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 +2215 -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_wsc_1754652157
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# train_wsc_1754652157
|
| 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_wsc_1754652157
|
|
|
|
| 16 |
|
| 17 |
# train_wsc_1754652157
|
| 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 wsc dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.3849
|
| 22 |
+
- Num Input Tokens Seen: 490000
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.38490504026412964,
|
| 4 |
+
"eval_runtime": 0.9049,
|
| 5 |
+
"eval_samples_per_second": 61.888,
|
| 6 |
+
"eval_steps_per_second": 15.472,
|
| 7 |
+
"num_input_tokens_seen": 490000,
|
| 8 |
+
"total_flos": 2.206447853568e+16,
|
| 9 |
+
"train_loss": 2.282332957649231,
|
| 10 |
+
"train_runtime": 172.1756,
|
| 11 |
+
"train_samples_per_second": 28.924,
|
| 12 |
+
"train_steps_per_second": 7.26
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.38490504026412964,
|
| 4 |
+
"eval_runtime": 0.9049,
|
| 5 |
+
"eval_samples_per_second": 61.888,
|
| 6 |
+
"eval_steps_per_second": 15.472,
|
| 7 |
+
"num_input_tokens_seen": 490000
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"num_input_tokens_seen": 490000,
|
| 4 |
+
"total_flos": 2.206447853568e+16,
|
| 5 |
+
"train_loss": 2.282332957649231,
|
| 6 |
+
"train_runtime": 172.1756,
|
| 7 |
+
"train_samples_per_second": 28.924,
|
| 8 |
+
"train_steps_per_second": 7.26
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2215 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1008,
|
| 3 |
+
"best_metric": 0.38490504026412964,
|
| 4 |
+
"best_model_checkpoint": "saves/prefix-tuning/llama-3-8b-instruct/train_wsc_1754652157/checkpoint-1008",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 63,
|
| 7 |
+
"global_step": 1250,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.04,
|
| 14 |
+
"grad_norm": 2.4841861724853516,
|
| 15 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 16 |
+
"loss": 15.2081,
|
| 17 |
+
"num_input_tokens_seen": 2144,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.08,
|
| 22 |
+
"grad_norm": 2.424849510192871,
|
| 23 |
+
"learning_rate": 3.6e-06,
|
| 24 |
+
"loss": 15.3148,
|
| 25 |
+
"num_input_tokens_seen": 4128,
|
| 26 |
+
"step": 10
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.12,
|
| 30 |
+
"grad_norm": 2.6069109439849854,
|
| 31 |
+
"learning_rate": 5.600000000000001e-06,
|
| 32 |
+
"loss": 15.3875,
|
| 33 |
+
"num_input_tokens_seen": 6240,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.16,
|
| 38 |
+
"grad_norm": 2.666384696960449,
|
| 39 |
+
"learning_rate": 7.6e-06,
|
| 40 |
+
"loss": 15.1604,
|
| 41 |
+
"num_input_tokens_seen": 8096,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.2,
|
| 46 |
+
"grad_norm": 2.383948564529419,
|
| 47 |
+
"learning_rate": 9.600000000000001e-06,
|
| 48 |
+
"loss": 15.358,
|
| 49 |
+
"num_input_tokens_seen": 10112,
|
| 50 |
+
"step": 25
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.24,
|
| 54 |
+
"grad_norm": 2.555549383163452,
|
| 55 |
+
"learning_rate": 1.16e-05,
|
| 56 |
+
"loss": 14.8947,
|
| 57 |
+
"num_input_tokens_seen": 12032,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.28,
|
| 62 |
+
"grad_norm": 2.3365485668182373,
|
| 63 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 64 |
+
"loss": 14.9376,
|
| 65 |
+
"num_input_tokens_seen": 13824,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.32,
|
| 70 |
+
"grad_norm": 2.37636137008667,
