End of training
Browse files- README.md +4 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +9 -0
- trainer_state.json +1127 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: train_cb_1757340241
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# train_cb_1757340241
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss:
|
| 21 |
-
- Num Input Tokens Seen:
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- p-tuning
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: train_cb_1757340241
|
|
|
|
| 16 |
|
| 17 |
# train_cb_1757340241
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the cb dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.2107
|
| 22 |
+
- Num Input Tokens Seen: 352296
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 2.21065354347229,
|
| 4 |
+
"eval_runtime": 0.981,
|
| 5 |
+
"eval_samples_per_second": 25.483,
|
| 6 |
+
"eval_steps_per_second": 7.135,
|
| 7 |
+
"num_input_tokens_seen": 352296,
|
| 8 |
+
"total_flos": 1.5863729653481472e+16,
|
| 9 |
+
"train_loss": 0.24723993078785084,
|
| 10 |
+
"train_runtime": 184.7232,
|
| 11 |
+
"train_samples_per_second": 12.18,
|
| 12 |
+
"train_steps_per_second": 3.086
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 2.21065354347229,
|
| 4 |
+
"eval_runtime": 0.981,
|
| 5 |
+
"eval_samples_per_second": 25.483,
|
| 6 |
+
"eval_steps_per_second": 7.135,
|
| 7 |
+
"num_input_tokens_seen": 352296
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"num_input_tokens_seen": 352296,
|
| 4 |
+
"total_flos": 1.5863729653481472e+16,
|
| 5 |
+
"train_loss": 0.24723993078785084,
|
| 6 |
+
"train_runtime": 184.7232,
|
| 7 |
+
"train_samples_per_second": 12.18,
|
| 8 |
+
"train_steps_per_second": 3.086
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1127 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 203,
|
| 3 |
+
"best_metric": 0.008254488930106163,
|
| 4 |
+
"best_model_checkpoint": "saves_stability/p-tuning/llama-3-8b-instruct/train_cb_1757340241/checkpoint-203",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 29,
|
| 7 |
+
"global_step": 570,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 77.47952270507812,
|
| 15 |
+
"learning_rate": 3.5087719298245615e-06,
|
| 16 |
+
"loss": 6.0986,
|
| 17 |
+
"num_input_tokens_seen": 3040,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.17543859649122806,
|
| 22 |
+
"grad_norm": 150.34814453125,
|
| 23 |
+
"learning_rate": 7.894736842105263e-06,
|
| 24 |
+
"loss": 3.9469,
|
| 25 |
+
"num_input_tokens_seen": 5984,
|
| 26 |
+
"step": 10
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.2631578947368421,
|
| 30 |
+
"grad_norm": 95.17002868652344,
|
| 31 |
+
"learning_rate": 1.2280701754385964e-05,
|
| 32 |
+
"loss": 2.7365,
|
| 33 |
+
"num_input_tokens_seen": 9952,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.3508771929824561,
|
| 38 |
+
"grad_norm": 34.78875732421875,
|
| 39 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 40 |
+
"loss": 1.7607,
|
| 41 |
+
"num_input_tokens_seen": 13120,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.43859649122807015,
|
| 46 |
+
"grad_norm": 94.67193603515625,
|
| 47 |
+
"learning_rate": 2.105263157894737e-05,
|
| 48 |
+
"loss": 1.2308,
|
| 49 |
+
"num_input_tokens_seen": 16128,
|
| 50 |
+
"step": 25
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.5087719298245614,
|
| 54 |
+
"eval_loss": 0.5064517259597778,
|
| 55 |
+
"eval_runtime": 0.9896,
|
| 56 |
+
"eval_samples_per_second": 25.263,
|
| 57 |
+
"eval_steps_per_second": 7.074,
|
| 58 |
+
"num_input_tokens_seen": 18528,
|
| 59 |
+
"step": 29
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.5263157894736842,
|
| 63 |
+
"grad_norm": 22.95918083190918,
|
| 64 |
+
"learning_rate": 2.5438596491228074e-05,
|
| 65 |
+
"loss": 0.7794,
|
| 66 |
+
"num_input_tokens_seen": 18944,
|
| 67 |
+
"step": 30
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.6140350877192983,
|
| 71 |
+
"grad_norm": 17.1202392578125,
|
| 72 |
+
"learning_rate": 2.9824561403508772e-05,
|
| 73 |
+
"loss": 0.3722,
|
| 74 |
+
"num_input_tokens_seen": 21984,
|
| 75 |
+
"step": 35
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.7017543859649122,
|
| 79 |
+
"grad_norm": 13.455132484436035,
|
| 80 |
+
"learning_rate": 3.421052631578947e-05,
|
| 81 |
+
"loss": 0.2511,
|
| 82 |
+
"num_input_tokens_seen": 25632,
|
| 83 |
+
"step": 40
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.7894736842105263,
|
| 87 |
+
"grad_norm": 43.001953125,
|
| 88 |
+
"learning_rate": 3.859649122807018e-05,
|
| 89 |
+
"loss": 0.359,
|
| 90 |
+
"num_input_tokens_seen": 28416,
|
| 91 |
+
"step": 45
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.8771929824561403,
|
| 95 |
+
"grad_norm": 14.863762855529785,
|
| 96 |
+
"learning_rate": 4.298245614035088e-05,
|
| 97 |
+
"loss": 0.3828,
|
| 98 |
+
"num_input_tokens_seen": 31264,
|
| 99 |
+
"step": 50
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.9649122807017544,
|
| 103 |
