.4789784626680819e-05,
|
| 499 |
+
"loss": 1.6379,
|
| 500 |
+
"num_input_tokens_seen": 2170880,
|
| 501 |
+
"step": 265
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.1297071129707112,
|
| 505 |
+
"grad_norm": 29.633745193481445,
|
| 506 |
+
"learning_rate": 1.4590177214200609e-05,
|
| 507 |
+
"loss": 1.6454,
|
| 508 |
+
"num_input_tokens_seen": 2211840,
|
| 509 |
+
"step": 270
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 1.1506276150627615,
|
| 513 |
+
"grad_norm": 9.598276138305664,
|
| 514 |
+
"learning_rate": 1.4388225135048137e-05,
|
| 515 |
+
"loss": 1.7242,
|
| 516 |
+
"num_input_tokens_seen": 2252800,
|
| 517 |
+
"step": 275
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 1.1715481171548117,
|
| 521 |
+
"grad_norm": 16.58527374267578,
|
| 522 |
+
"learning_rate": 1.4184031546528077e-05,
|
| 523 |
+
"loss": 1.646,
|
| 524 |
+
"num_input_tokens_seen": 2293760,
|
| 525 |
+
"step": 280
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.1924686192468619,
|
| 529 |
+
"grad_norm": 37.41482162475586,
|
| 530 |
+
"learning_rate": 1.3977700750910112e-05,
|
| 531 |
+
"loss": 1.7451,
|
| 532 |
+
"num_input_tokens_seen": 2334720,
|
| 533 |
+
"step": 285
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.213389121338912,
|
| 537 |
+
"grad_norm": 21.555646896362305,
|
| 538 |
+
"learning_rate": 1.3769338142151245e-05,
|
| 539 |
+
"loss": 1.7345,
|
| 540 |
+
"num_input_tokens_seen": 2375680,
|
| 541 |
+
"step": 290
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.2343096234309623,
|
| 545 |
+
"grad_norm": 31.439435958862305,
|
| 546 |
+
"learning_rate": 1.3559050152060465e-05,
|
| 547 |
+
"loss": 1.6293,
|
| 548 |
+
"num_input_tokens_seen": 2416640,
|
| 549 |
+
"step": 295
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.2552301255230125,
|
| 553 |
+
"grad_norm": 19.446046829223633,
|
| 554 |
+
"learning_rate": 1.3346944195933294e-05,
|
| 555 |
+
"loss": 1.5824,
|
| 556 |
+
"num_input_tokens_seen": 2457600,
|
| 557 |
+
"step": 300
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.2552301255230125,
|
| 561 |
+
"eval_accuracy": 0.18396226415094338,
|
| 562 |
+
"eval_loss": 1.8486742973327637,
|
| 563 |
+
"eval_runtime": 1.7646,
|
| 564 |
+
"eval_samples_per_second": 120.141,
|
| 565 |
+
"eval_steps_per_second": 15.301,
|
| 566 |
+
"num_input_tokens_seen": 2457600,
|
| 567 |
+
"step": 300
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.2761506276150627,
|
| 571 |
+
"grad_norm": 24.764543533325195,
|
| 572 |
+
"learning_rate": 1.313312861768394e-05,
|
| 573 |
+
"loss": 1.8709,
|
| 574 |
+
"num_input_tokens_seen": 2498560,
|
| 575 |
+
"step": 305
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.297071129707113,
|
| 579 |
+
"grad_norm": 27.600025177001953,
|
| 580 |
+
"learning_rate": 1.2917712634503148e-05,
|
| 581 |
+
"loss": 1.6984,
|
| 582 |
+
"num_input_tokens_seen": 2539520,
|
| 583 |
+
"step": 310
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.3179916317991631,
|
| 587 |
+
"grad_norm": 19.41140365600586,
|
| 588 |
+
"learning_rate": 1.2700806281069942e-05,
|
| 589 |
+
"loss": 1.6789,
|
| 590 |
+
"num_input_tokens_seen": 2580480,
|
| 591 |
+
"step": 315
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.3389121338912133,
|
| 595 |
+
"grad_norm": 15.373785018920898,
|
| 596 |
+
"learning_rate": 1.2482520353345819e-05,
|
| 597 |
+
"loss": 1.6207,
|
| 598 |
+
"num_input_tokens_seen": 2621440,
|
| 599 |
+
"step": 320
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.3598326359832635,
|
| 603 |
+
"grad_norm": 33.88786315917969,
|
| 604 |
+
"learning_rate": 1.2262966351980075e-05,