|
| 71 |
+
"learning_rate": 1.56e-05,
|
| 72 |
+
"loss": 15.0763,
|
| 73 |
+
"num_input_tokens_seen": 15840,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.36,
|
| 78 |
+
"grad_norm": 2.863443374633789,
|
| 79 |
+
"learning_rate": 1.76e-05,
|
| 80 |
+
"loss": 14.7281,
|
| 81 |
+
"num_input_tokens_seen": 17920,
|
| 82 |
+
"step": 45
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.4,
|
| 86 |
+
"grad_norm": 2.697338819503784,
|
| 87 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 88 |
+
"loss": 14.723,
|
| 89 |
+
"num_input_tokens_seen": 19712,
|
| 90 |
+
"step": 50
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.44,
|
| 94 |
+
"grad_norm": 2.416774272918701,
|
| 95 |
+
"learning_rate": 2.16e-05,
|
| 96 |
+
"loss": 14.5953,
|
| 97 |
+
"num_input_tokens_seen": 21952,
|
| 98 |
+
"step": 55
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.48,
|
| 102 |
+
"grad_norm": 2.4722342491149902,
|
| 103 |
+
"learning_rate": 2.36e-05,
|
| 104 |
+
"loss": 13.9838,
|
| 105 |
+
"num_input_tokens_seen": 24160,
|
| 106 |
+
"step": 60
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.504,
|
| 110 |
+
"eval_loss": 13.871087074279785,
|
| 111 |
+
"eval_runtime": 0.8458,
|
| 112 |
+
"eval_samples_per_second": 66.213,
|
| 113 |
+
"eval_steps_per_second": 16.553,
|
| 114 |
+
"num_input_tokens_seen": 25504,
|
| 115 |
+
"step": 63
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.52,
|
| 119 |
+
"grad_norm": 2.531996250152588,
|
| 120 |
+
"learning_rate": 2.5600000000000002e-05,
|
| 121 |
+
"loss": 13.8098,
|
| 122 |
+
"num_input_tokens_seen": 26112,
|
| 123 |
+
"step": 65
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.56,
|
| 127 |
+
"grad_norm": 2.4052157402038574,
|
| 128 |
+
"learning_rate": 2.7600000000000003e-05,
|
| 129 |
+
"loss": 13.7342,
|
| 130 |
+
"num_input_tokens_seen": 28064,
|
| 131 |
+
"step": 70
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.6,
|
| 135 |
+
"grad_norm": 2.380267381668091,
|
| 136 |
+
"learning_rate": 2.96e-05,
|
| 137 |
+
"loss": 13.545,
|
| 138 |
+
"num_input_tokens_seen": 29824,
|
| 139 |
+
"step": 75
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.64,
|
| 143 |
+
"grad_norm": 2.2633209228515625,
|
| 144 |
+
"learning_rate": 3.16e-05,
|
| 145 |
+
"loss": 13.168,
|
| 146 |
+
"num_input_tokens_seen": 31904,
|
| 147 |
+
"step": 80
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.68,
|
| 151 |
+
"grad_norm": 2.5283195972442627,
|
| 152 |
+
"learning_rate": 3.3600000000000004e-05,
|
| 153 |
+
"loss": 13.0662,
|
| 154 |
+
"num_input_tokens_seen": 33984,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.72,
|
| 159 |
+
"grad_norm": 2.547339916229248,
|
| 160 |
+
"learning_rate": 3.56e-05,
|
| 161 |
+
"loss": 12.3617,
|
| 162 |
+
"num_input_tokens_seen": 35776,
|
| 163 |
+
"step": 90
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.76,
|
| 167 |
+
"grad_norm": 2.478236675262451,
|
| 168 |
+
"learning_rate": 3.76e-05,
|
| 169 |
+
"loss": 12.2163,
|
| 170 |
+
"num_input_tokens_seen": 37472,
|
| 171 |
+
"step": 95
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.8,
|
| 175 |
+
"grad_norm": 2.3510525226593018,
|
| 176 |
+
"learning_rate": 3.960000000000001e-05,
|
| 177 |
+
"loss": 12.0549,
|
| 178 |
+
"num_input_tokens_seen": 39328,
|
| 179 |
+
"step": 100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.84,
|
| 183 |
+
"grad_norm": 2.5149877071380615,
|
| 184 |
+
"learning_rate": 4.16e-05,
|
| 185 |
+
"loss": 11.9157,
|
| 186 |
+
"num_input_tokens_seen": 41280,
|
| 187 |
