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 +2023 -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_1757081466
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# train_cb_1757081466
|
| 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: 0.
|
| 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 |
+
- prefix-tuning
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: train_cb_1757081466
|
|
|
|
| 16 |
|
| 17 |
# train_cb_1757081466
|
| 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: 0.3225
|
| 22 |
+
- Num Input Tokens Seen: 310888
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.32253095507621765,
|
| 4 |
+
"eval_runtime": 0.7285,
|
| 5 |
+
"eval_samples_per_second": 34.319,
|
| 6 |
+
"eval_steps_per_second": 17.846,
|
| 7 |
+
"num_input_tokens_seen": 310888,
|
| 8 |
+
"total_flos": 1.3999146128572416e+16,
|
| 9 |
+
"train_loss": 0.27997444444191766,
|
| 10 |
+
"train_runtime": 152.8701,
|
| 11 |
+
"train_samples_per_second": 14.718,
|
| 12 |
+
"train_steps_per_second": 7.392
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.32253095507621765,
|
| 4 |
+
"eval_runtime": 0.7285,
|
| 5 |
+
"eval_samples_per_second": 34.319,
|
| 6 |
+
"eval_steps_per_second": 17.846,
|
| 7 |
+
"num_input_tokens_seen": 310888
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"num_input_tokens_seen": 310888,
|
| 4 |
+
"total_flos": 1.3999146128572416e+16,
|
| 5 |
+
"train_loss": 0.27997444444191766,
|
| 6 |
+
"train_runtime": 152.8701,
|
| 7 |
+
"train_samples_per_second": 14.718,
|
| 8 |
+
"train_steps_per_second": 7.392
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2023 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 570,
|
| 3 |
+
"best_metric": 0.1859114021062851,
|
| 4 |
+
"best_model_checkpoint": "saves_stability/prefix-tuning/llama-3-8b-instruct/train_cb_1757081466/checkpoint-570",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 57,
|
| 7 |
+
"global_step": 1130,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.04424778761061947,
|
| 14 |
+
"grad_norm": 158.60247802734375,
|
| 15 |
+
"learning_rate": 1.7699115044247788e-06,
|
| 16 |
+
"loss": 7.2714,
|
| 17 |
+
"num_input_tokens_seen": 1248,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.08849557522123894,
|
| 22 |
+
"grad_norm": 164.1085662841797,
|
| 23 |
+
"learning_rate": 3.982300884955752e-06,
|
| 24 |
+
"loss": 6.2545,
|
| 25 |
+
"num_input_tokens_seen": 2864,
|
| 26 |
+
"step": 10
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.13274336283185842,
|
| 30 |
+
"grad_norm": 111.85543823242188,
|
| 31 |
+
"learning_rate": 6.194690265486726e-06,
|
| 32 |
+
"loss": 4.3354,
|
| 33 |
+
"num_input_tokens_seen": 4160,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.17699115044247787,
|
| 38 |
+
"grad_norm": 83.06703186035156,
|
| 39 |
+
"learning_rate": 8.407079646017701e-06,
|
| 40 |
+
"loss": 2.0766,
|
| 41 |
+
"num_input_tokens_seen": 5552,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.22123893805309736,
|
| 46 |
+
"grad_norm": 92.03155517578125,
|
| 47 |
+
"learning_rate": 1.0619469026548673e-05,
|
| 48 |
+
"loss": 1.4378,
|
| 49 |
+
"num_input_tokens_seen": 6848,
|
| 50 |
+
"step": 25
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.26548672566371684,
|
| 54 |
+
"grad_norm": 76.08715057373047,
|
| 55 |
+
"learning_rate": 1.2831858407079647e-05,
|
| 56 |
+
"loss": 0.4648,
|
| 57 |
+
"num_input_tokens_seen": 8016,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.30973451327433627,
|
| 62 |
+
"grad_norm": 62.18666076660156,
|
| 63 |
+
"learning_rate": 1.504424778761062e-05,
|
| 64 |
+
"loss": 0.572,
|
| 65 |
+
"num_input_tokens_seen": 9408,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.35398230088495575,
|
| 70 |
+
"grad_norm": 11.945180892944336,
|
| 71 |
+
"learning_rate": 1.7256637168141594e-05,
|
| 72 |
+
"loss": 0.2606,
|
| 73 |
+
"num_input_tokens_seen": 10736,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.39823008849557523,
|
| 78 |
+
"grad_norm": 270.0860900878906,
|
| 79 |
+
"learning_rate": 1.946902654867257e-05,
|
| 80 |
+
"loss": 0.5221,
|
| 81 |
+
"num_input_tokens_seen": 11984,
|
| 82 |
+
"step": 45
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.4424778761061947,
|
| 86 |
+
"grad_norm": 58.8084602355957,
|
| 87 |
+
"learning_rate": 2.1681415929203542e-05,
|
| 88 |
+
"loss": 0.5118,
|
| 89 |
+
"num_input_tokens_seen": 13360,
|
| 90 |
+
"step": 50
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.48672566371681414,
|
| 94 |
+
"grad_norm": 80.50281524658203,
|
| 95 |
+
"learning_rate": 2.3893805309734516e-05,
|
| 96 |
+
"loss": 0.5795,
|
| 97 |
+
"num_input_tokens_seen": 14800,
|
| 98 |
+
"step": 55
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.504424778761062,
|
| 102 |
+
"eval_loss": 0.3516029119491577,
|
| 103 |
+
"eval_runtime": 0.7089,
|
| 104 |
+
"eval_samples_per_second": 35.267,
|
| 105 |
+
"eval_steps_per_second": 18.339,
|
| 106 |
+
"num_input_tokens_seen": 15456,
|
| 107 |
+
"step": 57
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5309734513274337,
|
| 111 |
+
"grad_norm": 67.8954849243164,
|
| 112 |
+
"learning_rate": 2.610619469026549e-05,
|
| 113 |
+
"loss": 0.398,
|
| 114 |
+
"num_input_tokens_seen": 16192,
|
| 115 |
+
"step": 60
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.5752212389380531,
|
| 119 |
+
"grad_norm": 83.74840545654297,
|
| 120 |
+
"learning_rate": 2.831858407079646e-05,
|
| 121 |
+
"loss": 0.3393,
|
| 122 |
+
"num_input_tokens_seen": 18160,
|
| 123 |
+
"step": 65
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.6194690265486725,
|
| 127 |
+
"grad_norm": 4.390460968017578,
|
| 128 |
+
"learning_rate": 3.0530973451327434e-05,
|
| 129 |
+
"loss": 0.2629,
|
| 130 |
+
"num_input_tokens_seen": 19520,
|
| 131 |
+
"step": 70
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.6637168141592921,
|
| 135 |
+
"grad_norm": 55.11955261230469,
|
| 136 |
+
"learning_rate": 3.274336283185841e-05,
|
| 137 |
+
"loss": 0.6772,
|
| 138 |
+
"num_input_tokens_seen": 20816,
|
| 139 |
+
"step": 75
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.7079646017699115,
|
| 143 |
+
"grad_norm": 58.2765998840332,
|
| 144 |
+
"learning_rate": 3.495575221238938e-05,
|
| 145 |
+
"loss": 0.5141,
|
| 146 |
+
"num_input_tokens_seen": 22432,
|
| 147 |
+
"step": 80
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.7522123893805309,
|
| 151 |
+
"grad_norm": 40.18519592285156,
|
| 152 |
+
"learning_rate": 3.716814159292036e-05,
|
| 153 |
+
"loss": 0.968,
|
| 154 |
+
"num_input_tokens_seen": 23568,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.7964601769911505,
|
| 159 |
+
"grad_norm": 73.16150665283203,
|
| 160 |
+
"learning_rate": 3.938053097345133e-05,
|
| 161 |
+
"loss": 0.4512,
|
| 162 |
+
"num_input_tokens_seen": 25280,
|
| 163 |
+
"step": 90
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.8407079646017699,
|
| 167 |
+
"grad_norm": 35.98966598510742,
|
| 168 |
+
"learning_rate": 4.15929203539823e-05,
|
| 169 |
+
"loss": 0.7999,
|
| 170 |
+
"num_input_tokens_seen": 26656,
|
| 171 |
+
"step": 95
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.8849557522123894,
|
| 175 |
+
"grad_norm": 38.017478942871094,
|
| 176 |
+
"learning_rate": 4.380530973451328e-05,
|
| 177 |
+
"loss": 0.9526,
|
| 178 |
+
"num_input_tokens_seen": 28080,
|
| 179 |
+
"step": 100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.9292035398230089,
|
| 183 |
+
"grad_norm": 15.452778816223145,
|
| 184 |
+
"learning_rate": 4.601769911504425e-05,
|
| 185 |
+
"loss": 0.4326,
|
| 186 |
+
"num_input_tokens_seen": 29344,
|
| 187 |
+
"step": 105
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.9734513274336283,
|
| 191 |
+
"grad_norm": 29.611724853515625,
|
| 192 |
+
"learning_rate": 4.823008849557522e-05,
|
| 193 |
+
"loss": 0.2361,
|
| 194 |
+
"num_input_tokens_seen": 30576,
|
| 195 |
+
"step": 110
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.008849557522124,
|
| 199 |
+
"eval_loss": 1.4061414003372192,
|
| 200 |
+
"eval_runtime": 0.7082,
|
| 201 |
+
"eval_samples_per_second": 35.303,
|
| 202 |
+
"eval_steps_per_second": 18.357,
|
| 203 |
+
"num_input_tokens_seen": 31408,
|
| 204 |
+
"step": 114
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.0176991150442478,
|
| 208 |
+
"grad_norm": 30.109329223632812,
|
| 209 |
+
"learning_rate": 4.9999880720033856e-05,
|
| 210 |
+
"loss": 0.4565,
|
| 211 |
+
"num_input_tokens_seen": 31584,
|
| 212 |
+
"step": 115
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.0619469026548674,
|
| 216 |
+
"grad_norm": 15.066938400268555,
|
| 217 |
+
"learning_rate": 4.999570604073014e-05,
|
| 218 |
+
"loss": 0.6162,
|
| 219 |
+
"num_input_tokens_seen": 32688,
|
| 220 |
+
"step": 120