|
| 605 |
+
"loss": 1.6183,
|
| 606 |
+
"num_input_tokens_seen": 2662400,
|
| 607 |
+
"step": 325
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 1.3807531380753137,
|
| 611 |
+
"grad_norm": 28.47850799560547,
|
| 612 |
+
"learning_rate": 1.2042256425355165e-05,
|
| 613 |
+
"loss": 1.6355,
|
| 614 |
+
"num_input_tokens_seen": 2703360,
|
| 615 |
+
"step": 330
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.401673640167364,
|
| 619 |
+
"grad_norm": 23.044967651367188,
|
| 620 |
+
"learning_rate": 1.1820503312301218e-05,
|
| 621 |
+
"loss": 1.6644,
|
| 622 |
+
"num_input_tokens_seen": 2744320,
|
| 623 |
+
"step": 335
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 1.4225941422594142,
|
| 627 |
+
"grad_norm": 23.657499313354492,
|
| 628 |
+
"learning_rate": 1.1597820284508927e-05,
|
| 629 |
+
"loss": 1.6313,
|
| 630 |
+
"num_input_tokens_seen": 2785280,
|
| 631 |
+
"step": 340
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.4435146443514644,
|
| 635 |
+
"grad_norm": 40.53915023803711,
|
| 636 |
+
"learning_rate": 1.1374321088670277e-05,
|
| 637 |
+
"loss": 1.6638,
|
| 638 |
+
"num_input_tokens_seen": 2826240,
|
| 639 |
+
"step": 345
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.4644351464435146,
|
| 643 |
+
"grad_norm": 29.33917808532715,
|
| 644 |
+
"learning_rate": 1.1150119888376631e-05,
|
| 645 |
+
"loss": 1.8204,
|
| 646 |
+
"num_input_tokens_seen": 2867200,
|
| 647 |
+
"step": 350
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.4644351464435146,
|
| 651 |
+
"eval_accuracy": 0.18867924528301888,
|
| 652 |
+
"eval_loss": 1.6929585933685303,
|
| 653 |
+
"eval_runtime": 1.7722,
|
| 654 |
+
"eval_samples_per_second": 119.624,
|
| 655 |
+
"eval_steps_per_second": 15.235,
|
| 656 |
+
"num_input_tokens_seen": 2867200,
|
| 657 |
+
"step": 350
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 1.4853556485355648,
|
| 661 |
+
"grad_norm": 13.35934829711914,
|
| 662 |
+
"learning_rate": 1.0925331205803861e-05,
|
| 663 |
+
"loss": 1.7131,
|
| 664 |
+
"num_input_tokens_seen": 2908160,
|
| 665 |
+
"step": 355
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 1.506276150627615,
|
| 669 |
+
"grad_norm": 31.252412796020508,
|
| 670 |
+
"learning_rate": 1.0700069863214317e-05,
|
| 671 |
+
"loss": 1.591,
|
| 672 |
+
"num_input_tokens_seen": 2949120,
|
| 673 |
+
"step": 360
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 1.5271966527196654,
|
| 677 |
+
"grad_norm": 31.35171890258789,
|
| 678 |
+
"learning_rate": 1.047445092430552e-05,
|
| 679 |
+
"loss": 1.8913,
|
| 680 |
+
"num_input_tokens_seen": 2990080,
|
| 681 |
+
"step": 365
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.5481171548117154,
|
| 685 |
+
"grad_norm": 29.04215431213379,
|
| 686 |
+
"learning_rate": 1.0248589635435505e-05,
|
| 687 |
+
"loss": 1.6869,
|
| 688 |
+
"num_input_tokens_seen": 3031040,
|
| 689 |
+
"step": 370
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.5690376569037658,
|
| 693 |
+
"grad_norm": 26.4223575592041,
|
| 694 |
+
"learning_rate": 1.0022601366754889e-05,
|
| 695 |
+
"loss": 1.6154,
|
| 696 |
+
"num_input_tokens_seen": 3072000,
|
| 697 |
+
"step": 375
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.5899581589958158,
|
| 701 |
+
"grad_norm": 19.447614669799805,
|
| 702 |
+
"learning_rate": 9.796601553275697e-06,
|
| 703 |
+
"loss": 1.6481,
|
| 704 |
+
"num_input_tokens_seen": 3112960,
|
| 705 |
+
"step": 380
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 1.6108786610878663,
|
| 709 |
+
"grad_norm": 17.78988265991211,
|
| 710 |
+
"learning_rate": 9.570705635907038e-06,
|
| 711 |
+