+
"step": 105
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.88,
|
| 191 |
+
"grad_norm": 2.451765537261963,
|
| 192 |
+
"learning_rate": 4.36e-05,
|
| 193 |
+
"loss": 11.5939,
|
| 194 |
+
"num_input_tokens_seen": 43552,
|
| 195 |
+
"step": 110
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.92,
|
| 199 |
+
"grad_norm": 2.4347667694091797,
|
| 200 |
+
"learning_rate": 4.5600000000000004e-05,
|
| 201 |
+
"loss": 10.6461,
|
| 202 |
+
"num_input_tokens_seen": 45216,
|
| 203 |
+
"step": 115
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.96,
|
| 207 |
+
"grad_norm": 2.257194995880127,
|
| 208 |
+
"learning_rate": 4.76e-05,
|
| 209 |
+
"loss": 10.384,
|
| 210 |
+
"num_input_tokens_seen": 47360,
|
| 211 |
+
"step": 120
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.0,
|
| 215 |
+
"grad_norm": 2.9155919551849365,
|
| 216 |
+
"learning_rate": 4.96e-05,
|
| 217 |
+
"loss": 9.9251,
|
| 218 |
+
"num_input_tokens_seen": 49376,
|
| 219 |
+
"step": 125
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.008,
|
| 223 |
+
"eval_loss": 9.737223625183105,
|
| 224 |
+
"eval_runtime": 0.8536,
|
| 225 |
+
"eval_samples_per_second": 65.608,
|
| 226 |
+
"eval_steps_per_second": 16.402,
|
| 227 |
+
"num_input_tokens_seen": 49696,
|
| 228 |
+
"step": 126
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.04,
|
| 232 |
+
"grad_norm": 2.394456386566162,
|
| 233 |
+
"learning_rate": 4.9998440375027166e-05,
|
| 234 |
+
"loss": 9.5781,
|
| 235 |
+
"num_input_tokens_seen": 51200,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.08,
|
| 240 |
+
"grad_norm": 2.3213348388671875,
|
| 241 |
+
"learning_rate": 4.99921047320825e-05,
|
| 242 |
+
"loss": 8.9157,
|
| 243 |
+
"num_input_tokens_seen": 53216,
|
| 244 |
+
"step": 135
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.12,
|
| 248 |
+
"grad_norm": 2.2454257011413574,
|
| 249 |
+
"learning_rate": 4.998089682880117e-05,
|
| 250 |
+
"loss": 8.0389,
|
| 251 |
+
"num_input_tokens_seen": 55168,
|
| 252 |
+
"step": 140
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.16,
|
| 256 |
+
"grad_norm": 2.323469400405884,
|
| 257 |
+
"learning_rate": 4.9964818850186135e-05,
|
| 258 |
+
"loss": 8.6052,
|
| 259 |
+
"num_input_tokens_seen": 56960,
|
| 260 |
+
"step": 145
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.2,
|
| 264 |
+
"grad_norm": 2.213332414627075,
|
| 265 |
+
"learning_rate": 4.994387393067117e-05,
|
| 266 |
+
"loss": 7.8179,
|
| 267 |
+
"num_input_tokens_seen": 58880,
|
| 268 |
+
"step": 150
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.24,
|
| 272 |
+
"grad_norm": 1.9772940874099731,
|
| 273 |
+
"learning_rate": 4.9918066153509834e-05,
|
| 274 |
+
"loss": 7.3245,
|
| 275 |
+
"num_input_tokens_seen": 60672,
|
| 276 |
+
"step": 155
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.28,
|
| 280 |
+
"grad_norm": 2.1031250953674316,
|
| 281 |
+
"learning_rate": 4.988740054997943e-05,
|
| 282 |
+
"loss": 7.6592,
|
| 283 |
+
"num_input_tokens_seen": 62848,
|
| 284 |
+
"step": 160
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.32,
|
| 288 |
+
"grad_norm": 1.5588639974594116,
|
| 289 |
+
"learning_rate": 4.985188309840012e-05,
|
| 290 |
+
"loss": 6.5144,
|
| 291 |
+
"num_input_tokens_seen": 64448,
|
| 292 |
+
"step": 165
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.3599999999999999,
|
| 296 |
+
"grad_norm": 1.6640610694885254,
|
| 297 |
+
"learning_rate": 4.9811520722969465e-05,
|
| 298 |
+
"loss": 6.9579,
|
| 299 |
+
"num_input_tokens_seen": 66368,
|
| 300 |