+
"grad_norm": 5.473771095275879,
|
| 104 |
+
"learning_rate": 4.736842105263158e-05,
|
| 105 |
+
"loss": 0.3097,
|
| 106 |
+
"num_input_tokens_seen": 34528,
|
| 107 |
+
"step": 55
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.0175438596491229,
|
| 111 |
+
"eval_loss": 1.1115503311157227,
|
| 112 |
+
"eval_runtime": 0.9873,
|
| 113 |
+
"eval_samples_per_second": 25.323,
|
| 114 |
+
"eval_steps_per_second": 7.09,
|
| 115 |
+
"num_input_tokens_seen": 35960,
|
| 116 |
+
"step": 58
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.0526315789473684,
|
| 120 |
+
"grad_norm": 33.196285247802734,
|
| 121 |
+
"learning_rate": 4.999812487773597e-05,
|
| 122 |
+
"loss": 0.3996,
|
| 123 |
+
"num_input_tokens_seen": 37464,
|
| 124 |
+
"step": 60
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.1403508771929824,
|
| 128 |
+
"grad_norm": 24.45400047302246,
|
| 129 |
+
"learning_rate": 4.997703298253406e-05,
|
| 130 |
+
"loss": 0.4544,
|
| 131 |
+
"num_input_tokens_seen": 40568,
|
| 132 |
+
"step": 65
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.2280701754385965,
|
| 136 |
+
"grad_norm": 30.591833114624023,
|
| 137 |
+
"learning_rate": 4.993252512887069e-05,
|
| 138 |
+
"loss": 0.5608,
|
| 139 |
+
"num_input_tokens_seen": 43576,
|
| 140 |
+
"step": 70
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.3157894736842106,
|
| 144 |
+
"grad_norm": 15.184714317321777,
|
| 145 |
+
"learning_rate": 4.986464304284091e-05,
|
| 146 |
+
"loss": 0.3779,
|
| 147 |
+
"num_input_tokens_seen": 46392,
|
| 148 |
+
"step": 75
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.4035087719298245,
|
| 152 |
+
"grad_norm": 17.91200828552246,
|
| 153 |
+
"learning_rate": 4.977345036387331e-05,
|
| 154 |
+
"loss": 0.2358,
|
| 155 |
+
"num_input_tokens_seen": 49560,
|
| 156 |
+
"step": 80
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.4912280701754386,
|
| 160 |
+
"grad_norm": 16.38266372680664,
|
| 161 |
+
"learning_rate": 4.965903258506806e-05,
|
| 162 |
+
"loss": 2.2471,
|
| 163 |
+
"num_input_tokens_seen": 52184,
|
| 164 |
+
"step": 85
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.526315789473684,
|
| 168 |
+
"eval_loss": 0.1543671041727066,
|
| 169 |
+
"eval_runtime": 0.9763,
|
| 170 |
+
"eval_samples_per_second": 25.608,
|
| 171 |
+
"eval_steps_per_second": 7.17,
|
| 172 |
+
"num_input_tokens_seen": 53272,
|
| 173 |
+
"step": 87
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.5789473684210527,
|
| 177 |
+
"grad_norm": 3.6652369499206543,
|
| 178 |
+
"learning_rate": 4.952149697304716e-05,
|
| 179 |
+
"loss": 0.1703,
|
| 180 |
+
"num_input_tokens_seen": 55128,
|
| 181 |
+
"step": 90
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.6666666666666665,
|
| 185 |
+
"grad_norm": 4.727101802825928,
|
| 186 |
+
"learning_rate": 4.9360972467392056e-05,
|
| 187 |
+
"loss": 0.2016,
|
| 188 |
+
"num_input_tokens_seen": 58456,
|
| 189 |
+
"step": 95
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.7543859649122808,
|
| 193 |
+
"grad_norm": 3.6023104190826416,
|
| 194 |
+
"learning_rate": 4.917760955976277e-05,
|
| 195 |
+
"loss": 0.268,
|
| 196 |
+
"num_input_tokens_seen": 61272,
|
| 197 |
+
"step": 100
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.8421052631578947,
|
| 201 |
+
"grad_norm": 10.225960731506348,
|
| 202 |
+
"learning_rate": 4.897158015281209e-05,
|
| 203 |
+
"loss": 0.2217,
|
| 204 |
+
"num_input_tokens_seen": 64952,
|
| 205 |
+
"step": 105
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.9298245614035088,
|
| 209 |
+
"grad_norm": 2.08432936668396,
|
| 210 |
+
"learning_rate": 4.874307739902689e-05,
|
| 211 |
+
"loss": 0.2854,
|
| 212 |
+
"num_input_tokens_seen": 67896,
|
| 213 |
+
"step": 110
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.017543859649123,
|
| 217 |
+
"grad_norm": 40.928401947021484,
|
| 218 |
+
"learning_rate": 4.849231551964771e-05,
|
| 219 |
+
"loss": 0.1901,
|
| 220 |
+
"num_input_tokens_seen": 70592,
|
| 221 |
+
"step": 115
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.0350877192982457,
|
| 225 |
+
"eval_loss": 0.17909584939479828,
|
| 226 |
+
"eval_runtime": 0.9756,
|
| 227 |
+
"eval_samples_per_second": 25.624,
|
| 228 |
+
"eval_steps_per_second": 7.175,
|
| 229 |
+
"num_input_tokens_seen": 71200,
|
| 230 |
+
"step": 116
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.1052631578947367,
|
| 234 |
+
"grad_norm": 0.7482051253318787,
|
| 235 |
+
"learning_rate": 4.821952960383649e-05,
|
| 236 |
+
"loss": 0.0757,
|
| 237 |
+
"num_input_tokens_seen": 73728,
|
| 238 |
+
"step": 120
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.192982456140351,
|
| 242 |
+
"grad_norm": 0.3453710079193115,
|
| 243 |
+
"learning_rate": 4.7924975388280524e-05,
|
| 244 |
+
"loss": 0.3489,
|
| 245 |
+
"num_input_tokens_seen": 76992,
|
| 246 |
+
"step": 125
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.280701754385965,
|
| 250 |
+
"grad_norm": 12.241211891174316,
|
| 251 |
+