"loss": 1.7014,
|
| 712 |
+
"num_input_tokens_seen": 3153920,
|
| 713 |
+
"step": 385
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.6317991631799162,
|
| 717 |
+
"grad_norm": 16.625778198242188,
|
| 718 |
+
"learning_rate": 9.3450290024878e-06,
|
| 719 |
+
"loss": 1.6126,
|
| 720 |
+
"num_input_tokens_seen": 3194880,
|
| 721 |
+
"step": 390
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.6527196652719667,
|
| 725 |
+
"grad_norm": 8.680732727050781,
|
| 726 |
+
"learning_rate": 9.119686928846437e-06,
|
| 727 |
+
"loss": 1.5989,
|
| 728 |
+
"num_input_tokens_seen": 3235840,
|
| 729 |
+
"step": 395
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.6736401673640167,
|
| 733 |
+
"grad_norm": 27.154508590698242,
|
| 734 |
+
"learning_rate": 8.894794519917947e-06,
|
| 735 |
+
"loss": 1.6639,
|
| 736 |
+
"num_input_tokens_seen": 3276800,
|
| 737 |
+
"step": 400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.6736401673640167,
|
| 741 |
+
"eval_accuracy": 0.21226415094339623,
|
| 742 |
+
"eval_loss": 1.6967244148254395,
|
| 743 |
+
"eval_runtime": 1.7729,
|
| 744 |
+
"eval_samples_per_second": 119.581,
|
| 745 |
+
"eval_steps_per_second": 15.23,
|
| 746 |
+
"num_input_tokens_seen": 3276800,
|
| 747 |
+
"step": 400
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 1.694560669456067,
|
| 751 |
+
"grad_norm": 19.39278793334961,
|
| 752 |
+
"learning_rate": 8.670466650948208e-06,
|
| 753 |
+
"loss": 1.6022,
|
| 754 |
+
"num_input_tokens_seen": 3317760,
|
| 755 |
+
"step": 405
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.715481171548117,
|
| 759 |
+
"grad_norm": 24.7038516998291,
|
| 760 |
+
"learning_rate": 8.446817908815566e-06,
|
| 761 |
+
"loss": 1.6768,
|
| 762 |
+
"num_input_tokens_seen": 3358720,
|
| 763 |
+
"step": 410
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 1.7364016736401675,
|
| 767 |
+
"grad_norm": 28.1064395904541,
|
| 768 |
+
"learning_rate": 8.223962533499757e-06,
|
| 769 |
+
"loss": 1.666,
|
| 770 |
+
"num_input_tokens_seen": 3399680,
|
| 771 |
+
"step": 415
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 1.7573221757322175,
|
| 775 |
+
"grad_norm": 22.78620147705078,
|
| 776 |
+
"learning_rate": 8.002014359728046e-06,
|
| 777 |
+
"loss": 1.6043,
|
| 778 |
+
"num_input_tokens_seen": 3440640,
|
| 779 |
+
"step": 420
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.778242677824268,
|
| 783 |
+
"grad_norm": 20.595088958740234,
|
| 784 |
+
"learning_rate": 7.781086758828314e-06,
|
| 785 |
+
"loss": 1.589,
|
| 786 |
+
"num_input_tokens_seen": 3481600,
|
| 787 |
+
"step": 425
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.799163179916318,
|
| 791 |
+
"grad_norm": 27.415950775146484,
|
| 792 |
+
"learning_rate": 7.561292580818906e-06,
|
| 793 |
+
"loss": 1.746,
|
| 794 |
+
"num_input_tokens_seen": 3522560,
|
| 795 |
+
"step": 430
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 1.8200836820083683,
|
| 799 |
+
"grad_norm": 18.025171279907227,
|
| 800 |
+
"learning_rate": 7.342744096764727e-06,
|
| 801 |
+
"loss": 1.5013,
|
| 802 |
+
"num_input_tokens_seen": 3563520,
|
| 803 |
+
"step": 435
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 1.8410041841004183,
|
| 807 |
+
"grad_norm": 30.85523223876953,
|
| 808 |
+
"learning_rate": 7.125552941429077e-06,
|
| 809 |
+
"loss": 1.9142,
|
| 810 |
+
"num_input_tokens_seen": 3604480,
|
| 811 |
+
"step": 440
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.8619246861924688,
|
| 815 |
+
"grad_norm": 24.6289005279541,
|
| 816 |
+
"learning_rate": 6.909830056250527e-06,
|
| 817 |
+
"loss": 1.7215,
|
| 818 |
+