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.1061946902654867,
|
| 224 |
+
"grad_norm": 29.882238388061523,
|
| 225 |
+
"learning_rate": 4.998556850128691e-05,
|
| 226 |
+
"loss": 0.4465,
|
| 227 |
+
"num_input_tokens_seen": 33872,
|
| 228 |
+
"step": 125
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.1504424778761062,
|
| 232 |
+
"grad_norm": 11.128195762634277,
|
| 233 |
+
"learning_rate": 4.996947052006874e-05,
|
| 234 |
+
"loss": 0.3424,
|
| 235 |
+
"num_input_tokens_seen": 35136,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.1946902654867257,
|
| 240 |
+
"grad_norm": 4.366806507110596,
|
| 241 |
+
"learning_rate": 4.9947415937335635e-05,
|
| 242 |
+
"loss": 0.6297,
|
| 243 |
+
"num_input_tokens_seen": 36448,
|
| 244 |
+
"step": 135
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.238938053097345,
|
| 248 |
+
"grad_norm": 4.560500621795654,
|
| 249 |
+
"learning_rate": 4.9919410014326955e-05,
|
| 250 |
+
"loss": 0.2276,
|
| 251 |
+
"num_input_tokens_seen": 37568,
|
| 252 |
+
"step": 140
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.2831858407079646,
|
| 256 |
+
"grad_norm": 15.226495742797852,
|
| 257 |
+
"learning_rate": 4.98854594320063e-05,
|
| 258 |
+
"loss": 0.4086,
|
| 259 |
+
"num_input_tokens_seen": 38800,
|
| 260 |
+
"step": 145
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.3274336283185841,
|
| 264 |
+
"grad_norm": 7.096482753753662,
|
| 265 |
+
"learning_rate": 4.98455722894677e-05,
|
| 266 |
+
"loss": 0.4786,
|
| 267 |
+
"num_input_tokens_seen": 39824,
|
| 268 |
+
"step": 150
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.3716814159292037,
|
| 272 |
+
"grad_norm": 1.0929534435272217,
|
| 273 |
+
"learning_rate": 4.979975810200359e-05,
|
| 274 |
+
"loss": 0.2336,
|
| 275 |
+
"num_input_tokens_seen": 41312,
|
| 276 |
+
"step": 155
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.415929203539823,
|
| 280 |
+
"grad_norm": 1.4610778093338013,
|
| 281 |
+
"learning_rate": 4.974802779883484e-05,
|
| 282 |
+
"loss": 0.4179,
|
| 283 |
+
"num_input_tokens_seen": 42864,
|
| 284 |
+
"step": 160
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.4601769911504425,
|
| 288 |
+
"grad_norm": 1.1674245595932007,
|
| 289 |
+
"learning_rate": 4.969039372050356e-05,
|
| 290 |
+
"loss": 0.1715,
|
| 291 |
+
"num_input_tokens_seen": 44480,
|
| 292 |
+
"step": 165
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.504424778761062,
|
| 296 |
+
"grad_norm": 5.010025978088379,
|
| 297 |
+
"learning_rate": 4.9626869615929175e-05,
|
| 298 |
+
"loss": 0.1954,
|
| 299 |
+
"num_input_tokens_seen": 45904,
|
| 300 |
+
"step": 170
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.5132743362831858,
|
| 304 |
+
"eval_loss": 0.5750893950462341,
|
| 305 |
+
"eval_runtime": 0.7056,
|
| 306 |
+
"eval_samples_per_second": 35.43,
|
| 307 |
+
"eval_steps_per_second": 18.424,
|
| 308 |
+
"num_input_tokens_seen": 46128,
|
| 309 |
+
"step": 171
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.5486725663716814,
|
| 313 |
+
"grad_norm": 0.6192958354949951,
|
| 314 |
+
"learning_rate": 4.9557470639128554e-05,
|
| 315 |
+
"loss": 0.6675,
|
| 316 |
+
"num_input_tokens_seen": 47728,
|
| 317 |
+
"step": 175
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.592920353982301,
|
| 321 |
+
"grad_norm": 1.1015119552612305,
|
| 322 |
+
"learning_rate": 4.9482213345600936e-05,
|
| 323 |
+
"loss": 0.0695,
|
| 324 |
+
"num_input_tokens_seen": 49072,
|
| 325 |
+
"step": 180
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.6371681415929205,
|
| 329 |
+
"grad_norm": 0.21914945542812347,
|
| 330 |
+
"learning_rate": 4.940111568837854e-05,
|
| 331 |
+
"loss": 0.3679,
|
| 332 |
+
"num_input_tokens_seen": 50224,
|
| 333 |
+
"step": 185
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.6814159292035398,
|
| 337 |
+
"grad_norm": 3.072237730026245,
|
| 338 |
+
"learning_rate": 4.931419701374377e-05,
|
| 339 |
+
"loss": 0.5304,
|
| 340 |
+
"num_input_tokens_seen": 51392,
|
| 341 |
+
"step": 190
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.7256637168141593,
|
| 345 |
+
"grad_norm": 1.34825599193573,
|
| 346 |
+
"learning_rate": 4.922147805661402e-05,
|
| 347 |
+
"loss": 0.1569,
|
| 348 |
+
"num_input_tokens_seen": 53040,
|
| 349 |
+
"step": 195
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.7699115044247788,
|
| 353 |
+
"grad_norm": 1.5443557500839233,
|
| 354 |
+
"learning_rate": 4.91229809355953e-05,
|
| 355 |
+
"loss": 0.3948,
|
| 356 |
+
"num_input_tokens_seen": 54080,
|
| 357 |
+
"step": 200
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.8141592920353982,
|
| 361 |
+
"grad_norm": 0.7244002223014832,
|
| 362 |
+
"learning_rate": 4.901872914770569e-05,
|
| 363 |
+
"loss": 0.151,
|
| 364 |
+
"num_input_tokens_seen": 55456,
|
| 365 |
+
"step": 205
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.8584070796460177,
|
| 369 |
+
"grad_norm": 11.362713813781738,
|
| 370 |
+
"learning_rate": 4.8908747562769995e-05,
|
| 371 |
+
"loss": 0.5248,
|
| 372 |
+
"num_input_tokens_seen": 56928,
|
| 373 |
+
"step": 210
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.9026548672566372,
|
| 377 |
+
"grad_norm": 2.929269313812256,
|
| 378 |
+
"learning_rate": 4.8793062417486976e-05,
|
| 379 |
+
"loss": 0.2828,
|
| 380 |
+
"num_input_tokens_seen": 58672,
|
| 381 |
+
"step": 215
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.9469026548672566,
|
| 385 |
+
"grad_norm": 6.6553635597229,
|
| 386 |
+
"learning_rate": 4.867170130917034e-05,
|
| 387 |
+
"loss": 0.3496,
|
| 388 |
+
"num_input_tokens_seen": 60064,
|
| 389 |
+
"step": 220
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.991150442477876,
|
| 393 |
+
"grad_norm": 7.771035671234131,
|
| 394 |
+
"learning_rate": 4.8544693189165324e-05,
|
| 395 |
+
"loss": 0.4297,
|
| 396 |
+
"num_input_tokens_seen": 61760,
|
| 397 |
+
"step": 225
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 2.017699115044248,
|
| 401 |
+
"eval_loss": 0.4387664794921875,
|
| 402 |
+
"eval_runtime": 0.7069,
|
| 403 |
+
"eval_samples_per_second": 35.364,
|
| 404 |
+
"eval_steps_per_second": 18.389,
|
| 405 |
+
"num_input_tokens_seen": 62512,
|
| 406 |
+
"step": 228
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.0353982300884956,
|
| 410 |
+
"grad_norm": 3.6984031200408936,
|
| 411 |
+
"learning_rate": 4.841206835594222e-05,
|
| 412 |
+
"loss": 0.1663,
|
| 413 |
+
"num_input_tokens_seen": 63088,
|
| 414 |
+
"step": 230
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.079646017699115,
|
| 418 |
+
"grad_norm": 12.891488075256348,
|
| 419 |
+
"learning_rate": 4.8273858447868424e-05,
|
| 420 |
+
"loss": 0.3115,
|
| 421 |
+
"num_input_tokens_seen": 64400,
|
| 422 |
+
"step": 235
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.1238938053097347,
|
| 426 |
+
"grad_norm": 0.34429389238357544,
|
| 427 |
+
"learning_rate": 4.813009643566101e-05,
|
| 428 |
+
"loss": 0.0553,
|
| 429 |
+
"num_input_tokens_seen": 65904,
|
| 430 |
+
"step": 240
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.168141592920354,
|
| 434 |
+
"grad_norm": 1.5679643154144287,
|
| 435 |
+
"learning_rate": 4.798081661452135e-05,
|
| 436 |
+
"loss": 0.0229,
|
| 437 |
+
"num_input_tokens_seen": 67184,
|
| 438 |
+
"step": 245
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.2123893805309733,
|
| 442 |
+
"grad_norm": 12.214455604553223,
|
| 443 |
+
"learning_rate": 4.7826054595953815e-05,
|
| 444 |
+
"loss": 0.7841,
|
| 445 |
+
"num_input_tokens_seen": 68528,
|
| 446 |
+
"step": 250
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.256637168141593,
|
| 450 |
+
"grad_norm": 4.900900363922119,
|
| 451 |
+
"learning_rate": 4.766584729927049e-05,
|
| 452 |
+
"loss": 0.2819,
|
| 453 |
+
"num_input_tokens_seen": 70016,
|
| 454 |
+
"step": 255
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 2.3008849557522124,
|
| 458 |
+
"grad_norm": 2.5725162029266357,
|
| 459 |
+
"learning_rate": 4.750023294278381e-05,
|
| 460 |
+
"loss": 0.1761,
|
| 461 |
+
"num_input_tokens_seen": 71280,
|
| 462 |
+
"step": 260
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.3451327433628317,
|
| 466 |
+
"grad_norm": 1.1505086421966553,
|
| 467 |
+
"learning_rate": 4.732925103468944e-05,
|
| 468 |
+
"loss": 0.2276,
|
| 469 |
+
"num_input_tokens_seen": 72960,
|
| 470 |
+
"step": 265
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 2.3893805309734515,
|
| 474 |
+
"grad_norm": 6.388667583465576,
|
| 475 |
+
"learning_rate": 4.715294236364135e-05,
|
| 476 |
+
"loss": 0.5257,
|
| 477 |
+
"num_input_tokens_seen": 74272,
|
| 478 |
+
"step": 270
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 2.433628318584071,
|
| 482 |
+
"grad_norm": 5.725376605987549,
|
| 483 |
+
"learning_rate": 4.6971348989021446e-05,
|
| 484 |
+
"loss": 0.3409,
|