"learning_rate": 4.760892901743944e-05,
|
| 252 |
+
"loss": 0.2132,
|
| 253 |
+
"num_input_tokens_seen": 79712,
|
| 254 |
+
"step": 130
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.3684210526315788,
|
| 258 |
+
"grad_norm": 4.8105597496032715,
|
| 259 |
+
"learning_rate": 4.727168678465988e-05,
|
| 260 |
+
"loss": 0.1799,
|
| 261 |
+
"num_input_tokens_seen": 82944,
|
| 262 |
+
"step": 135
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 2.456140350877193,
|
| 266 |
+
"grad_norm": 1.7109187841415405,
|
| 267 |
+
"learning_rate": 4.6913564854400595e-05,
|
| 268 |
+
"loss": 0.0346,
|
| 269 |
+
"num_input_tokens_seen": 85568,
|
| 270 |
+
"step": 140
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.543859649122807,
|
| 274 |
+
"grad_norm": 10.28818130493164,
|
| 275 |
+
"learning_rate": 4.6534898965828405e-05,
|
| 276 |
+
"loss": 0.2718,
|
| 277 |
+
"num_input_tokens_seen": 89088,
|
| 278 |
+
"step": 145
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.543859649122807,
|
| 282 |
+
"eval_loss": 0.04383612796664238,
|
| 283 |
+
"eval_runtime": 0.9808,
|
| 284 |
+
"eval_samples_per_second": 25.49,
|
| 285 |
+
"eval_steps_per_second": 7.137,
|
| 286 |
+
"num_input_tokens_seen": 89088,
|
| 287 |
+
"step": 145
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 2.6315789473684212,
|
| 291 |
+
"grad_norm": 3.1198360919952393,
|
| 292 |
+
"learning_rate": 4.613604411806285e-05,
|
| 293 |
+
"loss": 0.2177,
|
| 294 |
+
"num_input_tokens_seen": 92128,
|
| 295 |
+
"step": 150
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.719298245614035,
|
| 299 |
+
"grad_norm": 4.574709415435791,
|
| 300 |
+
"learning_rate": 4.5717374237364665e-05,
|
| 301 |
+
"loss": 0.0509,
|
| 302 |
+
"num_input_tokens_seen": 95520,
|
| 303 |
+
"step": 155
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.807017543859649,
|
| 307 |
+
"grad_norm": 11.19549560546875,
|
| 308 |
+
"learning_rate": 4.5279281826580056e-05,
|
| 309 |
+
"loss": 0.2005,
|
| 310 |
+
"num_input_tokens_seen": 98848,
|
| 311 |
+
"step": 160
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.8947368421052633,
|
| 315 |
+
"grad_norm": 12.849621772766113,
|
| 316 |
+
"learning_rate": 4.482217759716946e-05,
|
| 317 |
+
"loss": 0.2062,
|
| 318 |
+
"num_input_tokens_seen": 101856,
|
| 319 |
+
"step": 165
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.982456140350877,
|
| 323 |
+
"grad_norm": 0.13913767039775848,
|
| 324 |
+
"learning_rate": 4.434649008416565e-05,
|
| 325 |
+
"loss": 0.1133,
|
| 326 |
+
"num_input_tokens_seen": 105632,
|
| 327 |
+
"step": 170
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.0526315789473686,
|
| 331 |
+
"eval_loss": 0.06627459824085236,
|
| 332 |
+
"eval_runtime": 0.9746,
|
| 333 |
+
"eval_samples_per_second": 25.651,
|
| 334 |
+
"eval_steps_per_second": 7.182,
|
| 335 |
+
"num_input_tokens_seen": 107504,
|
| 336 |
+
"step": 174
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.0701754385964914,
|
| 340 |
+
"grad_norm": 2.0158233642578125,
|
| 341 |
+
"learning_rate": 4.385266524442241e-05,
|
| 342 |
+
"loss": 0.0206,
|
| 343 |
+
"num_input_tokens_seen": 108208,
|
| 344 |
+
"step": 175
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.1578947368421053,
|
| 348 |
+
"grad_norm": 8.511082649230957,
|
| 349 |
+
"learning_rate": 4.334116603853007e-05,
|
| 350 |
+
"loss": 0.1147,
|
| 351 |
+
"num_input_tokens_seen": 111280,
|
| 352 |
+
"step": 180
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.245614035087719,
|
| 356 |
+
"grad_norm": 0.18950973451137543,
|
| 357 |
+
"learning_rate": 4.2812471996790206e-05,
|
| 358 |
+
"loss": 0.0554,
|
| 359 |
+
"num_input_tokens_seen": 114512,
|
| 360 |
+
"step": 185
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 3.3333333333333335,
|
| 364 |
+
"grad_norm": 2.283494472503662,
|
| 365 |
+
"learning_rate": 4.226707876965611e-05,
|
| 366 |
+
"loss": 0.0259,
|
| 367 |
+
"num_input_tokens_seen": 117616,
|
| 368 |
+
"step": 190
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.4210526315789473,
|
| 372 |
+
"grad_norm": 0.019897565245628357,
|
| 373 |
+
"learning_rate": 4.1705497663060767e-05,
|
| 374 |
+
"loss": 0.121,
|
| 375 |
+
"num_input_tokens_seen": 121456,
|
| 376 |
+
"step": 195
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 3.5087719298245617,
|
| 380 |
+
"grad_norm": 0.41271084547042847,
|
| 381 |
+
"learning_rate": 4.1128255159067665e-05,
|
| 382 |
+
"loss": 0.0569,
|
| 383 |
+
"num_input_tokens_seen": 124816,
|
| 384 |
+
"step": 200
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 3.56140350877193,
|
| 388 |
+
"eval_loss": 0.008254488930106163,
|
| 389 |
+
"eval_runtime": 0.9742,
|
| 390 |
+
"eval_samples_per_second": 25.662,
|
| 391 |
+
"eval_steps_per_second": 7.185,
|
| 392 |
+
"num_input_tokens_seen": 126384,
|
| 393 |
+
"step": 203
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.5964912280701755,
|
| 397 |
+
"grad_norm": 0.01498648151755333,
|
| 398 |
+