"num_input_tokens_seen": 3645440,
|
| 819 |
+
"step": 445
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.8828451882845187,
|
| 823 |
+
"grad_norm": 19.287546157836914,
|
| 824 |
+
"learning_rate": 6.695685632673919e-06,
|
| 825 |
+
"loss": 1.5446,
|
| 826 |
+
"num_input_tokens_seen": 3686400,
|
| 827 |
+
"step": 450
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.8828451882845187,
|
| 831 |
+
"eval_accuracy": 0.22169811320754718,
|
| 832 |
+
"eval_loss": 1.656185507774353,
|
| 833 |
+
"eval_runtime": 1.7255,
|
| 834 |
+
"eval_samples_per_second": 122.861,
|
| 835 |
+
"eval_steps_per_second": 15.647,
|
| 836 |
+
"num_input_tokens_seen": 3686400,
|
| 837 |
+
"step": 450
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 1.9037656903765692,
|
| 841 |
+
"grad_norm": 10.62146282196045,
|
| 842 |
+
"learning_rate": 6.48322905586449e-06,
|
| 843 |
+
"loss": 1.6482,
|
| 844 |
+
"num_input_tokens_seen": 3727360,
|
| 845 |
+
"step": 455
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.9246861924686192,
|
| 849 |
+
"grad_norm": 31.397525787353516,
|
| 850 |
+
"learning_rate": 6.272568848833809e-06,
|
| 851 |
+
"loss": 1.6448,
|
| 852 |
+
"num_input_tokens_seen": 3768320,
|
| 853 |
+
"step": 460
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.9456066945606696,
|
| 857 |
+
"grad_norm": 22.08171844482422,
|
| 858 |
+
"learning_rate": 6.06381261700614e-06,
|
| 859 |
+
"loss": 1.6913,
|
| 860 |
+
"num_input_tokens_seen": 3809280,
|
| 861 |
+
"step": 465
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 1.9665271966527196,
|
| 865 |
+
"grad_norm": 9.43496322631836,
|
| 866 |
+
"learning_rate": 5.857066993253501e-06,
|
| 867 |
+
"loss": 1.5766,
|
| 868 |
+
"num_input_tokens_seen": 3850240,
|
| 869 |
+
"step": 470
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.98744769874477,
|
| 873 |
+
"grad_norm": 31.67534065246582,
|
| 874 |
+
"learning_rate": 5.652437583427478e-06,
|
| 875 |
+
"loss": 1.7616,
|
| 876 |
+
"num_input_tokens_seen": 3891200,
|
| 877 |
+
"step": 475
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.00836820083682,
|
| 881 |
+
"grad_norm": 22.396835327148438,
|
| 882 |
+
"learning_rate": 5.450028912415672e-06,
|
| 883 |
+
"loss": 1.6519,
|
| 884 |
+
"num_input_tokens_seen": 3932160,
|
| 885 |
+
"step": 480
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.0292887029288704,
|
| 889 |
+
"grad_norm": 17.868091583251953,
|
| 890 |
+
"learning_rate": 5.249944370750293e-06,
|
| 891 |
+
"loss": 1.5843,
|
| 892 |
+
"num_input_tokens_seen": 3973120,
|
| 893 |
+
"step": 485
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.0502092050209204,
|
| 897 |
+
"grad_norm": 26.96571159362793,
|
| 898 |
+
"learning_rate": 5.0522861617961694e-06,
|
| 899 |
+
"loss": 1.5715,
|
| 900 |
+
"num_input_tokens_seen": 4014080,
|
| 901 |
+
"step": 490
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 2.071129707112971,
|
| 905 |
+
"grad_norm": 10.547337532043457,
|
| 906 |
+
"learning_rate": 4.857155249545197e-06,
|
| 907 |
+
"loss": 1.6168,
|
| 908 |
+
"num_input_tokens_seen": 4055040,
|
| 909 |
+
"step": 495
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.092050209205021,
|
| 913 |
+
"grad_norm": 33.86396408081055,
|
| 914 |
+
"learning_rate": 4.66465130704382e-06,
|
| 915 |
+
"loss": 1.569,
|
| 916 |
+
"num_input_tokens_seen": 4096000,
|
| 917 |
+
"step": 500
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 2.092050209205021,
|
| 921 |
+
"eval_accuracy": 0.21226415094339623,
|
| 922 |
+
"eval_loss": 1.615054726600647,
|
| 923 |
+
"eval_runtime": 1.7232,
|
| 924 |
+
"eval_samples_per_second": 123.027,
|