| 485 |
+
"num_input_tokens_seen": 75520,
|
| 486 |
+
"step": 275
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 2.47787610619469,
|
| 490 |
+
"grad_norm": 174.96688842773438,
|
| 491 |
+
"learning_rate": 4.6784514230906146e-05,
|
| 492 |
+
"loss": 0.4146,
|
| 493 |
+
"num_input_tokens_seen": 76672,
|
| 494 |
+
"step": 280
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.52212389380531,
|
| 498 |
+
"grad_norm": 7.057203769683838,
|
| 499 |
+
"learning_rate": 4.659248265973205e-05,
|
| 500 |
+
"loss": 0.2269,
|
| 501 |
+
"num_input_tokens_seen": 78272,
|
| 502 |
+
"step": 285
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 2.52212389380531,
|
| 506 |
+
"eval_loss": 0.22981101274490356,
|
| 507 |
+
"eval_runtime": 0.7102,
|
| 508 |
+
"eval_samples_per_second": 35.203,
|
| 509 |
+
"eval_steps_per_second": 18.306,
|
| 510 |
+
"num_input_tokens_seen": 78272,
|
| 511 |
+
"step": 285
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.566371681415929,
|
| 515 |
+
"grad_norm": 4.738373756408691,
|
| 516 |
+
"learning_rate": 4.639530008566349e-05,
|
| 517 |
+
"loss": 0.3573,
|
| 518 |
+
"num_input_tokens_seen": 79296,
|
| 519 |
+
"step": 290
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.6106194690265485,
|
| 523 |
+
"grad_norm": 0.413061261177063,
|
| 524 |
+
"learning_rate": 4.6193013547664255e-05,
|
| 525 |
+
"loss": 0.2362,
|
| 526 |
+
"num_input_tokens_seen": 80704,
|
| 527 |
+
"step": 295
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.6548672566371683,
|
| 531 |
+
"grad_norm": 4.458314895629883,
|
| 532 |
+
"learning_rate": 4.5985671302276165e-05,
|
| 533 |
+
"loss": 0.1831,
|
| 534 |
+
"num_input_tokens_seen": 82080,
|
| 535 |
+
"step": 300
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.6991150442477876,
|
| 539 |
+
"grad_norm": 0.6618825793266296,
|
| 540 |
+
"learning_rate": 4.577332281210727e-05,
|
| 541 |
+
"loss": 0.249,
|
| 542 |
+
"num_input_tokens_seen": 83520,
|
| 543 |
+
"step": 305
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.7433628318584073,
|
| 547 |
+
"grad_norm": 0.4620368778705597,
|
| 548 |
+
"learning_rate": 4.55560187340322e-05,
|
| 549 |
+
"loss": 0.328,
|
| 550 |
+
"num_input_tokens_seen": 84832,
|
| 551 |
+
"step": 310
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 2.7876106194690267,
|
| 555 |
+
"grad_norm": 0.35645490884780884,
|
| 556 |
+
"learning_rate": 4.533381090710776e-05,
|
| 557 |
+
"loss": 0.0369,
|
| 558 |
+
"num_input_tokens_seen": 86144,
|
| 559 |
+
"step": 315
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.831858407079646,
|
| 563 |
+
"grad_norm": 9.479719161987305,
|
| 564 |
+
"learning_rate": 4.5106752340206435e-05,
|
| 565 |
+
"loss": 0.5349,
|
| 566 |
+
"num_input_tokens_seen": 87568,
|
| 567 |
+
"step": 320
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.8761061946902657,
|
| 571 |
+
"grad_norm": 0.5001351833343506,
|
| 572 |
+
"learning_rate": 4.4874897199370805e-05,
|
| 573 |
+
"loss": 0.1872,
|
| 574 |
+
"num_input_tokens_seen": 89024,
|
| 575 |
+
"step": 325
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.920353982300885,
|
| 579 |
+
"grad_norm": 8.476178169250488,
|
| 580 |
+
"learning_rate": 4.463830079489196e-05,
|
| 581 |
+
"loss": 0.4609,
|
| 582 |
+
"num_input_tokens_seen": 90320,
|
| 583 |
+
"step": 330
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.9646017699115044,
|
| 587 |
+
"grad_norm": 0.24967752397060394,
|
| 588 |
+
"learning_rate": 4.4397019568114987e-05,
|
| 589 |
+
"loss": 0.1262,
|
| 590 |
+
"num_input_tokens_seen": 91952,
|
| 591 |
+
"step": 335
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.0088495575221237,
|
| 595 |
+
"grad_norm": 1.342038631439209,
|
| 596 |
+
"learning_rate": 4.415111107797445e-05,
|
| 597 |
+
"loss": 0.545,
|
| 598 |
+
"num_input_tokens_seen": 93496,
|
| 599 |
+
"step": 340
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 3.0265486725663715,
|
| 603 |
+
"eval_loss": 0.43351349234580994,
|
| 604 |
+
"eval_runtime": 0.7069,
|
| 605 |
+
"eval_samples_per_second": 35.365,
|
| 606 |
+
"eval_steps_per_second": 18.39,
|
| 607 |
+
"num_input_tokens_seen": 94136,
|
| 608 |
+
"step": 342
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.0530973451327434,
|
| 612 |
+
"grad_norm": 0.784528374671936,
|
| 613 |
+
"learning_rate": 4.390063398726356e-05,
|
| 614 |
+
"loss": 0.1923,
|
| 615 |
+
"num_input_tokens_seen": 94984,
|
| 616 |
+
"step": 345
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 3.0973451327433628,
|
| 620 |
+
"grad_norm": 8.556395530700684,
|
| 621 |
+
"learning_rate": 4.3645648048639734e-05,
|
| 622 |
+
"loss": 0.1364,
|
| 623 |
+
"num_input_tokens_seen": 96296,
|
| 624 |
+
"step": 350
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.1415929203539825,
|
| 628 |
+
"grad_norm": 0.44090715050697327,
|
| 629 |
+
"learning_rate": 4.338621409037031e-05,
|
| 630 |
+
"loss": 0.2228,
|
| 631 |
+
"num_input_tokens_seen": 97528,
|
| 632 |
+
"step": 355
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 3.185840707964602,
|
| 636 |
+
"grad_norm": 4.060224533081055,
|
| 637 |
+
"learning_rate": 4.312239400182166e-05,
|
| 638 |
+
"loss": 0.1388,
|
| 639 |
+
"num_input_tokens_seen": 98760,
|
| 640 |
+
"step": 360
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 3.230088495575221,
|
| 644 |
+
"grad_norm": 2.259648084640503,
|
| 645 |
+
"learning_rate": 4.285425071869511e-05,
|
| 646 |
+
"loss": 0.5518,
|
| 647 |
+
"num_input_tokens_seen": 99864,
|
| 648 |
+
"step": 365
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 3.274336283185841,
|
| 652 |
+
"grad_norm": 1.4717310667037964,
|
| 653 |
+
"learning_rate": 4.258184820801331e-05,
|
| 654 |
+
"loss": 0.1861,
|
| 655 |
+
"num_input_tokens_seen": 101208,
|
| 656 |
+
"step": 370
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 3.3185840707964602,
|
| 660 |
+
"grad_norm": 0.34744587540626526,
|
| 661 |
+
"learning_rate": 4.230525145286057e-05,
|
| 662 |
+
"loss": 0.0693,
|
| 663 |
+
"num_input_tokens_seen": 102520,
|
| 664 |
+
"step": 375
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 3.3628318584070795,
|
| 668 |
+
"grad_norm": 6.193620204925537,
|
| 669 |
+
"learning_rate": 4.2024526436880785e-05,
|
| 670 |
+
"loss": 0.2851,
|
| 671 |
+
"num_input_tokens_seen": 104024,
|
| 672 |
+
"step": 380
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 3.4070796460176993,
|
| 676 |
+
"grad_norm": 2.856996774673462,
|
| 677 |
+
"learning_rate": 4.173974012853665e-05,
|
| 678 |
+
"loss": 0.2516,
|
| 679 |
+
"num_input_tokens_seen": 105784,
|
| 680 |
+
"step": 385
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 3.4513274336283186,
|
| 684 |
+
"grad_norm": 1.0180028676986694,
|
| 685 |
+
"learning_rate": 4.1450960465134025e-05,
|
| 686 |
+
"loss": 0.4014,
|
| 687 |
+
"num_input_tokens_seen": 107000,
|
| 688 |
+
"step": 390
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 3.495575221238938,
|
| 692 |
+
"grad_norm": 3.6202290058135986,
|
| 693 |
+
"learning_rate": 4.1158256336615096e-05,
|
| 694 |
+
"loss": 0.3603,
|
| 695 |
+
"num_input_tokens_seen": 108536,
|
| 696 |
+
"step": 395
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 3.5309734513274336,
|
| 700 |
+
"eval_loss": 0.23764507472515106,
|
| 701 |
+
"eval_runtime": 0.7108,
|
| 702 |
+
"eval_samples_per_second": 35.174,
|
| 703 |
+
"eval_steps_per_second": 18.29,
|
| 704 |
+
"num_input_tokens_seen": 109880,
|
| 705 |
+
"step": 399
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.5398230088495577,
|
| 709 |
+
"grad_norm": 3.1358914375305176,
|
| 710 |
+
"learning_rate": 4.086169756912431e-05,
|
| 711 |
+
"loss": 0.1939,
|
| 712 |
+
"num_input_tokens_seen": 110088,
|
| 713 |
+
"step": 400
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 3.584070796460177,
|
| 717 |
+
"grad_norm": 1.656354546546936,
|
| 718 |
+
"learning_rate": 4.056135490835098e-05,
|
| 719 |
+
"loss": 0.11,
|
| 720 |
+
"num_input_tokens_seen": 111752,
|
| 721 |
+
"step": 405
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 3.6283185840707963,
|
| 725 |
+
"grad_norm": 0.7589633464813232,
|
| 726 |
+
"learning_rate": 4.025730000265251e-05,
|
| 727 |
+
"loss": 0.3031,
|
| 728 |
+
"num_input_tokens_seen": 112808,
|
| 729 |
+
"step": 410
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.672566371681416,
|
| 733 |
+
"grad_norm": 4.3416032791137695,
|
| 734 |
+
"learning_rate": 3.994960538596232e-05,
|
| 735 |
+
"loss": 0.1994,
|
| 736 |
+
"num_input_tokens_seen": 114264,
|
| 737 |
+
"step": 415
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 3.7168141592920354,
|
| 741 |
+
"grad_norm": 5.597128868103027,
|
| 742 |
+
"learning_rate": 3.963834446048644e-05,
|
| 743 |
+
"loss": 0.2133,
|
| 744 |
+
"num_input_tokens_seen": 115608,
|
| 745 |
+
"step": 420