"learning_rate": 4.053589242229412e-05,
|
| 399 |
+
"loss": 0.0283,
|
| 400 |
+
"num_input_tokens_seen": 128048,
|
| 401 |
+
"step": 205
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 3.6842105263157894,
|
| 405 |
+
"grad_norm": 0.07060320675373077,
|
| 406 |
+
"learning_rate": 3.9928964792569655e-05,
|
| 407 |
+
"loss": 0.5359,
|
| 408 |
+
"num_input_tokens_seen": 131184,
|
| 409 |
+
"step": 210
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 3.7719298245614032,
|
| 413 |
+
"grad_norm": 0.04119057208299637,
|
| 414 |
+
"learning_rate": 3.930804126430513e-05,
|
| 415 |
+
"loss": 0.0541,
|
| 416 |
+
"num_input_tokens_seen": 134480,
|
| 417 |
+
"step": 215
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 3.8596491228070176,
|
| 421 |
+
"grad_norm": 0.5880560874938965,
|
| 422 |
+
"learning_rate": 3.867370395306068e-05,
|
| 423 |
+
"loss": 0.0073,
|
| 424 |
+
"num_input_tokens_seen": 137616,
|
| 425 |
+
"step": 220
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 3.9473684210526314,
|
| 429 |
+
"grad_norm": 15.249713897705078,
|
| 430 |
+
"learning_rate": 3.8026547549812665e-05,
|
| 431 |
+
"loss": 0.1801,
|
| 432 |
+
"num_input_tokens_seen": 140496,
|
| 433 |
+
"step": 225
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 4.035087719298246,
|
| 437 |
+
"grad_norm": 4.175070762634277,
|
| 438 |
+
"learning_rate": 3.736717876343106e-05,
|
| 439 |
+
"loss": 0.107,
|
| 440 |
+
"num_input_tokens_seen": 142832,
|
| 441 |
+
"step": 230
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.0701754385964914,
|
| 445 |
+
"eval_loss": 0.06282971054315567,
|
| 446 |
+
"eval_runtime": 0.9755,
|
| 447 |
+
"eval_samples_per_second": 25.629,
|
| 448 |
+
"eval_steps_per_second": 7.176,
|
| 449 |
+
"num_input_tokens_seen": 143952,
|
| 450 |
+
"step": 232
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 4.12280701754386,
|
| 454 |
+
"grad_norm": 16.625675201416016,
|
| 455 |
+
"learning_rate": 3.66962157518902e-05,
|
| 456 |
+
"loss": 0.0352,
|
| 457 |
+
"num_input_tokens_seen": 145936,
|
| 458 |
+
"step": 235
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 4.2105263157894735,
|
| 462 |
+
"grad_norm": 0.653355598449707,
|
| 463 |
+
"learning_rate": 3.601428754274584e-05,
|
| 464 |
+
"loss": 0.008,
|
| 465 |
+
"num_input_tokens_seen": 148496,
|
| 466 |
+
"step": 240
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 4.298245614035087,
|
| 470 |
+
"grad_norm": 0.017996955662965775,
|
| 471 |
+
"learning_rate": 3.532203344342212e-05,
|
| 472 |
+
"loss": 0.0121,
|
| 473 |
+
"num_input_tokens_seen": 151920,
|
| 474 |
+
"step": 245
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 4.385964912280702,
|
| 478 |
+
"grad_norm": 0.013032309710979462,
|
| 479 |
+
"learning_rate": 3.4620102441861143e-05,
|
| 480 |
+
"loss": 0.0006,
|
| 481 |
+
"num_input_tokens_seen": 155760,
|
| 482 |
+
"step": 250
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.473684210526316,
|
| 486 |
+
"grad_norm": 6.339408874511719,
|
| 487 |
+
"learning_rate": 3.390915259809696e-05,
|
| 488 |
+
"loss": 0.0833,
|
| 489 |
+
"num_input_tokens_seen": 158608,
|
| 490 |
+
"step": 255
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.56140350877193,
|
| 494 |
+
"grad_norm": 0.04958834499120712,
|
| 495 |
+
"learning_rate": 3.318985042732461e-05,
|
| 496 |
+
"loss": 0.0284,
|
| 497 |
+
"num_input_tokens_seen": 161232,
|
| 498 |
+
"step": 260
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.578947368421053,
|
| 502 |
+
"eval_loss": 0.13766826689243317,
|
| 503 |
+
"eval_runtime": 0.9766,
|
| 504 |
+
"eval_samples_per_second": 25.6,
|
| 505 |
+
"eval_steps_per_second": 7.168,
|
| 506 |
+
"num_input_tokens_seen": 161840,
|
| 507 |
+
"step": 261
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 4.649122807017544,
|
| 511 |
+
"grad_norm": 0.39925575256347656,
|
| 512 |
+
"learning_rate": 3.246287027504237e-05,
|
| 513 |
+
"loss": 0.0059,
|
| 514 |
+
"num_input_tokens_seen": 164336,
|
| 515 |
+
"step": 265
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.7368421052631575,
|
| 519 |
+
"grad_norm": 11.792527198791504,
|
| 520 |
+
"learning_rate": 3.172889368485311e-05,
|
| 521 |
+
"loss": 0.1597,
|
| 522 |
+
"num_input_tokens_seen": 167568,
|
| 523 |
+
"step": 270
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 4.824561403508772,
|
| 527 |
+
"grad_norm": 0.0264922883361578,
|
| 528 |
+
"learning_rate": 3.0988608759517475e-05,
|
| 529 |
+
"loss": 0.0515,
|
| 530 |
+
"num_input_tokens_seen": 170672,
|
| 531 |
+
"step": 275
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.912280701754386,
|
| 535 |
+
"grad_norm": 19.58041763305664,
|
| 536 |
+
"learning_rate": 3.0242709515857758e-05,
|
| 537 |
+
"loss": 0.1949,
|
| 538 |
+
"num_input_tokens_seen": 173648,
|
| 539 |
+
"step": 280
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 5.0,
|
| 543 |
+
"grad_norm": 0.016560431569814682,
|
| 544 |
+
"learning_rate": 2.949189523411747e-05,
|
| 545 |
+