| 925 |
+
"eval_steps_per_second": 15.669,
|
| 926 |
+
"num_input_tokens_seen": 4096000,
|
| 927 |
+
"step": 500
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.1129707112970713,
|
| 931 |
+
"grad_norm": 27.13317108154297,
|
| 932 |
+
"learning_rate": 4.474872665479974e-06,
|
| 933 |
+
"loss": 1.5502,
|
| 934 |
+
"num_input_tokens_seen": 4136960,
|
| 935 |
+
"step": 505
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.1338912133891212,
|
| 939 |
+
"grad_norm": 31.335521697998047,
|
| 940 |
+
"learning_rate": 4.2879162639553925e-06,
|
| 941 |
+
"loss": 1.6434,
|
| 942 |
+
"num_input_tokens_seen": 4177920,
|
| 943 |
+
"step": 510
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 2.1548117154811717,
|
| 947 |
+
"grad_norm": 24.757951736450195,
|
| 948 |
+
"learning_rate": 4.103877599969056e-06,
|
| 949 |
+
"loss": 1.6334,
|
| 950 |
+
"num_input_tokens_seen": 4218880,
|
| 951 |
+
"step": 515
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 2.1757322175732217,
|
| 955 |
+
"grad_norm": 23.463459014892578,
|
| 956 |
+
"learning_rate": 3.922850680636968e-06,
|
| 957 |
+
"loss": 1.6167,
|
| 958 |
+
"num_input_tokens_seen": 4259840,
|
| 959 |
+
"step": 520
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 2.196652719665272,
|
| 963 |
+
"grad_norm": 22.036346435546875,
|
| 964 |
+
"learning_rate": 3.744927974673237e-06,
|
| 965 |
+
"loss": 1.6218,
|
| 966 |
+
"num_input_tokens_seen": 4300800,
|
| 967 |
+
"step": 525
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 2.217573221757322,
|
| 971 |
+
"grad_norm": 26.18072509765625,
|
| 972 |
+
"learning_rate": 3.5702003651569883e-06,
|
| 973 |
+
"loss": 1.5746,
|
| 974 |
+
"num_input_tokens_seen": 4341760,
|
| 975 |
+
"step": 530
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.2384937238493725,
|
| 979 |
+
"grad_norm": 15.42270565032959,
|
| 980 |
+
"learning_rate": 3.3987571031091735e-06,
|
| 981 |
+
"loss": 1.5657,
|
| 982 |
+
"num_input_tokens_seen": 4382720,
|
| 983 |
+
"step": 535
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 2.2594142259414225,
|
| 987 |
+
"grad_norm": 18.54688262939453,
|
| 988 |
+
"learning_rate": 3.230685761903094e-06,
|
| 989 |
+
"loss": 1.5928,
|
| 990 |
+
"num_input_tokens_seen": 4423680,
|
| 991 |
+
"step": 540
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.280334728033473,
|
| 995 |
+
"grad_norm": 27.211170196533203,
|
| 996 |
+
"learning_rate": 3.0660721925318483e-06,
|
| 997 |
+
"loss": 1.5975,
|
| 998 |
+
"num_input_tokens_seen": 4464640,
|
| 999 |
+
"step": 545
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.301255230125523,
|
| 1003 |
+
"grad_norm": 26.673534393310547,
|
| 1004 |
+
"learning_rate": 2.905000479755531e-06,
|
| 1005 |
+
"loss": 1.5797,
|
| 1006 |
+
"num_input_tokens_seen": 4505600,
|
| 1007 |
+
"step": 550
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.301255230125523,
|
| 1011 |
+
"eval_accuracy": 0.23113207547169812,
|
| 1012 |
+
"eval_loss": 1.6243550777435303,
|
| 1013 |
+
"eval_runtime": 1.7276,
|
| 1014 |
+
"eval_samples_per_second": 122.712,
|
| 1015 |
+
"eval_steps_per_second": 15.628,
|
| 1016 |
+
"num_input_tokens_seen": 4505600,
|
| 1017 |
+
"step": 550
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.3221757322175733,
|
| 1021 |
+
"grad_norm": 26.736270904541016,
|
| 1022 |
+
"learning_rate": 2.7475528991507106e-06,
|
| 1023 |
+
"loss": 1.6087,
|
| 1024 |
+
"num_input_tokens_seen": 4546560,
|
| 1025 |
+
"step": 555
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 2.3430962343096233,
|
| 1029 |
+
"grad_norm": 10.619952201843262,
|
| 1030 |
+
"learning_rate": 2.5938098750839414e-06,