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 3.7610619469026547,
|
| 749 |
+
"grad_norm": 8.884173393249512,
|
| 750 |
+
"learning_rate": 3.9323591479193074e-05,
|
| 751 |
+
"loss": 0.278,
|
| 752 |
+
"num_input_tokens_seen": 116888,
|
| 753 |
+
"step": 425
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.8053097345132745,
|
| 757 |
+
"grad_norm": 0.49469271302223206,
|
| 758 |
+
"learning_rate": 3.90054215280991e-05,
|
| 759 |
+
"loss": 0.2608,
|
| 760 |
+
"num_input_tokens_seen": 118648,
|
| 761 |
+
"step": 430
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.849557522123894,
|
| 765 |
+
"grad_norm": 1.2819952964782715,
|
| 766 |
+
"learning_rate": 3.8683910508357934e-05,
|
| 767 |
+
"loss": 0.0295,
|
| 768 |
+
"num_input_tokens_seen": 119864,
|
| 769 |
+
"step": 435
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 3.893805309734513,
|
| 773 |
+
"grad_norm": 0.17074313759803772,
|
| 774 |
+
"learning_rate": 3.83591351181529e-05,
|
| 775 |
+
"loss": 0.332,
|
| 776 |
+
"num_input_tokens_seen": 121144,
|
| 777 |
+
"step": 440
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 3.938053097345133,
|
| 781 |
+
"grad_norm": 2.5671634674072266,
|
| 782 |
+
"learning_rate": 3.803117283440044e-05,
|
| 783 |
+
"loss": 0.059,
|
| 784 |
+
"num_input_tokens_seen": 122504,
|
| 785 |
+
"step": 445
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 3.982300884955752,
|
| 789 |
+
"grad_norm": 3.7636024951934814,
|
| 790 |
+
"learning_rate": 3.7700101894267615e-05,
|
| 791 |
+
"loss": 0.3693,
|
| 792 |
+
"num_input_tokens_seen": 123784,
|
| 793 |
+
"step": 450
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 4.0265486725663715,
|
| 797 |
+
"grad_norm": 23.92598533630371,
|
| 798 |
+
"learning_rate": 3.736600127650817e-05,
|
| 799 |
+
"loss": 0.4205,
|
| 800 |
+
"num_input_tokens_seen": 124728,
|
| 801 |
+
"step": 455
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 4.035398230088496,
|
| 805 |
+
"eval_loss": 0.2482675462961197,
|
| 806 |
+
"eval_runtime": 0.7058,
|
| 807 |
+
"eval_samples_per_second": 35.421,
|
| 808 |
+
"eval_steps_per_second": 18.419,
|
| 809 |
+
"num_input_tokens_seen": 125144,
|
| 810 |
+
"step": 456
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 4.070796460176991,
|
| 814 |
+
"grad_norm": 7.132582664489746,
|
| 815 |
+
"learning_rate": 3.702895068262177e-05,
|
| 816 |
+
"loss": 0.1596,
|
| 817 |
+
"num_input_tokens_seen": 126264,
|
| 818 |
+
"step": 460
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 4.115044247787611,
|
| 822 |
+
"grad_norm": 6.178963661193848,
|
| 823 |
+
"learning_rate": 3.66890305178407e-05,
|
| 824 |
+
"loss": 0.3773,
|
| 825 |
+
"num_input_tokens_seen": 127400,
|
| 826 |
+
"step": 465
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 4.15929203539823,
|
| 830 |
+
"grad_norm": 0.4468452036380768,
|
| 831 |
+
"learning_rate": 3.63463218719489e-05,
|
| 832 |
+
"loss": 0.1251,
|
| 833 |
+
"num_input_tokens_seen": 128808,
|
| 834 |
+
"step": 470
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 4.20353982300885,
|
| 838 |
+
"grad_norm": 1.1716653108596802,
|
| 839 |
+
"learning_rate": 3.600090649993741e-05,
|
| 840 |
+
"loss": 0.0315,
|
| 841 |
+
"num_input_tokens_seen": 130184,
|
| 842 |
+
"step": 475
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 4.247787610619469,
|
| 846 |
+
"grad_norm": 2.446228265762329,
|
| 847 |
+
"learning_rate": 3.565286680250138e-05,
|
| 848 |
+
"loss": 0.1617,
|
| 849 |
+
"num_input_tokens_seen": 131352,
|
| 850 |
+
"step": 480
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 4.292035398230088,
|
| 854 |
+
"grad_norm": 16.807741165161133,
|
| 855 |
+
"learning_rate": 3.530228580638291e-05,
|
| 856 |
+
"loss": 0.2307,
|
| 857 |
+
"num_input_tokens_seen": 132632,
|
| 858 |
+
"step": 485
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 4.336283185840708,
|
| 862 |
+
"grad_norm": 14.887909889221191,
|
| 863 |
+
"learning_rate": 3.494924714456454e-05,
|
| 864 |
+
"loss": 0.2609,
|
| 865 |
+
"num_input_tokens_seen": 133816,
|
| 866 |
+
"step": 490
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 4.380530973451328,
|
| 870 |
+
"grad_norm": 3.152334690093994,
|
| 871 |
+
"learning_rate": 3.459383503631823e-05,
|
| 872 |
+
"loss": 0.0534,
|
| 873 |
+
"num_input_tokens_seen": 135336,
|
| 874 |
+
"step": 495
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 4.424778761061947,
|
| 878 |
+
"grad_norm": 0.1275576651096344,
|
| 879 |
+
"learning_rate": 3.423613426711431e-05,
|
| 880 |
+
"loss": 0.4134,
|
| 881 |
+
"num_input_tokens_seen": 136504,
|
| 882 |
+
"step": 500
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 4.469026548672566,
|
| 886 |
+
"grad_norm": 1.5385409593582153,
|
| 887 |
+
"learning_rate": 3.3876230168395525e-05,
|
| 888 |
+
"loss": 0.0234,
|
| 889 |
+
"num_input_tokens_seen": 137656,
|
| 890 |
+
"step": 505
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 4.513274336283186,
|
| 894 |
+
"grad_norm": 7.704578876495361,
|
| 895 |
+
"learning_rate": 3.3514208597220705e-05,
|
| 896 |
+
"loss": 0.5197,
|
| 897 |
+
"num_input_tokens_seen": 138776,
|
| 898 |
+
"step": 510
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 4.539823008849558,
|
| 902 |
+
"eval_loss": 0.18963883817195892,
|
| 903 |
+
"eval_runtime": 0.7031,
|
| 904 |
+
"eval_samples_per_second": 35.558,
|
| 905 |
+
"eval_steps_per_second": 18.49,
|
| 906 |
+
"num_input_tokens_seen": 139544,
|
| 907 |
+
"step": 513
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 4.557522123893805,
|
| 911 |
+
"grad_norm": 0.17451319098472595,
|
| 912 |
+
"learning_rate": 3.315015591578314e-05,
|
| 913 |
+
"loss": 0.2338,
|
| 914 |
+
"num_input_tokens_seen": 140296,
|
| 915 |
+
"step": 515
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 4.601769911504425,
|
| 919 |
+
"grad_norm": 4.632303714752197,
|
| 920 |
+
"learning_rate": 3.278415897080839e-05,
|
| 921 |
+
"loss": 0.0408,
|
| 922 |
+
"num_input_tokens_seen": 141720,
|
| 923 |
+
"step": 520
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 4.646017699115045,
|
| 927 |
+
"grad_norm": 0.036415815353393555,
|
| 928 |
+
"learning_rate": 3.2416305072836557e-05,
|
| 929 |
+
"loss": 0.4261,
|
| 930 |
+
"num_input_tokens_seen": 143336,
|
| 931 |
+
"step": 525
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 4.6902654867256635,
|
| 935 |
+
"grad_norm": 2.5997607707977295,
|
| 936 |
+
"learning_rate": 3.204668197539385e-05,
|
| 937 |
+
"loss": 0.0505,
|
| 938 |
+
"num_input_tokens_seen": 145016,
|
| 939 |
+
"step": 530
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 4.734513274336283,
|
| 943 |
+
"grad_norm": 0.5314985513687134,
|
| 944 |
+
"learning_rate": 3.167537785405854e-05,
|
| 945 |
+
"loss": 0.1154,
|
| 946 |
+
"num_input_tokens_seen": 146296,
|
| 947 |
+
"step": 535
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 4.778761061946903,
|
| 951 |
+
"grad_norm": 3.233059883117676,
|
| 952 |
+
"learning_rate": 3.13024812854262e-05,
|
| 953 |
+
"loss": 0.0516,
|
| 954 |
+
"num_input_tokens_seen": 147656,
|
| 955 |
+
"step": 540
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 4.823008849557522,
|
| 959 |
+
"grad_norm": 5.46919059753418,
|
| 960 |
+
"learning_rate": 3.092808122597924e-05,
|
| 961 |
+
"loss": 0.3412,
|
| 962 |
+
"num_input_tokens_seen": 148808,
|
| 963 |
+
"step": 545
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 4.867256637168142,
|
| 967 |
+
"grad_norm": 0.19922271370887756,
|
| 968 |
+
"learning_rate": 3.0552266990865934e-05,
|
| 969 |
+
"loss": 0.0287,
|
| 970 |
+
"num_input_tokens_seen": 150680,
|
| 971 |
+
"step": 550
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 4.911504424778761,
|
| 975 |
+
"grad_norm": 0.04063204303383827,
|
| 976 |
+
"learning_rate": 3.017512823259373e-05,
|
| 977 |
+
"loss": 0.0647,
|
| 978 |
+
"num_input_tokens_seen": 152392,
|
| 979 |
+
"step": 555
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 4.95575221238938,
|
| 983 |
+
"grad_norm": 0.041708074510097504,
|
| 984 |
+
"learning_rate": 2.9796754919642228e-05,
|
| 985 |
+
"loss": 0.1071,
|
| 986 |
+
"num_input_tokens_seen": 153896,
|
| 987 |
+
"step": 560
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 5.0,
|
| 991 |
+
"grad_norm": 5.79632568359375,
|
| 992 |
+
"learning_rate": 2.941723731500068e-05,
|
| 993 |
+
"loss": 0.2245,
|
| 994 |
+
"num_input_tokens_seen": 155240,
|
| 995 |
+
"step": 565
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 5.04424778761062,
|
| 999 |
+
"grad_norm": 5.1361565589904785,
|
| 1000 |
+
"learning_rate": 2.9036665954635266e-05,
|
| 1001 |
+
"loss": 0.1052,
|
| 1002 |
+
"num_input_tokens_seen": 156296,
|
| 1003 |
+
"step": 570
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 5.04424778761062,
|
| 1007 |