"loss": 0.0996,
|
| 546 |
+
"num_input_tokens_seen": 176200,
|
| 547 |
+
"step": 285
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 5.087719298245614,
|
| 551 |
+
"grad_norm": 0.0974491760134697,
|
| 552 |
+
"learning_rate": 2.8736869802386364e-05,
|
| 553 |
+
"loss": 0.0148,
|
| 554 |
+
"num_input_tokens_seen": 179816,
|
| 555 |
+
"step": 290
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.087719298245614,
|
| 559 |
+
"eval_loss": 0.07785650342702866,
|
| 560 |
+
"eval_runtime": 0.9762,
|
| 561 |
+
"eval_samples_per_second": 25.61,
|
| 562 |
+
"eval_steps_per_second": 7.171,
|
| 563 |
+
"num_input_tokens_seen": 179816,
|
| 564 |
+
"step": 290
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 5.175438596491228,
|
| 568 |
+
"grad_norm": 4.775076866149902,
|
| 569 |
+
"learning_rate": 2.797834105670559e-05,
|
| 570 |
+
"loss": 0.0184,
|
| 571 |
+
"num_input_tokens_seen": 182920,
|
| 572 |
+
"step": 295
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 5.2631578947368425,
|
| 576 |
+
"grad_norm": 3.1460559368133545,
|
| 577 |
+
"learning_rate": 2.7217020117471793e-05,
|
| 578 |
+
"loss": 0.0464,
|
| 579 |
+
"num_input_tokens_seen": 185928,
|
| 580 |
+
"step": 300
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 5.350877192982456,
|
| 584 |
+
"grad_norm": 6.037375450134277,
|
| 585 |
+
"learning_rate": 2.6453620722761896e-05,
|
| 586 |
+
"loss": 0.0261,
|
| 587 |
+
"num_input_tokens_seen": 188392,
|
| 588 |
+
"step": 305
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 5.43859649122807,
|
| 592 |
+
"grad_norm": 0.015429499559104443,
|
| 593 |
+
"learning_rate": 2.5688858559204053e-05,
|
| 594 |
+
"loss": 0.073,
|
| 595 |
+
"num_input_tokens_seen": 191624,
|
| 596 |
+
"step": 310
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 5.526315789473684,
|
| 600 |
+
"grad_norm": 0.04696372151374817,
|
| 601 |
+
"learning_rate": 2.492345059102164e-05,
|
| 602 |
+
"loss": 0.0724,
|
| 603 |
+
"num_input_tokens_seen": 194376,
|
| 604 |
+
"step": 315
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 5.5964912280701755,
|
| 608 |
+
"eval_loss": 0.14411476254463196,
|
| 609 |
+
"eval_runtime": 0.9774,
|
| 610 |
+
"eval_samples_per_second": 25.578,
|
| 611 |
+
"eval_steps_per_second": 7.162,
|
| 612 |
+
"num_input_tokens_seen": 197416,
|
| 613 |
+
"step": 319
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 5.614035087719298,
|
| 617 |
+
"grad_norm": 1.3863834142684937,
|
| 618 |
+
"learning_rate": 2.4158114387879616e-05,
|
| 619 |
+
"loss": 0.0098,
|
| 620 |
+
"num_input_tokens_seen": 198088,
|
| 621 |
+
"step": 320
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 5.701754385964913,
|
| 625 |
+
"grad_norm": 0.08153573423624039,
|
| 626 |
+
"learning_rate": 2.3393567452163252e-05,
|
| 627 |
+
"loss": 0.0694,
|
| 628 |
+
"num_input_tokens_seen": 200680,
|
| 629 |
+
"step": 325
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 5.7894736842105265,
|
| 633 |
+
"grad_norm": 0.051107168197631836,
|
| 634 |
+
"learning_rate": 2.2630526546319914e-05,
|
| 635 |
+
"loss": 0.0005,
|
| 636 |
+
"num_input_tokens_seen": 203784,
|
| 637 |
+
"step": 330
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 5.87719298245614,
|
| 641 |
+
"grad_norm": 0.04321357235312462,
|
| 642 |
+
"learning_rate": 2.186970702089457e-05,
|
| 643 |
+
"loss": 0.0005,
|
| 644 |
+
"num_input_tokens_seen": 207112,
|
| 645 |
+
"step": 335
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 5.964912280701754,
|
| 649 |
+
"grad_norm": 0.01020878180861473,
|
| 650 |
+
"learning_rate": 2.111182214388893e-05,
|
| 651 |
+
"loss": 0.0006,
|
| 652 |
+
"num_input_tokens_seen": 209800,
|
| 653 |
+
"step": 340
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 6.052631578947368,
|
| 657 |
+
"grad_norm": 0.015666935592889786,
|
| 658 |
+
"learning_rate": 2.0357582432072957e-05,
|
| 659 |
+
"loss": 0.0177,
|
| 660 |
+
"num_input_tokens_seen": 212896,
|
| 661 |
+
"step": 345
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 6.105263157894737,
|
| 665 |
+
"eval_loss": 0.11291729658842087,
|
| 666 |
+
"eval_runtime": 0.9886,
|
| 667 |
+
"eval_samples_per_second": 25.288,
|
| 668 |
+
"eval_steps_per_second": 7.081,
|
| 669 |
+
"num_input_tokens_seen": 214432,
|
| 670 |
+
"step": 348
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 6.140350877192983,
|
| 674 |
+
"grad_norm": 0.40531566739082336,
|
| 675 |
+
"learning_rate": 1.9607694984875754e-05,
|
| 676 |
+
"loss": 0.0279,
|
| 677 |
+
"num_input_tokens_seen": 215456,
|
| 678 |
+
"step": 350
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 6.228070175438597,
|
| 682 |
+
"grad_norm": 0.01669851690530777,
|
| 683 |
+
"learning_rate": 1.8862862821480025e-05,
|
| 684 |
+
"loss": 0.017,
|
| 685 |
+
"num_input_tokens_seen": 218944,
|
| 686 |
+
"step": 355
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 6.315789473684211,
|
| 690 |
+
"grad_norm": 0.0227603018283844,
|
| 691 |
+
"learning_rate": 1.8123784221741964e-05,