|
| 1031 |
+
"loss": 1.5212,
|
| 1032 |
+
"num_input_tokens_seen": 4587520,
|
| 1033 |
+
"step": 560
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 2.3640167364016738,
|
| 1037 |
+
"grad_norm": 30.245092391967773,
|
| 1038 |
+
"learning_rate": 2.443849939630959e-06,
|
| 1039 |
+
"loss": 1.538,
|
| 1040 |
+
"num_input_tokens_seen": 4628480,
|
| 1041 |
+
"step": 565
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 2.3849372384937237,
|
| 1045 |
+
"grad_norm": 17.95905113220215,
|
| 1046 |
+
"learning_rate": 2.2977496924624223e-06,
|
| 1047 |
+
"loss": 1.5952,
|
| 1048 |
+
"num_input_tokens_seen": 4669440,
|
| 1049 |
+
"step": 570
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 2.405857740585774,
|
| 1053 |
+
"grad_norm": 18.98442840576172,
|
| 1054 |
+
"learning_rate": 2.155583761716703e-06,
|
| 1055 |
+
"loss": 1.6038,
|
| 1056 |
+
"num_input_tokens_seen": 4710400,
|
| 1057 |
+
"step": 575
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 2.426778242677824,
|
| 1061 |
+
"grad_norm": 23.98601722717285,
|
| 1062 |
+
"learning_rate": 2.0174247658798054e-06,
|
| 1063 |
+
"loss": 1.6029,
|
| 1064 |
+
"num_input_tokens_seen": 4751360,
|
| 1065 |
+
"step": 580
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.4476987447698746,
|
| 1069 |
+
"grad_norm": 28.006078720092773,
|
| 1070 |
+
"learning_rate": 1.8833432766917514e-06,
|
| 1071 |
+
"loss": 1.659,
|
| 1072 |
+
"num_input_tokens_seen": 4792320,
|
| 1073 |
+
"step": 585
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.4686192468619246,
|
| 1077 |
+
"grad_norm": 34.81071090698242,
|
| 1078 |
+
"learning_rate": 1.7534077830984697e-06,
|
| 1079 |
+
"loss": 1.5578,
|
| 1080 |
+
"num_input_tokens_seen": 4833280,
|
| 1081 |
+
"step": 590
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.489539748953975,
|
| 1085 |
+
"grad_norm": 27.956401824951172,
|
| 1086 |
+
"learning_rate": 1.6276846562676085e-06,
|
| 1087 |
+
"loss": 1.5686,
|
| 1088 |
+
"num_input_tokens_seen": 4874240,
|
| 1089 |
+
"step": 595
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.510460251046025,
|
| 1093 |
+
"grad_norm": 16.93408203125,
|
| 1094 |
+
"learning_rate": 1.506238115686044e-06,
|
| 1095 |
+
"loss": 1.5543,
|
| 1096 |
+
"num_input_tokens_seen": 4915200,
|
| 1097 |
+
"step": 600
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.510460251046025,
|
| 1101 |
+
"eval_accuracy": 0.2028301886792453,
|
| 1102 |
+
"eval_loss": 1.6461066007614136,
|
| 1103 |
+
"eval_runtime": 1.7336,
|
| 1104 |
+
"eval_samples_per_second": 122.287,
|
| 1105 |
+
"eval_steps_per_second": 15.574,
|
| 1106 |
+
"num_input_tokens_seen": 4915200,
|
| 1107 |
+
"step": 600
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 2.5313807531380754,
|
| 1111 |
+
"grad_norm": 20.968040466308594,
|
| 1112 |
+
"learning_rate": 1.3891301963565473e-06,
|
| 1113 |
+
"loss": 1.6177,
|
| 1114 |
+
"num_input_tokens_seen": 4956160,
|
| 1115 |
+
"step": 605
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.5523012552301254,
|
| 1119 |
+
"grad_norm": 24.227895736694336,
|
| 1120 |
+
"learning_rate": 1.2764207171102206e-06,
|
| 1121 |
+
"loss": 1.6219,
|
| 1122 |
+
"num_input_tokens_seen": 4997120,
|
| 1123 |
+
"step": 610
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 2.573221757322176,
|
| 1127 |
+
"grad_norm": 18.555587768554688,
|
| 1128 |
+
"learning_rate": 1.1681672500509866e-06,
|
| 1129 |
+
"loss": 1.5723,
|
| 1130 |
+
"num_input_tokens_seen": 5038080,
|
| 1131 |
+
"step": 615
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.594142259414226,
|
| 1135 |
+
"grad_norm": 32.787696838378906,