+
"eval_loss": 0.1859114021062851,
|
| 1008 |
+
"eval_runtime": 0.7062,
|
| 1009 |
+
"eval_samples_per_second": 35.402,
|
| 1010 |
+
"eval_steps_per_second": 18.409,
|
| 1011 |
+
"num_input_tokens_seen": 156296,
|
| 1012 |
+
"step": 570
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 5.088495575221239,
|
| 1016 |
+
"grad_norm": 0.14968609809875488,
|
| 1017 |
+
"learning_rate": 2.8655131625891274e-05,
|
| 1018 |
+
"loss": 0.0817,
|
| 1019 |
+
"num_input_tokens_seen": 157688,
|
| 1020 |
+
"step": 575
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 5.132743362831858,
|
| 1024 |
+
"grad_norm": 4.9272942543029785,
|
| 1025 |
+
"learning_rate": 2.8272725345835278e-05,
|
| 1026 |
+
"loss": 0.1101,
|
| 1027 |
+
"num_input_tokens_seen": 159016,
|
| 1028 |
+
"step": 580
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 5.176991150442478,
|
| 1032 |
+
"grad_norm": 0.2512366473674774,
|
| 1033 |
+
"learning_rate": 2.7889538339542527e-05,
|
| 1034 |
+
"loss": 0.1267,
|
| 1035 |
+
"num_input_tokens_seen": 160552,
|
| 1036 |
+
"step": 585
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 5.221238938053097,
|
| 1040 |
+
"grad_norm": 0.1694510281085968,
|
| 1041 |
+
"learning_rate": 2.7505662018334698e-05,
|
| 1042 |
+
"loss": 0.1858,
|
| 1043 |
+
"num_input_tokens_seen": 161912,
|
| 1044 |
+
"step": 590
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 5.265486725663717,
|
| 1048 |
+
"grad_norm": 0.2537103295326233,
|
| 1049 |
+
"learning_rate": 2.712118795797329e-05,
|
| 1050 |
+
"loss": 0.238,
|
| 1051 |
+
"num_input_tokens_seen": 163288,
|
| 1052 |
+
"step": 595
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 5.3097345132743365,
|
| 1056 |
+
"grad_norm": 1.7408089637756348,
|
| 1057 |
+
"learning_rate": 2.6736207876813646e-05,
|
| 1058 |
+
"loss": 0.0122,
|
| 1059 |
+
"num_input_tokens_seen": 164920,
|
| 1060 |
+
"step": 600
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 5.353982300884955,
|
| 1064 |
+
"grad_norm": 0.0584169365465641,
|
| 1065 |
+
"learning_rate": 2.635081361392513e-05,
|
| 1066 |
+
"loss": 0.0537,
|
| 1067 |
+
"num_input_tokens_seen": 166296,
|
| 1068 |
+
"step": 605
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 5.398230088495575,
|
| 1072 |
+
"grad_norm": 0.6817641854286194,
|
| 1073 |
+
"learning_rate": 2.596509710718235e-05,
|
| 1074 |
+
"loss": 0.0053,
|
| 1075 |
+
"num_input_tokens_seen": 167720,
|
| 1076 |
+
"step": 610
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 5.442477876106195,
|
| 1080 |
+
"grad_norm": 0.03988633677363396,
|
| 1081 |
+
"learning_rate": 2.5579150371332956e-05,
|
| 1082 |
+
"loss": 0.1432,
|
| 1083 |
+
"num_input_tokens_seen": 169480,
|
| 1084 |
+
"step": 615
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 5.486725663716814,
|
| 1088 |
+
"grad_norm": 0.04661005362868309,
|
| 1089 |
+
"learning_rate": 2.5193065476046955e-05,
|
| 1090 |
+
"loss": 0.4185,
|
| 1091 |
+
"num_input_tokens_seen": 171032,
|
| 1092 |
+
"step": 620
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 5.530973451327434,
|
| 1096 |
+
"grad_norm": 4.454204082489014,
|
| 1097 |
+
"learning_rate": 2.480693452395305e-05,
|
| 1098 |
+
"loss": 0.3257,
|
| 1099 |
+
"num_input_tokens_seen": 172072,
|
| 1100 |
+
"step": 625
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 5.548672566371682,
|
| 1104 |
+
"eval_loss": 0.20005442202091217,
|
| 1105 |
+
"eval_runtime": 0.7078,
|
| 1106 |
+
"eval_samples_per_second": 35.322,
|
| 1107 |
+
"eval_steps_per_second": 18.367,
|
| 1108 |
+
"num_input_tokens_seen": 172504,
|
| 1109 |
+
"step": 627
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 5.575221238938053,
|
| 1113 |
+
"grad_norm": 0.203868567943573,
|
| 1114 |
+
"learning_rate": 2.4420849628667046e-05,
|
| 1115 |
+
"loss": 0.1605,
|
| 1116 |
+
"num_input_tokens_seen": 173368,
|
| 1117 |
+
"step": 630
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 5.619469026548672,
|
| 1121 |
+
"grad_norm": 8.949444770812988,
|
| 1122 |
+
"learning_rate": 2.4034902892817648e-05,
|
| 1123 |
+
"loss": 0.2972,
|
| 1124 |
+
"num_input_tokens_seen": 174520,
|
| 1125 |
+
"step": 635
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 5.663716814159292,
|
| 1129 |
+
"grad_norm": 0.9069784283638,
|
| 1130 |
+
"learning_rate": 2.3649186386074872e-05,
|
| 1131 |
+
"loss": 0.1017,
|
| 1132 |
+
"num_input_tokens_seen": 175976,
|
| 1133 |
+
"step": 640
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 5.707964601769912,
|
| 1137 |
+
"grad_norm": 1.71530282497406,
|
| 1138 |
+
"learning_rate": 2.3263792123186353e-05,
|
| 1139 |
+
"loss": 0.1992,
|
| 1140 |
+
"num_input_tokens_seen": 177240,
|
| 1141 |
+
"step": 645
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 5.752212389380531,
|
| 1145 |
+
"grad_norm": 5.149680137634277,
|
| 1146 |
+
"learning_rate": 2.2878812042026713e-05,
|
| 1147 |
+
"loss": 0.2113,
|
| 1148 |
+
"num_input_tokens_seen": 178808,
|
| 1149 |
+
"step": 650
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 5.79646017699115,
|
| 1153 |
+
"grad_norm": 0.13400527834892273,
|
| 1154 |
+
"learning_rate": 2.2494337981665304e-05,
|
| 1155 |
+
"loss": 0.1821,
|
| 1156 |
+
"num_input_tokens_seen": 180024,
|
| 1157 |
+
"step": 655
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 5.84070796460177,
|
| 1161 |
+
"grad_norm": 0.34926721453666687,
|
| 1162 |
+
"learning_rate": 2.2110461660457482e-05,
|
| 1163 |
+
"loss": 0.0488,
|
| 1164 |
+
"num_input_tokens_seen": 181368,
|
| 1165 |
+
"step": 660
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 5.88495575221239,
|
| 1169 |
+
"grad_norm": 0.0739203616976738,
|
| 1170 |
+
"learning_rate": 2.1727274654164728e-05,
|
| 1171 |
+
"loss": 0.0232,
|
| 1172 |
+
"num_input_tokens_seen": 182696,
|
| 1173 |
+
"step": 665
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 5.929203539823009,
|
| 1177 |
+
"grad_norm": 0.5862051248550415,
|
| 1178 |
+
"learning_rate": 2.1344868374108732e-05,
|
| 1179 |
+
"loss": 0.0654,
|
| 1180 |
+
"num_input_tokens_seen": 184232,
|
| 1181 |
+
"step": 670
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 5.9734513274336285,
|
| 1185 |
+
"grad_norm": 1.7623151540756226,
|
| 1186 |
+
"learning_rate": 2.096333404536474e-05,
|
| 1187 |
+
"loss": 0.1758,
|
| 1188 |
+
"num_input_tokens_seen": 185496,
|
| 1189 |
+
"step": 675
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 6.017699115044247,
|
| 1193 |
+
"grad_norm": 0.030710099264979362,
|
| 1194 |
+
"learning_rate": 2.0582762684999327e-05,
|
| 1195 |
+
"loss": 0.0065,
|
| 1196 |
+
"num_input_tokens_seen": 186464,
|
| 1197 |
+
"step": 680
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 6.053097345132743,
|
| 1201 |
+
"eval_loss": 0.2582325339317322,
|
| 1202 |
+
"eval_runtime": 0.7044,
|
| 1203 |
+
"eval_samples_per_second": 35.491,
|
| 1204 |
+
"eval_steps_per_second": 18.455,
|
| 1205 |
+
"num_input_tokens_seen": 187664,
|
| 1206 |
+
"step": 684
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 6.061946902654867,
|
| 1210 |
+
"grad_norm": 0.4586619734764099,
|
| 1211 |
+
"learning_rate": 2.0203245080357775e-05,
|
| 1212 |
+
"loss": 0.0646,
|
| 1213 |
+
"num_input_tokens_seen": 187936,
|
| 1214 |
+
"step": 685
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 6.106194690265487,
|
| 1218 |
+
"grad_norm": 0.207475945353508,
|
| 1219 |
+
"learning_rate": 1.982487176740627e-05,
|
| 1220 |
+
"loss": 0.1359,
|
| 1221 |
+
"num_input_tokens_seen": 189600,
|
| 1222 |
+
"step": 690
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 6.150442477876107,
|
| 1226 |
+
"grad_norm": 16.1010684967041,
|
| 1227 |
+
"learning_rate": 1.944773300913407e-05,
|
| 1228 |
+
"loss": 0.0468,
|
| 1229 |
+
"num_input_tokens_seen": 190896,
|
| 1230 |
+
"step": 695
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 6.1946902654867255,
|
| 1234 |
+
"grad_norm": 16.56742286682129,
|
| 1235 |
+
"learning_rate": 1.907191877402076e-05,
|
| 1236 |
+
"loss": 0.2096,
|
| 1237 |
+
"num_input_tokens_seen": 192496,
|
| 1238 |
+
"step": 700
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 6.238938053097345,
|
| 1242 |
+
"grad_norm": 0.22490589320659637,
|
| 1243 |
+
"learning_rate": 1.8697518714573807e-05,
|
| 1244 |
+
"loss": 0.0786,
|
| 1245 |
+
"num_input_tokens_seen": 193968,
|
| 1246 |
+
"step": 705
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 6.283185840707965,
|
| 1250 |
+
"grad_norm": 0.7650446891784668,
|
| 1251 |
+
"learning_rate": 1.832462214594146e-05,
|
| 1252 |
+
"loss": 0.3287,
|
| 1253 |
+
"num_input_tokens_seen": 195232,
|
| 1254 |
+
"step": 710
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 6.327433628318584,
|
| 1258 |
+
"grad_norm": 0.919759213924408,
|
| 1259 |
+
"learning_rate": 1.7953318024606157e-05,
|
| 1260 |
+
"loss": 0.0111,