|
| 692 |
+
"loss": 0.0005,
|
| 693 |
+
"num_input_tokens_seen": 221824,
|
| 694 |
+
"step": 360
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 6.4035087719298245,
|
| 698 |
+
"grad_norm": 4.3280110359191895,
|
| 699 |
+
"learning_rate": 1.73911520715541e-05,
|
| 700 |
+
"loss": 0.0118,
|
| 701 |
+
"num_input_tokens_seen": 225568,
|
| 702 |
+
"step": 365
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 6.491228070175438,
|
| 706 |
+
"grad_norm": 0.0335407480597496,
|
| 707 |
+
"learning_rate": 1.666565321326512e-05,
|
| 708 |
+
"loss": 0.0013,
|
| 709 |
+
"num_input_tokens_seen": 228960,
|
| 710 |
+
"step": 370
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 6.578947368421053,
|
| 714 |
+
"grad_norm": 0.008555182255804539,
|
| 715 |
+
"learning_rate": 1.5947967801765345e-05,
|
| 716 |
+
"loss": 0.0043,
|
| 717 |
+
"num_input_tokens_seen": 231936,
|
| 718 |
+
"step": 375
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 6.614035087719298,
|
| 722 |
+
"eval_loss": 0.15904298424720764,
|
| 723 |
+
"eval_runtime": 0.9821,
|
| 724 |
+
"eval_samples_per_second": 25.455,
|
| 725 |
+
"eval_steps_per_second": 7.127,
|
| 726 |
+
"num_input_tokens_seen": 233280,
|
| 727 |
+
"step": 377
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 6.666666666666667,
|
| 731 |
+
"grad_norm": 0.006874398794025183,
|
| 732 |
+
"learning_rate": 1.5238768666841907e-05,
|
| 733 |
+
"loss": 0.0105,
|
| 734 |
+
"num_input_tokens_seen": 234944,
|
| 735 |
+
"step": 380
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 6.754385964912281,
|
| 739 |
+
"grad_norm": 0.015940073877573013,
|
| 740 |
+
"learning_rate": 1.4538720682400969e-05,
|
| 741 |
+
"loss": 0.0003,
|
| 742 |
+
"num_input_tokens_seen": 238080,
|
| 743 |
+
"step": 385
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 6.842105263157895,
|
| 747 |
+
"grad_norm": 0.030423445627093315,
|
| 748 |
+
"learning_rate": 1.3848480143148839e-05,
|
| 749 |
+
"loss": 0.0018,
|
| 750 |
+
"num_input_tokens_seen": 241120,
|
| 751 |
+
"step": 390
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 6.9298245614035086,
|
| 755 |
+
"grad_norm": 0.00575337465852499,
|
| 756 |
+
"learning_rate": 1.3168694149315796e-05,
|
| 757 |
+
"loss": 0.0006,
|
| 758 |
+
"num_input_tokens_seen": 244800,
|
| 759 |
+
"step": 395
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 7.017543859649122,
|
| 763 |
+
"grad_norm": 0.006840342190116644,
|
| 764 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 765 |
+
"loss": 0.0005,
|
| 766 |
+
"num_input_tokens_seen": 247408,
|
| 767 |
+
"step": 400
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 7.105263157894737,
|
| 771 |
+
"grad_norm": 0.022249391302466393,
|
| 772 |
+
"learning_rate": 1.1843024595699805e-05,
|
| 773 |
+
"loss": 0.0003,
|
| 774 |
+
"num_input_tokens_seen": 250512,
|
| 775 |
+
"step": 405
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 7.12280701754386,
|
| 779 |
+
"eval_loss": 0.16361309587955475,
|
| 780 |
+
"eval_runtime": 0.9703,
|
| 781 |
+
"eval_samples_per_second": 25.764,
|
| 782 |
+
"eval_steps_per_second": 7.214,
|
| 783 |
+
"num_input_tokens_seen": 251120,
|
| 784 |
+
"step": 406
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 7.192982456140351,
|
| 788 |
+
"grad_norm": 0.010337313637137413,
|
| 789 |
+
"learning_rate": 1.1198383850594758e-05,
|
| 790 |
+
"loss": 0.0005,
|
| 791 |
+
"num_input_tokens_seen": 254352,
|
| 792 |
+
"step": 410
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 7.280701754385965,
|
| 796 |
+
"grad_norm": 7.584884166717529,
|
| 797 |
+
"learning_rate": 1.0566682115126344e-05,
|
| 798 |
+
"loss": 0.0078,
|
| 799 |
+
"num_input_tokens_seen": 257264,
|
| 800 |
+
"step": 415
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 7.368421052631579,
|
| 804 |
+
"grad_norm": 0.010941845364868641,
|
| 805 |
+
"learning_rate": 9.948511609419675e-06,
|
| 806 |
+
"loss": 0.0003,
|
| 807 |
+
"num_input_tokens_seen": 260048,
|
| 808 |
+
"step": 420
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 7.456140350877193,
|
| 812 |
+
"grad_norm": 0.010121169500052929,
|
| 813 |
+
"learning_rate": 9.344451868077353e-06,
|
| 814 |
+
"loss": 0.0002,
|
| 815 |
+
"num_input_tokens_seen": 263568,
|
| 816 |
+
"step": 425
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 7.543859649122807,
|
| 820 |
+
"grad_norm": 0.007580413483083248,
|
| 821 |
+
"learning_rate": 8.755069196866014e-06,
|
| 822 |
+
"loss": 0.0002,
|
| 823 |
+
"num_input_tokens_seen": 266640,
|
| 824 |
+
"step": 430
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 7.631578947368421,
|
| 828 |
+
"grad_norm": 0.003944739233702421,
|
| 829 |
+
"learning_rate": 8.180916141804906e-06,
|
| 830 |
+
"loss": 0.0001,
|
| 831 |
+
"num_input_tokens_seen": 270128,
|
| 832 |
+
"step": 435
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 7.631578947368421,
|
| 836 |
+
"eval_loss": 0.15004655718803406,
|
| 837 |
+
"eval_runtime": 0.9753,
|