|
| 1136 |
+
"learning_rate": 1.0644250911477306e-06,
|
| 1137 |
+
"loss": 1.5555,
|
| 1138 |
+
"num_input_tokens_seen": 5079040,
|
| 1139 |
+
"step": 620
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 2.6150627615062763,
|
| 1143 |
+
"grad_norm": 30.15579605102539,
|
| 1144 |
+
"learning_rate": 9.652472319890372e-07,
|
| 1145 |
+
"loss": 1.5715,
|
| 1146 |
+
"num_input_tokens_seen": 5120000,
|
| 1147 |
+
"step": 625
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 2.6359832635983262,
|
| 1151 |
+
"grad_norm": 27.77329444885254,
|
| 1152 |
+
"learning_rate": 8.706843327150605e-07,
|
| 1153 |
+
"loss": 1.5831,
|
| 1154 |
+
"num_input_tokens_seen": 5160960,
|
| 1155 |
+
"step": 630
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 2.6569037656903767,
|
| 1159 |
+
"grad_norm": 21.68170928955078,
|
| 1160 |
+
"learning_rate": 7.807846961402699e-07,
|
| 1161 |
+
"loss": 1.5731,
|
| 1162 |
+
"num_input_tokens_seen": 5201920,
|
| 1163 |
+
"step": 635
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 2.6778242677824267,
|
| 1167 |
+
"grad_norm": 14.29506778717041,
|
| 1168 |
+
"learning_rate": 6.955942430803298e-07,
|
| 1169 |
+
"loss": 1.6046,
|
| 1170 |
+
"num_input_tokens_seen": 5242880,
|
| 1171 |
+
"step": 640
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.698744769874477,
|
| 1175 |
+
"grad_norm": 26.25128746032715,
|
| 1176 |
+
"learning_rate": 6.151564888957084e-07,
|
| 1177 |
+
"loss": 1.5819,
|
| 1178 |
+
"num_input_tokens_seen": 5283840,
|
| 1179 |
+
"step": 645
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.719665271966527,
|
| 1183 |
+
"grad_norm": 20.036659240722656,
|
| 1184 |
+
"learning_rate": 5.395125212639895e-07,
|
| 1185 |
+
"loss": 1.5691,
|
| 1186 |
+
"num_input_tokens_seen": 5324800,
|
| 1187 |
+
"step": 650
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.719665271966527,
|
| 1191 |
+
"eval_accuracy": 0.20754716981132076,
|
| 1192 |
+
"eval_loss": 1.6239819526672363,
|
| 1193 |
+
"eval_runtime": 1.7176,
|
| 1194 |
+
"eval_samples_per_second": 123.431,
|
| 1195 |
+
"eval_steps_per_second": 15.72,
|
| 1196 |
+
"num_input_tokens_seen": 5324800,
|
| 1197 |
+
"step": 650
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 2.7405857740585775,
|
| 1201 |
+
"grad_norm": 24.415496826171875,
|
| 1202 |
+
"learning_rate": 4.6870097919224923e-07,
|
| 1203 |
+
"loss": 1.6025,
|
| 1204 |
+
"num_input_tokens_seen": 5365760,
|
| 1205 |
+
"step": 655
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 2.7615062761506275,
|
| 1209 |
+
"grad_norm": 28.49026870727539,
|
| 1210 |
+
"learning_rate": 4.0275803328021946e-07,
|
| 1211 |
+
"loss": 1.6376,
|
| 1212 |
+
"num_input_tokens_seen": 5406720,
|
| 1213 |
+
"step": 660
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.782426778242678,
|
| 1217 |
+
"grad_norm": 27.395978927612305,
|
| 1218 |
+
"learning_rate": 3.417173672443075e-07,
|
| 1219 |
+
"loss": 1.5553,
|
| 1220 |
+
"num_input_tokens_seen": 5447680,
|
| 1221 |
+
"step": 665
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 2.803347280334728,
|
| 1225 |
+
"grad_norm": 20.28337287902832,
|
| 1226 |
+
"learning_rate": 2.8561016071192884e-07,
|
| 1227 |
+
"loss": 1.5418,
|
| 1228 |
+
"num_input_tokens_seen": 5488640,
|
| 1229 |
+
"step": 670
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.8242677824267783,
|
| 1233 |
+
"grad_norm": 10.513223648071289,
|
| 1234 |
+
"learning_rate": 2.3446507329492274e-07,
|
| 1235 |
+
"loss": 1.5828,
|
| 1236 |
+
"num_input_tokens_seen": 5529600,
|
| 1237 |
+
"step": 675
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 2.8451882845188283,