|
| 1261 |
+
"num_input_tokens_seen": 196880,
|
| 1262 |
+
"step": 715
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 6.371681415929204,
|
| 1266 |
+
"grad_norm": 0.16696296632289886,
|
| 1267 |
+
"learning_rate": 1.7583694927163453e-05,
|
| 1268 |
+
"loss": 0.122,
|
| 1269 |
+
"num_input_tokens_seen": 198336,
|
| 1270 |
+
"step": 720
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 6.415929203539823,
|
| 1274 |
+
"grad_norm": 0.09723463654518127,
|
| 1275 |
+
"learning_rate": 1.7215841029191616e-05,
|
| 1276 |
+
"loss": 0.1655,
|
| 1277 |
+
"num_input_tokens_seen": 199632,
|
| 1278 |
+
"step": 725
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 6.460176991150442,
|
| 1282 |
+
"grad_norm": 5.961605072021484,
|
| 1283 |
+
"learning_rate": 1.6849844084216874e-05,
|
| 1284 |
+
"loss": 0.0792,
|
| 1285 |
+
"num_input_tokens_seen": 200944,
|
| 1286 |
+
"step": 730
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 6.504424778761062,
|
| 1290 |
+
"grad_norm": 0.780555784702301,
|
| 1291 |
+
"learning_rate": 1.648579140277931e-05,
|
| 1292 |
+
"loss": 0.0897,
|
| 1293 |
+
"num_input_tokens_seen": 202448,
|
| 1294 |
+
"step": 735
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 6.548672566371682,
|
| 1298 |
+
"grad_norm": 7.634582996368408,
|
| 1299 |
+
"learning_rate": 1.612376983160449e-05,
|
| 1300 |
+
"loss": 0.0299,
|
| 1301 |
+
"num_input_tokens_seen": 203952,
|
| 1302 |
+
"step": 740
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 6.557522123893805,
|
| 1306 |
+
"eval_loss": 0.2114013135433197,
|
| 1307 |
+
"eval_runtime": 0.7072,
|
| 1308 |
+
"eval_samples_per_second": 35.351,
|
| 1309 |
+
"eval_steps_per_second": 18.382,
|
| 1310 |
+
"num_input_tokens_seen": 204128,
|
| 1311 |
+
"step": 741
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 6.592920353982301,
|
| 1315 |
+
"grad_norm": 0.6068083643913269,
|
| 1316 |
+
"learning_rate": 1.576386573288569e-05,
|
| 1317 |
+
"loss": 0.0098,
|
| 1318 |
+
"num_input_tokens_seen": 205104,
|
| 1319 |
+
"step": 745
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 6.6371681415929205,
|
| 1323 |
+
"grad_norm": 0.24441823363304138,
|
| 1324 |
+
"learning_rate": 1.5406164963681774e-05,
|
| 1325 |
+
"loss": 0.0353,
|
| 1326 |
+
"num_input_tokens_seen": 206656,
|
| 1327 |
+
"step": 750
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 6.68141592920354,
|
| 1331 |
+
"grad_norm": 0.02317521534860134,
|
| 1332 |
+
"learning_rate": 1.5050752855435457e-05,
|
| 1333 |
+
"loss": 0.0924,
|
| 1334 |
+
"num_input_tokens_seen": 207648,
|
| 1335 |
+
"step": 755
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 6.725663716814159,
|
| 1339 |
+
"grad_norm": 0.12233875691890717,
|
| 1340 |
+
"learning_rate": 1.4697714193617097e-05,
|
| 1341 |
+
"loss": 0.1835,
|
| 1342 |
+
"num_input_tokens_seen": 208928,
|
| 1343 |
+
"step": 760
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 6.769911504424779,
|
| 1347 |
+
"grad_norm": 0.06515863537788391,
|
| 1348 |
+
"learning_rate": 1.4347133197498619e-05,
|
| 1349 |
+
"loss": 0.3059,
|
| 1350 |
+
"num_input_tokens_seen": 210432,
|
| 1351 |
+
"step": 765
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 6.814159292035399,
|
| 1355 |
+
"grad_norm": 7.31368350982666,
|
| 1356 |
+
"learning_rate": 1.3999093500062587e-05,
|
| 1357 |
+
"loss": 0.2314,
|
| 1358 |
+
"num_input_tokens_seen": 211680,
|
| 1359 |
+
"step": 770
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 6.8584070796460175,
|
| 1363 |
+
"grad_norm": 0.24847674369812012,
|
| 1364 |
+
"learning_rate": 1.3653678128051106e-05,
|
| 1365 |
+
"loss": 0.0097,
|
| 1366 |
+
"num_input_tokens_seen": 212832,
|
| 1367 |
+
"step": 775
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 6.902654867256637,
|
| 1371 |
+
"grad_norm": 0.3456866443157196,
|
| 1372 |
+
"learning_rate": 1.3310969482159297e-05,
|
| 1373 |
+
"loss": 0.0788,
|
| 1374 |
+
"num_input_tokens_seen": 214144,
|
| 1375 |
+
"step": 780
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 6.946902654867257,
|
| 1379 |
+
"grad_norm": 0.45530736446380615,
|
| 1380 |
+
"learning_rate": 1.2971049317378237e-05,
|
| 1381 |
+
"loss": 0.0036,
|
| 1382 |
+
"num_input_tokens_seen": 215440,
|
| 1383 |
+
"step": 785
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 6.991150442477876,
|
| 1387 |
+
"grad_norm": 0.15640869736671448,
|
| 1388 |
+
"learning_rate": 1.263399872349183e-05,
|
| 1389 |
+
"loss": 0.0031,
|
| 1390 |
+
"num_input_tokens_seen": 217088,
|
| 1391 |
+
"step": 790
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 7.035398230088496,
|
| 1395 |
+
"grad_norm": 10.463488578796387,
|
| 1396 |
+
"learning_rate": 1.2299898105732385e-05,
|
| 1397 |
+
"loss": 0.1574,
|
| 1398 |
+
"num_input_tokens_seen": 218712,
|
| 1399 |
+
"step": 795
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 7.061946902654867,
|
| 1403 |
+
"eval_loss": 0.31536608934402466,
|
| 1404 |
+
"eval_runtime": 0.7066,
|
| 1405 |
+
"eval_samples_per_second": 35.379,
|
| 1406 |
+
"eval_steps_per_second": 18.397,
|
| 1407 |
+
"num_input_tokens_seen": 219400,
|
| 1408 |
+
"step": 798
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 7.079646017699115,
|
| 1412 |
+
"grad_norm": 0.05625517666339874,
|
| 1413 |
+
"learning_rate": 1.1968827165599564e-05,
|
| 1414 |
+
"loss": 0.123,
|
| 1415 |
+
"num_input_tokens_seen": 219944,
|
| 1416 |
+
"step": 800
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 7.123893805309734,
|
| 1420 |
+
"grad_norm": 0.06226060166954994,
|
| 1421 |
+
"learning_rate": 1.1640864881847105e-05,
|
| 1422 |
+
"loss": 0.1137,
|
| 1423 |
+
"num_input_tokens_seen": 221304,
|
| 1424 |
+
"step": 805
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 7.168141592920354,
|
| 1428 |
+
"grad_norm": 0.20677171647548676,
|
| 1429 |
+
"learning_rate": 1.1316089491642076e-05,
|
| 1430 |
+
"loss": 0.0069,
|
| 1431 |
+
"num_input_tokens_seen": 222680,
|
| 1432 |
+
"step": 810
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 7.212389380530974,
|
| 1436 |
+
"grad_norm": 6.473257064819336,
|
| 1437 |
+
"learning_rate": 1.099457847190091e-05,
|
| 1438 |
+
"loss": 0.0752,
|
| 1439 |
+
"num_input_tokens_seen": 224120,
|
| 1440 |
+
"step": 815
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 7.256637168141593,
|
| 1444 |
+
"grad_norm": 2.6501564979553223,
|
| 1445 |
+
"learning_rate": 1.067640852080693e-05,
|
| 1446 |
+
"loss": 0.0092,
|
| 1447 |
+
"num_input_tokens_seen": 225448,
|
| 1448 |
+
"step": 820
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 7.300884955752212,
|
| 1452 |
+
"grad_norm": 9.59801959991455,
|
| 1453 |
+
"learning_rate": 1.0361655539513565e-05,
|
| 1454 |
+
"loss": 0.0214,
|
| 1455 |
+
"num_input_tokens_seen": 226840,
|
| 1456 |
+
"step": 825
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 7.345132743362832,
|
| 1460 |
+
"grad_norm": 0.01992979645729065,
|
| 1461 |
+
"learning_rate": 1.0050394614037687e-05,
|
| 1462 |
+
"loss": 0.0416,
|
| 1463 |
+
"num_input_tokens_seen": 227928,
|
| 1464 |
+
"step": 830
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 7.389380530973451,
|
| 1468 |
+
"grad_norm": 0.19040854275226593,
|
| 1469 |
+
"learning_rate": 9.742699997347498e-06,
|
| 1470 |
+
"loss": 0.0043,
|
| 1471 |
+
"num_input_tokens_seen": 229336,
|
| 1472 |
+
"step": 835
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 7.433628318584071,
|
| 1476 |
+
"grad_norm": 0.17141713201999664,
|
| 1477 |
+
"learning_rate": 9.438645091649029e-06,
|
| 1478 |
+
"loss": 0.0196,
|
| 1479 |
+
"num_input_tokens_seen": 231016,
|
| 1480 |
+
"step": 840
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 7.477876106194691,
|
| 1484 |
+
"grad_norm": 0.08842453360557556,
|
| 1485 |
+
"learning_rate": 9.138302430875695e-06,
|
| 1486 |
+
"loss": 0.0075,
|
| 1487 |
+
"num_input_tokens_seen": 232296,
|
| 1488 |
+
"step": 845
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 7.522123893805309,
|
| 1492 |
+
"grad_norm": 0.12026006728410721,
|
| 1493 |
+
"learning_rate": 8.841743663384913e-06,
|
| 1494 |
+
"loss": 0.002,
|
| 1495 |
+
"num_input_tokens_seen": 233432,
|
| 1496 |
+
"step": 850
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 7.566371681415929,
|
| 1500 |
+
"grad_norm": 0.1048886701464653,
|
| 1501 |
+
"learning_rate": 8.549039534865979e-06,
|
| 1502 |
+
"loss": 0.0121,
|
| 1503 |
+
"num_input_tokens_seen": 235224,
|
| 1504 |
+
"step": 855
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 7.566371681415929,
|
| 1508 |
+
"eval_loss": 0.29263150691986084,
|
| 1509 |
+
"eval_runtime": 0.7097,
|
| 1510 |
+
"eval_samples_per_second": 35.224,
|
| 1511 |
+
"eval_steps_per_second": 18.316,
|
| 1512 |
+
"num_input_tokens_seen": 235224,
|
| 1513 |
+
"step": 855