| 838 |
+
"eval_samples_per_second": 25.634,
|
| 839 |
+
"eval_steps_per_second": 7.177,
|
| 840 |
+
"num_input_tokens_seen": 270128,
|
| 841 |
+
"step": 435
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 7.719298245614035,
|
| 845 |
+
"grad_norm": 0.011167265474796295,
|
| 846 |
+
"learning_rate": 7.622530971154199e-06,
|
| 847 |
+
"loss": 0.0003,
|
| 848 |
+
"num_input_tokens_seen": 273136,
|
| 849 |
+
"step": 440
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 7.807017543859649,
|
| 853 |
+
"grad_norm": 0.08406668901443481,
|
| 854 |
+
"learning_rate": 7.080437170788723e-06,
|
| 855 |
+
"loss": 0.0039,
|
| 856 |
+
"num_input_tokens_seen": 276464,
|
| 857 |
+
"step": 445
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 7.894736842105263,
|
| 861 |
+
"grad_norm": 0.17748688161373138,
|
| 862 |
+
"learning_rate": 6.555142953430158e-06,
|
| 863 |
+
"loss": 0.0004,
|
| 864 |
+
"num_input_tokens_seen": 278960,
|
| 865 |
+
"step": 450
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 7.982456140350877,
|
| 869 |
+
"grad_norm": 0.011380401439964771,
|
| 870 |
+
"learning_rate": 6.0471407821978135e-06,
|
| 871 |
+
"loss": 0.0003,
|
| 872 |
+
"num_input_tokens_seen": 282512,
|
| 873 |
+
"step": 455
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 8.070175438596491,
|
| 877 |
+
"grad_norm": 0.02105165459215641,
|
| 878 |
+
"learning_rate": 5.556906908924655e-06,
|
| 879 |
+
"loss": 0.0004,
|
| 880 |
+
"num_input_tokens_seen": 285496,
|
| 881 |
+
"step": 460
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 8.140350877192983,
|
| 885 |
+
"eval_loss": 0.16394011676311493,
|
| 886 |
+
"eval_runtime": 0.9867,
|
| 887 |
+
"eval_samples_per_second": 25.337,
|
| 888 |
+
"eval_steps_per_second": 7.094,
|
| 889 |
+
"num_input_tokens_seen": 288216,
|
| 890 |
+
"step": 464
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 8.157894736842104,
|
| 894 |
+
"grad_norm": 0.005344858393073082,
|
| 895 |
+
"learning_rate": 5.084900927671393e-06,
|
| 896 |
+
"loss": 0.0002,
|
| 897 |
+
"num_input_tokens_seen": 289080,
|
| 898 |
+
"step": 465
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 8.24561403508772,
|
| 902 |
+
"grad_norm": 0.006819204892963171,
|
| 903 |
+
"learning_rate": 4.631565343857239e-06,
|
| 904 |
+
"loss": 0.0003,
|
| 905 |
+
"num_input_tokens_seen": 292120,
|
| 906 |
+
"step": 470
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 8.333333333333334,
|
| 910 |
+
"grad_norm": 0.004894046578556299,
|
| 911 |
+
"learning_rate": 4.19732515941125e-06,
|
| 912 |
+
"loss": 0.0004,
|
| 913 |
+
"num_input_tokens_seen": 295384,
|
| 914 |
+
"step": 475
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 8.421052631578947,
|
| 918 |
+
"grad_norm": 0.004802997689694166,
|
| 919 |
+
"learning_rate": 3.7825874743331907e-06,
|
| 920 |
+
"loss": 0.0002,
|
| 921 |
+
"num_input_tokens_seen": 298712,
|
| 922 |
+
"step": 480
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 8.508771929824562,
|
| 926 |
+
"grad_norm": 0.023499688133597374,
|
| 927 |
+
"learning_rate": 3.3877411050374424e-06,
|
| 928 |
+
"loss": 0.0002,
|
| 929 |
+
"num_input_tokens_seen": 301688,
|
| 930 |
+
"step": 485
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 8.596491228070175,
|
| 934 |
+
"grad_norm": 0.005042481701821089,
|
| 935 |
+
"learning_rate": 3.013156219837776e-06,
|
| 936 |
+
"loss": 0.0002,
|
| 937 |
+
"num_input_tokens_seen": 304504,
|
| 938 |
+
"step": 490
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 8.649122807017545,
|
| 942 |
+
"eval_loss": 0.16186748445034027,
|
| 943 |
+
"eval_runtime": 0.9799,
|
| 944 |
+
"eval_samples_per_second": 25.514,
|
| 945 |
+
"eval_steps_per_second": 7.144,
|
| 946 |
+
"num_input_tokens_seen": 306648,
|
| 947 |
+
"step": 493
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 8.68421052631579,
|
| 951 |
+
"grad_norm": 0.004729500040411949,
|
| 952 |
+
"learning_rate": 2.659183991914696e-06,
|
| 953 |
+
"loss": 0.0001,
|
| 954 |
+
"num_input_tokens_seen": 308056,
|
| 955 |
+
"step": 495
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 8.771929824561404,
|
| 959 |
+
"grad_norm": 0.007018934935331345,
|
| 960 |
+
"learning_rate": 2.326156270090735e-06,
|
| 961 |
+
"loss": 0.0002,
|
| 962 |
+
"num_input_tokens_seen": 310584,
|
| 963 |
+
"step": 500
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 8.859649122807017,
|
| 967 |
+
"grad_norm": 0.008993715047836304,
|
| 968 |
+
"learning_rate": 2.0143852677223075e-06,
|
| 969 |
+
"loss": 0.0002,
|
| 970 |
+
"num_input_tokens_seen": 313528,
|
| 971 |
+
"step": 505
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 8.947368421052632,
|
| 975 |
+
"grad_norm": 0.007980038411915302,
|
| 976 |
+
"learning_rate": 1.7241632699998123e-06,
|
| 977 |
+
"loss": 0.0006,
|
| 978 |
+
"num_input_tokens_seen": 316472,
|
| 979 |
+
"step": 510
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 9.035087719298245,
|
| 983 |
+
"grad_norm": 0.05837644264101982,