|
| 1241 |
+
"grad_norm": 21.558700561523438,
|
| 1242 |
+
"learning_rate": 1.8830822995019593e-07,
|
| 1243 |
+
"loss": 1.5803,
|
| 1244 |
+
"num_input_tokens_seen": 5570560,
|
| 1245 |
+
"step": 680
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 2.8661087866108788,
|
| 1249 |
+
"grad_norm": 33.30543518066406,
|
| 1250 |
+
"learning_rate": 1.4716320763507152e-07,
|
| 1251 |
+
"loss": 1.5931,
|
| 1252 |
+
"num_input_tokens_seen": 5611520,
|
| 1253 |
+
"step": 685
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 2.8870292887029287,
|
| 1257 |
+
"grad_norm": 32.0794792175293,
|
| 1258 |
+
"learning_rate": 1.1105102326415929e-07,
|
| 1259 |
+
"loss": 1.5912,
|
| 1260 |
+
"num_input_tokens_seen": 5652480,
|
| 1261 |
+
"step": 690
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 2.907949790794979,
|
| 1265 |
+
"grad_norm": 17.398990631103516,
|
| 1266 |
+
"learning_rate": 7.999012297389419e-08,
|
| 1267 |
+
"loss": 1.5175,
|
| 1268 |
+
"num_input_tokens_seen": 5693440,
|
| 1269 |
+
"step": 695
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.928870292887029,
|
| 1273 |
+
"grad_norm": 27.844022750854492,
|
| 1274 |
+
"learning_rate": 5.399637270023683e-08,
|
| 1275 |
+
"loss": 1.5852,
|
| 1276 |
+
"num_input_tokens_seen": 5734400,
|
| 1277 |
+
"step": 700
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.928870292887029,
|
| 1281 |
+
"eval_accuracy": 0.2169811320754717,
|
| 1282 |
+
"eval_loss": 1.622738242149353,
|
| 1283 |
+
"eval_runtime": 1.7709,
|
| 1284 |
+
"eval_samples_per_second": 119.711,
|
| 1285 |
+
"eval_steps_per_second": 15.246,
|
| 1286 |
+
"num_input_tokens_seen": 5734400,
|
| 1287 |
+
"step": 700
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 2.9497907949790796,
|
| 1291 |
+
"grad_norm": 20.81698226928711,
|
| 1292 |
+
"learning_rate": 3.308305007433399e-08,
|
| 1293 |
+
"loss": 1.5199,
|
| 1294 |
+
"num_input_tokens_seen": 5775360,
|
| 1295 |
+
"step": 705
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 2.9707112970711296,
|
| 1299 |
+
"grad_norm": 21.446496963500977,
|
| 1300 |
+
"learning_rate": 1.726083764029607e-08,
|
| 1301 |
+
"loss": 1.5656,
|
| 1302 |
+
"num_input_tokens_seen": 5816320,
|
| 1303 |
+
"step": 710
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 2.99163179916318,
|
| 1307 |
+
"grad_norm": 15.635223388671875,
|
| 1308 |
+
"learning_rate": 6.5378173985441994e-09,
|
| 1309 |
+
"loss": 1.545,
|
| 1310 |
+
"num_input_tokens_seen": 5857280,
|
| 1311 |
+
"step": 715
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.0,
|
| 1315 |
+
"num_input_tokens_seen": 5873664,
|
| 1316 |
+
"step": 717,
|
| 1317 |
+
"total_flos": 1.0654234806583296e+16,
|
| 1318 |
+
"train_loss": 1.7262512909318612,
|
| 1319 |
+
"train_runtime": 402.0358,
|
| 1320 |
+
"train_samples_per_second": 14.215,
|
| 1321 |
+
"train_steps_per_second": 1.783
|
| 1322 |
+
}
|
| 1323 |
+
],
|
| 1324 |
+
"logging_steps": 5,
|
| 1325 |
+
"max_steps": 717,
|
| 1326 |
+
"num_input_tokens_seen": 5873664,
|
| 1327 |
+
"num_train_epochs": 3,
|
| 1328 |
+
"save_steps": 50,
|
| 1329 |
+
"stateful_callbacks": {
|
| 1330 |
+
"TrainerControl": {
|
| 1331 |
+
"args": {
|
| 1332 |
+
"should_epoch_stop": false,
|
| 1333 |
+
"should_evaluate": false,
|
| 1334 |
+
"should_log": false,
|
| 1335 |
+
"should_save": true,
|
| 1336 |
+
"should_training_stop": true
|
| 1337 |
+
},
|
| 1338 |
+
"attributes": {}
|
| 1339 |
+
}
|
| 1340 |
+
},
|
| 1341 |
+
"total_flos": 1.0654234806583296e+16,
|
| 1342 |
+
"train_batch_size": 4,
|
| 1343 |
+
"trial_name": null,
|
| 1344 |
+
"trial_params": null
|
| 1345 |
+
}
|