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 7.610619469026549,
|
| 1517 |
+
"grad_norm": 8.880864143371582,
|
| 1518 |
+
"learning_rate": 8.260259871463352e-06,
|
| 1519 |
+
"loss": 0.1689,
|
| 1520 |
+
"num_input_tokens_seen": 236456,
|
| 1521 |
+
"step": 860
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 7.654867256637168,
|
| 1525 |
+
"grad_norm": 0.07442810386419296,
|
| 1526 |
+
"learning_rate": 7.975473563119216e-06,
|
| 1527 |
+
"loss": 0.0009,
|
| 1528 |
+
"num_input_tokens_seen": 237912,
|
| 1529 |
+
"step": 865
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 7.699115044247788,
|
| 1533 |
+
"grad_norm": 0.12946204841136932,
|
| 1534 |
+
"learning_rate": 7.69474854713943e-06,
|
| 1535 |
+
"loss": 0.001,
|
| 1536 |
+
"num_input_tokens_seen": 239304,
|
| 1537 |
+
"step": 870
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 7.743362831858407,
|
| 1541 |
+
"grad_norm": 10.251002311706543,
|
| 1542 |
+
"learning_rate": 7.4181517919866914e-06,
|
| 1543 |
+
"loss": 0.0955,
|
| 1544 |
+
"num_input_tokens_seen": 240616,
|
| 1545 |
+
"step": 875
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 7.787610619469026,
|
| 1549 |
+
"grad_norm": 1.2198227643966675,
|
| 1550 |
+
"learning_rate": 7.1457492813049e-06,
|
| 1551 |
+
"loss": 0.0057,
|
| 1552 |
+
"num_input_tokens_seen": 241928,
|
| 1553 |
+
"step": 880
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 7.831858407079646,
|
| 1557 |
+
"grad_norm": 0.03483714163303375,
|
| 1558 |
+
"learning_rate": 6.877605998178344e-06,
|
| 1559 |
+
"loss": 0.0418,
|
| 1560 |
+
"num_input_tokens_seen": 243432,
|
| 1561 |
+
"step": 885
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 7.876106194690266,
|
| 1565 |
+
"grad_norm": 8.647575378417969,
|
| 1566 |
+
"learning_rate": 6.613785909629686e-06,
|
| 1567 |
+
"loss": 0.0719,
|
| 1568 |
+
"num_input_tokens_seen": 244904,
|
| 1569 |
+
"step": 890
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 7.920353982300885,
|
| 1573 |
+
"grad_norm": 34.08659744262695,
|
| 1574 |
+
"learning_rate": 6.354351951360268e-06,
|
| 1575 |
+
"loss": 0.1715,
|
| 1576 |
+
"num_input_tokens_seen": 246152,
|
| 1577 |
+
"step": 895
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 7.964601769911504,
|
| 1581 |
+
"grad_norm": 0.06069398671388626,
|
| 1582 |
+
"learning_rate": 6.099366012736438e-06,
|
| 1583 |
+
"loss": 0.1907,
|
| 1584 |
+
"num_input_tokens_seen": 247320,
|
| 1585 |
+
"step": 900
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 8.008849557522124,
|
| 1589 |
+
"grad_norm": 0.18082427978515625,
|
| 1590 |
+
"learning_rate": 5.848888922025553e-06,
|
| 1591 |
+
"loss": 0.0098,
|
| 1592 |
+
"num_input_tokens_seen": 248808,
|
| 1593 |
+
"step": 905
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 8.053097345132743,
|
| 1597 |
+
"grad_norm": 0.009733038023114204,
|
| 1598 |
+
"learning_rate": 5.602980431885024e-06,
|
| 1599 |
+
"loss": 0.0126,
|
| 1600 |
+
"num_input_tokens_seen": 250056,
|
| 1601 |
+
"step": 910
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 8.070796460176991,
|
| 1605 |
+
"eval_loss": 0.2721284031867981,
|
| 1606 |
+
"eval_runtime": 0.7046,
|
| 1607 |
+
"eval_samples_per_second": 35.482,
|
| 1608 |
+
"eval_steps_per_second": 18.45,
|
| 1609 |
+
"num_input_tokens_seen": 250648,
|
| 1610 |
+
"step": 912
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 8.097345132743364,
|
| 1614 |
+
"grad_norm": 0.2778746783733368,
|
| 1615 |
+
"learning_rate": 5.361699205108042e-06,
|
| 1616 |
+
"loss": 0.0037,
|
| 1617 |
+
"num_input_tokens_seen": 251560,
|
| 1618 |
+
"step": 915
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 8.141592920353983,
|
| 1622 |
+
"grad_norm": 0.22230306267738342,
|
| 1623 |
+
"learning_rate": 5.125102800629203e-06,
|
| 1624 |
+
"loss": 0.0022,
|
| 1625 |
+
"num_input_tokens_seen": 252808,
|
| 1626 |
+
"step": 920
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 8.185840707964601,
|
| 1630 |
+
"grad_norm": 0.12998662889003754,
|
| 1631 |
+
"learning_rate": 4.893247659793568e-06,
|
| 1632 |
+
"loss": 0.0372,
|
| 1633 |
+
"num_input_tokens_seen": 254536,
|
| 1634 |
+
"step": 925
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 8.230088495575222,
|
| 1638 |
+
"grad_norm": 2.2539122104644775,
|
| 1639 |
+
"learning_rate": 4.666189092892245e-06,
|
| 1640 |
+
"loss": 0.0098,
|
| 1641 |
+
"num_input_tokens_seen": 256136,
|
| 1642 |
+
"step": 930
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 8.274336283185841,
|
| 1646 |
+
"grad_norm": 0.064559206366539,
|
| 1647 |
+
"learning_rate": 4.443981265967806e-06,
|
| 1648 |
+
"loss": 0.0667,
|
| 1649 |
+
"num_input_tokens_seen": 257320,
|
| 1650 |
+
"step": 935
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 8.31858407079646,
|
| 1654 |
+
"grad_norm": 9.912572860717773,
|
| 1655 |
+
"learning_rate": 4.226677187892739e-06,
|
| 1656 |
+
"loss": 0.1145,
|
| 1657 |
+
"num_input_tokens_seen": 258920,
|
| 1658 |
+
"step": 940
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 8.36283185840708,
|
| 1662 |
+
"grad_norm": 0.22164621949195862,
|
| 1663 |
+
"learning_rate": 4.014328697723835e-06,
|
| 1664 |
+
"loss": 0.0039,
|
| 1665 |
+
"num_input_tokens_seen": 260216,
|
| 1666 |
+
"step": 945
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 8.4070796460177,
|
| 1670 |
+
"grad_norm": 0.22859039902687073,
|
| 1671 |
+
"learning_rate": 3.806986452335748e-06,
|
| 1672 |
+
"loss": 0.0035,
|
| 1673 |
+
"num_input_tokens_seen": 261736,
|
| 1674 |
+
"step": 950
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 8.451327433628318,
|
| 1678 |
+
"grad_norm": 0.026006024330854416,
|
| 1679 |
+
"learning_rate": 3.604699914336515e-06,
|
| 1680 |
+
"loss": 0.0594,
|
| 1681 |
+
"num_input_tokens_seen": 262920,
|
| 1682 |
+
"step": 955
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 8.495575221238939,
|
| 1686 |
+
"grad_norm": 6.112636566162109,
|
| 1687 |
+
"learning_rate": 3.4075173402679574e-06,
|
| 1688 |
+
"loss": 0.0197,
|
| 1689 |
+
"num_input_tokens_seen": 264216,
|
| 1690 |
+
"step": 960
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 8.539823008849558,
|
| 1694 |
+
"grad_norm": 0.274373859167099,
|
| 1695 |
+
"learning_rate": 3.215485769093862e-06,
|
| 1696 |
+
"loss": 0.005,
|
| 1697 |
+
"num_input_tokens_seen": 265320,
|
| 1698 |
+
"step": 965
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 8.575221238938052,
|
| 1702 |
+
"eval_loss": 0.28286850452423096,
|
| 1703 |
+
"eval_runtime": 0.7016,
|
| 1704 |
+
"eval_samples_per_second": 35.635,
|
| 1705 |
+
"eval_steps_per_second": 18.53,
|
| 1706 |
+
"num_input_tokens_seen": 266456,
|
| 1707 |
+
"step": 969
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 8.584070796460177,
|
| 1711 |
+
"grad_norm": 0.031764913350343704,
|
| 1712 |
+
"learning_rate": 3.028651010978556e-06,
|
| 1713 |
+
"loss": 0.0301,
|
| 1714 |
+
"num_input_tokens_seen": 266776,
|
| 1715 |
+
"step": 970
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 8.628318584070797,
|
| 1719 |
+
"grad_norm": 0.6587602496147156,
|
| 1720 |
+
"learning_rate": 2.8470576363586633e-06,
|
| 1721 |
+
"loss": 0.005,
|
| 1722 |
+
"num_input_tokens_seen": 268296,
|
| 1723 |
+
"step": 975
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 8.672566371681416,
|
| 1727 |
+
"grad_norm": 0.06793249398469925,
|
| 1728 |
+
"learning_rate": 2.67074896531056e-06,
|
| 1729 |
+
"loss": 0.0104,
|
| 1730 |
+
"num_input_tokens_seen": 269528,
|
| 1731 |
+
"step": 980
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 8.716814159292035,
|
| 1735 |
+
"grad_norm": 0.20474901795387268,
|
| 1736 |
+
"learning_rate": 2.499767057216193e-06,
|
| 1737 |
+
"loss": 0.0425,
|
| 1738 |
+
"num_input_tokens_seen": 270808,
|
| 1739 |
+
"step": 985
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 8.761061946902656,
|
| 1743 |
+
"grad_norm": 0.015458387322723866,
|
| 1744 |
+
"learning_rate": 2.334152700729511e-06,
|
| 1745 |
+
"loss": 0.0033,
|
| 1746 |
+
"num_input_tokens_seen": 271992,
|
| 1747 |
+
"step": 990
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 8.805309734513274,
|
| 1751 |
+
"grad_norm": 0.03912809118628502,
|
| 1752 |
+
"learning_rate": 2.1739454040461794e-06,
|
| 1753 |
+
"loss": 0.0018,
|
| 1754 |
+
"num_input_tokens_seen": 273656,
|
| 1755 |
+
"step": 995
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 8.849557522123893,
|
| 1759 |
+
"grad_norm": 0.008100362494587898,
|
| 1760 |
+
"learning_rate": 2.019183385478654e-06,
|
| 1761 |
+
"loss": 0.0033,
|
| 1762 |
+
"num_input_tokens_seen": 275352,
|
| 1763 |
+
"step": 1000
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 8.893805309734514,
|
| 1767 |
+
"grad_norm": 0.06251242011785507,
|
| 1768 |
+
"learning_rate": 1.8699035643389928e-06,
|
| 1769 |
+
"loss": 0.0358,