|
| 984 |
+
"learning_rate": 1.4557623599303903e-06,
|
| 985 |
+
"loss": 0.0002,
|
| 986 |
+
"num_input_tokens_seen": 318912,
|
| 987 |
+
"step": 515
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 9.12280701754386,
|
| 991 |
+
"grad_norm": 0.006253806408494711,
|
| 992 |
+
"learning_rate": 1.2094341632602064e-06,
|
| 993 |
+
"loss": 0.0002,
|
| 994 |
+
"num_input_tokens_seen": 321888,
|
| 995 |
+
"step": 520
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 9.157894736842104,
|
| 999 |
+
"eval_loss": 0.1608547419309616,
|
| 1000 |
+
"eval_runtime": 0.9706,
|
| 1001 |
+
"eval_samples_per_second": 25.757,
|
| 1002 |
+
"eval_steps_per_second": 7.212,
|
| 1003 |
+
"num_input_tokens_seen": 323296,
|
| 1004 |
+
"step": 522
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 9.210526315789474,
|
| 1008 |
+
"grad_norm": 0.006381010636687279,
|
| 1009 |
+
"learning_rate": 9.85409612575411e-07,
|
| 1010 |
+
"loss": 0.0002,
|
| 1011 |
+
"num_input_tokens_seen": 325248,
|
| 1012 |
+
"step": 525
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 9.298245614035087,
|
| 1016 |
+
"grad_norm": 0.005374109372496605,
|
| 1017 |
+
"learning_rate": 7.838987308029427e-07,
|
| 1018 |
+
"loss": 0.0002,
|
| 1019 |
+
"num_input_tokens_seen": 328224,
|
| 1020 |
+
"step": 530
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 9.385964912280702,
|
| 1024 |
+
"grad_norm": 0.005015636328607798,
|
| 1025 |
+
"learning_rate": 6.050904343141095e-07,
|
| 1026 |
+
"loss": 0.0001,
|
| 1027 |
+
"num_input_tokens_seen": 330880,
|
| 1028 |
+
"step": 535
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 9.473684210526315,
|
| 1032 |
+
"grad_norm": 0.0028100141789764166,
|
| 1033 |
+
"learning_rate": 4.491523558155714e-07,
|
| 1034 |
+
"loss": 0.0006,
|
| 1035 |
+
"num_input_tokens_seen": 334176,
|
| 1036 |
+
"step": 540
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 9.56140350877193,
|
| 1040 |
+
"grad_norm": 0.005245237145572901,
|
| 1041 |
+
"learning_rate": 3.162306871937387e-07,
|
| 1042 |
+
"loss": 0.0002,
|
| 1043 |
+
"num_input_tokens_seen": 337280,
|
| 1044 |
+
"step": 545
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 9.649122807017545,
|
| 1048 |
+
"grad_norm": 0.06211718171834946,
|
| 1049 |
+
"learning_rate": 2.064500424599436e-07,
|
| 1050 |
+
"loss": 0.0002,
|
| 1051 |
+
"num_input_tokens_seen": 340480,
|
| 1052 |
+
"step": 550
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 9.666666666666666,
|
| 1056 |
+
"eval_loss": 0.16464732587337494,
|
| 1057 |
+
"eval_runtime": 0.9709,
|
| 1058 |
+
"eval_samples_per_second": 25.75,
|
| 1059 |
+
"eval_steps_per_second": 7.21,
|
| 1060 |
+
"num_input_tokens_seen": 340960,
|
| 1061 |
+
"step": 551
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 9.736842105263158,
|
| 1065 |
+
"grad_norm": 0.009912996552884579,
|
| 1066 |
+
"learning_rate": 1.1991334092484318e-07,
|
| 1067 |
+
"loss": 0.0002,
|
| 1068 |
+
"num_input_tokens_seen": 343328,
|
| 1069 |
+
"step": 555
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 9.824561403508772,
|
| 1073 |
+
"grad_norm": 0.030118485912680626,
|
| 1074 |
+
"learning_rate": 5.6701710711626334e-08,
|
| 1075 |
+
"loss": 0.0003,
|
| 1076 |
+
"num_input_tokens_seen": 345984,
|
| 1077 |
+
"step": 560
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 9.912280701754385,
|
| 1081 |
+
"grad_norm": 0.0043678064830601215,
|
| 1082 |
+
"learning_rate": 1.6874412698408836e-08,
|
| 1083 |
+
"loss": 0.0002,
|
| 1084 |
+
"num_input_tokens_seen": 349280,
|
| 1085 |
+
"step": 565
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 10.0,
|
| 1089 |
+
"grad_norm": 0.2342500537633896,
|
| 1090 |
+
"learning_rate": 4.687849611939576e-10,
|
| 1091 |
+
"loss": 0.0004,
|
| 1092 |
+
"num_input_tokens_seen": 352296,
|
| 1093 |
+
"step": 570
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 10.0,
|
| 1097 |
+
"num_input_tokens_seen": 352296,
|
| 1098 |
+
"step": 570,
|
| 1099 |
+
"total_flos": 1.5863729653481472e+16,
|
| 1100 |
+
"train_loss": 0.24723993078785084,
|
| 1101 |
+
"train_runtime": 184.7232,
|
| 1102 |
+
"train_samples_per_second": 12.18,
|
| 1103 |
+
"train_steps_per_second": 3.086
|
| 1104 |
+
}
|
| 1105 |
+
],
|
| 1106 |
+
"logging_steps": 5,
|
| 1107 |
+
"max_steps": 570,
|
| 1108 |
+
"num_input_tokens_seen": 352296,
|
| 1109 |
+
"num_train_epochs": 10,
|
| 1110 |
+
"save_steps": 29,
|
| 1111 |
+
"stateful_callbacks": {
|
| 1112 |
+
"TrainerControl": {
|
| 1113 |
+
"args": {
|
| 1114 |
+
"should_epoch_stop": false,
|
| 1115 |
+
"should_evaluate": false,
|
| 1116 |
+
"should_log": false,
|
| 1117 |
+
"should_save": true,
|
| 1118 |
+
"should_training_stop": true
|
| 1119 |
+
},
|
| 1120 |
+
"attributes": {}
|
| 1121 |
+
}
|
| 1122 |
+
},
|
| 1123 |
+
"total_flos": 1.5863729653481472e+16,
|
| 1124 |
+
"train_batch_size": 4,
|
| 1125 |
+
"trial_name": null,
|
| 1126 |
+
"trial_params": null
|
| 1127 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|