|
| 1770 |
+
"num_input_tokens_seen": 276472,
|
| 1771 |
+
"step": 1005
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 8.938053097345133,
|
| 1775 |
+
"grad_norm": 0.2489873468875885,
|
| 1776 |
+
"learning_rate": 1.7261415521315798e-06,
|
| 1777 |
+
"loss": 0.0032,
|
| 1778 |
+
"num_input_tokens_seen": 277832,
|
| 1779 |
+
"step": 1010
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 8.982300884955752,
|
| 1783 |
+
"grad_norm": 5.699662685394287,
|
| 1784 |
+
"learning_rate": 1.587931644057783e-06,
|
| 1785 |
+
"loss": 0.1213,
|
| 1786 |
+
"num_input_tokens_seen": 279352,
|
| 1787 |
+
"step": 1015
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 9.026548672566372,
|
| 1791 |
+
"grad_norm": 0.5742537975311279,
|
| 1792 |
+
"learning_rate": 1.455306810834678e-06,
|
| 1793 |
+
"loss": 0.0038,
|
| 1794 |
+
"num_input_tokens_seen": 280608,
|
| 1795 |
+
"step": 1020
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 9.070796460176991,
|
| 1799 |
+
"grad_norm": 0.05869961529970169,
|
| 1800 |
+
"learning_rate": 1.3282986908296713e-06,
|
| 1801 |
+
"loss": 0.003,
|
| 1802 |
+
"num_input_tokens_seen": 281872,
|
| 1803 |
+
"step": 1025
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 9.079646017699115,
|
| 1807 |
+
"eval_loss": 0.3063960075378418,
|
| 1808 |
+
"eval_runtime": 0.7061,
|
| 1809 |
+
"eval_samples_per_second": 35.404,
|
| 1810 |
+
"eval_steps_per_second": 18.41,
|
| 1811 |
+
"num_input_tokens_seen": 282256,
|
| 1812 |
+
"step": 1026
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 9.11504424778761,
|
| 1816 |
+
"grad_norm": 0.13193757832050323,
|
| 1817 |
+
"learning_rate": 1.2069375825130292e-06,
|
| 1818 |
+
"loss": 0.0052,
|
| 1819 |
+
"num_input_tokens_seen": 283472,
|
| 1820 |
+
"step": 1030
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 9.15929203539823,
|
| 1824 |
+
"grad_norm": 0.051749859005212784,
|
| 1825 |
+
"learning_rate": 1.091252437230003e-06,
|
| 1826 |
+
"loss": 0.002,
|
| 1827 |
+
"num_input_tokens_seen": 284992,
|
| 1828 |
+
"step": 1035
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 9.20353982300885,
|
| 1832 |
+
"grad_norm": 0.025474902242422104,
|
| 1833 |
+
"learning_rate": 9.812708522943164e-07,
|
| 1834 |
+
"loss": 0.0017,
|
| 1835 |
+
"num_input_tokens_seen": 286464,
|
| 1836 |
+
"step": 1040
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 9.247787610619469,
|
| 1840 |
+
"grad_norm": 0.036191411316394806,
|
| 1841 |
+
"learning_rate": 8.770190644047077e-07,
|
| 1842 |
+
"loss": 0.0058,
|
| 1843 |
+
"num_input_tokens_seen": 287792,
|
| 1844 |
+
"step": 1045
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 9.29203539823009,
|
| 1848 |
+
"grad_norm": 0.053131312131881714,
|
| 1849 |
+
"learning_rate": 7.785219433859847e-07,
|
| 1850 |
+
"loss": 0.1187,
|
| 1851 |
+
"num_input_tokens_seen": 288960,
|
| 1852 |
+
"step": 1050
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 9.336283185840708,
|
| 1856 |
+
"grad_norm": 0.1995568573474884,
|
| 1857 |
+
"learning_rate": 6.858029862562343e-07,
|
| 1858 |
+
"loss": 0.0019,
|
| 1859 |
+
"num_input_tokens_seen": 290336,
|
| 1860 |
+
"step": 1055
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 9.380530973451327,
|
| 1864 |
+
"grad_norm": 0.17870356142520905,
|
| 1865 |
+
"learning_rate": 5.988843116214616e-07,
|
| 1866 |
+
"loss": 0.0015,
|
| 1867 |
+
"num_input_tokens_seen": 291728,
|
| 1868 |
+
"step": 1060
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 9.424778761061948,
|
| 1872 |
+
"grad_norm": 0.6506099700927734,
|
| 1873 |
+
"learning_rate": 5.177866543990689e-07,
|
| 1874 |
+
"loss": 0.0149,
|
| 1875 |
+
"num_input_tokens_seen": 292944,
|
| 1876 |
+
"step": 1065
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 9.469026548672566,
|
| 1880 |
+
"grad_norm": 0.015886101871728897,
|
| 1881 |
+
"learning_rate": 4.425293608714548e-07,
|
| 1882 |
+
"loss": 0.0004,
|
| 1883 |
+
"num_input_tokens_seen": 294448,
|
| 1884 |
+
"step": 1070
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 9.513274336283185,
|
| 1888 |
+
"grad_norm": 0.15230748057365417,
|
| 1889 |
+
"learning_rate": 3.7313038407083003e-07,
|
| 1890 |
+
"loss": 0.0722,
|
| 1891 |
+
"num_input_tokens_seen": 295744,
|
| 1892 |
+
"step": 1075
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 9.557522123893806,
|
| 1896 |
+
"grad_norm": 0.10186933726072311,
|
| 1897 |
+
"learning_rate": 3.0960627949644107e-07,
|
| 1898 |
+
"loss": 0.1136,
|
| 1899 |
+
"num_input_tokens_seen": 297280,
|
| 1900 |
+
"step": 1080
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 9.584070796460177,
|
| 1904 |
+
"eval_loss": 0.322214275598526,
|
| 1905 |
+
"eval_runtime": 0.7055,
|
| 1906 |
+
"eval_samples_per_second": 35.436,
|
| 1907 |
+
"eval_steps_per_second": 18.427,
|
| 1908 |
+
"num_input_tokens_seen": 297968,
|
| 1909 |
+
"step": 1083
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 9.601769911504425,
|
| 1913 |
+
"grad_norm": 0.13636958599090576,
|
| 1914 |
+
"learning_rate": 2.5197220116515987e-07,
|
| 1915 |
+
"loss": 0.0127,
|
| 1916 |
+
"num_input_tokens_seen": 298624,
|
| 1917 |
+
"step": 1085
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 9.646017699115044,
|
| 1921 |
+
"grad_norm": 0.14171549677848816,
|
| 1922 |
+
"learning_rate": 2.0024189799641435e-07,
|
| 1923 |
+
"loss": 0.0334,
|
| 1924 |
+
"num_input_tokens_seen": 299712,
|
| 1925 |
+
"step": 1090
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 9.690265486725664,
|
| 1929 |
+
"grad_norm": 0.017514703795313835,
|
| 1930 |
+
"learning_rate": 1.5442771053230665e-07,
|
| 1931 |
+
"loss": 0.0054,
|
| 1932 |
+
"num_input_tokens_seen": 301152,
|
| 1933 |
+
"step": 1095
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 9.734513274336283,
|
| 1937 |
+
"grad_norm": 0.021520795300602913,
|
| 1938 |
+
"learning_rate": 1.1454056799370672e-07,
|
| 1939 |
+
"loss": 0.0009,
|
| 1940 |
+
"num_input_tokens_seen": 302768,
|
| 1941 |
+
"step": 1100
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 9.778761061946902,
|
| 1945 |
+
"grad_norm": 0.09183833003044128,
|
| 1946 |
+
"learning_rate": 8.058998567304633e-08,
|
| 1947 |
+
"loss": 0.0201,
|
| 1948 |
+
"num_input_tokens_seen": 304176,
|
| 1949 |
+
"step": 1105
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 9.823008849557523,
|
| 1953 |
+
"grad_norm": 0.08951763063669205,
|
| 1954 |
+
"learning_rate": 5.2584062664368105e-08,
|
| 1955 |
+
"loss": 0.0018,
|
| 1956 |
+
"num_input_tokens_seen": 305328,
|
| 1957 |
+
"step": 1110
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 9.867256637168142,
|
| 1961 |
+
"grad_norm": 0.404524564743042,
|
| 1962 |
+
"learning_rate": 3.0529479931265535e-08,
|
| 1963 |
+
"loss": 0.0009,
|
| 1964 |
+
"num_input_tokens_seen": 306672,
|
| 1965 |
+
"step": 1115
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 9.91150442477876,
|
| 1969 |
+
"grad_norm": 11.300862312316895,
|
| 1970 |
+
"learning_rate": 1.443149871309113e-08,
|
| 1971 |
+
"loss": 0.0429,
|
| 1972 |
+
"num_input_tokens_seen": 307920,
|
| 1973 |
+
"step": 1120
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 9.955752212389381,
|
| 1977 |
+
"grad_norm": 0.10828626900911331,
|
| 1978 |
+
"learning_rate": 4.293959269863201e-09,
|
| 1979 |
+
"loss": 0.0011,
|
| 1980 |
+
"num_input_tokens_seen": 309424,
|
| 1981 |
+
"step": 1125
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 10.0,
|
| 1985 |
+
"grad_norm": 0.08019658923149109,
|
| 1986 |
+
"learning_rate": 1.192799661514643e-10,
|
| 1987 |
+
"loss": 0.0185,
|
| 1988 |
+
"num_input_tokens_seen": 310888,
|
| 1989 |
+
"step": 1130
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 10.0,
|
| 1993 |
+
"num_input_tokens_seen": 310888,
|
| 1994 |
+
"step": 1130,
|
| 1995 |
+
"total_flos": 1.3999146128572416e+16,
|
| 1996 |
+
"train_loss": 0.27997444444191766,
|
| 1997 |
+
"train_runtime": 152.8701,
|
| 1998 |
+
"train_samples_per_second": 14.718,
|
| 1999 |
+
"train_steps_per_second": 7.392
|
| 2000 |
+
}
|
| 2001 |
+
],
|
| 2002 |
+
"logging_steps": 5,
|
| 2003 |
+
"max_steps": 1130,
|
| 2004 |
+
"num_input_tokens_seen": 310888,
|
| 2005 |
+
"num_train_epochs": 10,
|
| 2006 |
+
"save_steps": 57,
|
| 2007 |
+
"stateful_callbacks": {
|
| 2008 |
+
"TrainerControl": {
|
| 2009 |
+
"args": {
|
| 2010 |
+
"should_epoch_stop": false,
|
| 2011 |
+
"should_evaluate": false,
|
| 2012 |
+
"should_log": false,
|
| 2013 |
+
"should_save": true,
|
| 2014 |
+
"should_training_stop": true
|
| 2015 |
+
},
|
| 2016 |
+
"attributes": {}
|
| 2017 |
+
}
|
| 2018 |
+
},
|
| 2019 |
+
"total_flos": 1.3999146128572416e+16,
|
| 2020 |
+
"train_batch_size": 2,
|
| 2021 |
+
"trial_name": null,
|
| 2022 |
+
"trial_params": null
|
| 2023 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|