+
"step": 170
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.4,
|
| 304 |
+
"grad_norm": 1.4669989347457886,
|
| 305 |
+
"learning_rate": 4.976632129241252e-05,
|
| 306 |
+
"loss": 5.7934,
|
| 307 |
+
"num_input_tokens_seen": 68128,
|
| 308 |
+
"step": 175
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.44,
|
| 312 |
+
"grad_norm": 1.5640122890472412,
|
| 313 |
+
"learning_rate": 4.971629361844785e-05,
|
| 314 |
+
"loss": 6.0098,
|
| 315 |
+
"num_input_tokens_seen": 70112,
|
| 316 |
+
"step": 180
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.48,
|
| 320 |
+
"grad_norm": 1.9767147302627563,
|
| 321 |
+
"learning_rate": 4.966144745406961e-05,
|
| 322 |
+
"loss": 6.0466,
|
| 323 |
+
"num_input_tokens_seen": 72384,
|
| 324 |
+
"step": 185
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.512,
|
| 328 |
+
"eval_loss": 5.308993816375732,
|
| 329 |
+
"eval_runtime": 0.8536,
|
| 330 |
+
"eval_samples_per_second": 65.601,
|
| 331 |
+
"eval_steps_per_second": 16.4,
|
| 332 |
+
"num_input_tokens_seen": 74112,
|
| 333 |
+
"step": 189
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.52,
|
| 337 |
+
"grad_norm": 1.4375022649765015,
|
| 338 |
+
"learning_rate": 4.960179349164621e-05,
|
| 339 |
+
"loss": 5.7175,
|
| 340 |
+
"num_input_tokens_seen": 74752,
|
| 341 |
+
"step": 190
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.56,
|
| 345 |
+
"grad_norm": 1.8162568807601929,
|
| 346 |
+
"learning_rate": 4.953734336083583e-05,
|
| 347 |
+
"loss": 5.0706,
|
| 348 |
+
"num_input_tokens_seen": 76640,
|
| 349 |
+
"step": 195
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.6,
|
| 353 |
+
"grad_norm": 2.0760273933410645,
|
| 354 |
+
"learning_rate": 4.946810962631916e-05,
|
| 355 |
+
"loss": 5.3219,
|
| 356 |
+
"num_input_tokens_seen": 78784,
|
| 357 |
+
"step": 200
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.6400000000000001,
|
| 361 |
+
"grad_norm": 1.7913469076156616,
|
| 362 |
+
"learning_rate": 4.9394105785349944e-05,
|
| 363 |
+
"loss": 4.6854,
|
| 364 |
+
"num_input_tokens_seen": 80768,
|
| 365 |
+
"step": 205
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.6800000000000002,
|
| 369 |
+
"grad_norm": 1.471130609512329,
|
| 370 |
+
"learning_rate": 4.9315346265123594e-05,
|
| 371 |
+
"loss": 4.0843,
|
| 372 |
+
"num_input_tokens_seen": 82848,
|
| 373 |
+
"step": 210
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.72,
|
| 377 |
+
"grad_norm": 1.8904441595077515,
|
| 378 |
+
"learning_rate": 4.923184641996463e-05,
|
| 379 |
+
"loss": 4.0373,
|
| 380 |
+
"num_input_tokens_seen": 84768,
|
| 381 |
+
"step": 215
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.76,
|
| 385 |
+
"grad_norm": 2.13264536857605,
|
| 386 |
+
"learning_rate": 4.914362252833332e-05,
|
| 387 |
+
"loss": 3.6419,
|
| 388 |
+
"num_input_tokens_seen": 86848,
|
| 389 |
+
"step": 220
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.8,
|
| 393 |
+
"grad_norm": 1.7601622343063354,
|
| 394 |
+
"learning_rate": 4.905069178965215e-05,
|
| 395 |
+
"loss": 3.3145,
|
| 396 |
+
"num_input_tokens_seen": 88736,
|
| 397 |
+
"step": 225
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 1.8399999999999999,
|
| 401 |
+
"grad_norm": 1.8561395406723022,
|
| 402 |
+
"learning_rate": 4.8953072320952745e-05,
|
| 403 |
+
"loss": 3.4327,
|
| 404 |
+
"num_input_tokens_seen": 90848,
|
| 405 |
+
"step": 230
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.88,
|
| 409 |
+
"grad_norm": 1.9273947477340698,
|
| 410 |
+
"learning_rate": 4.885078315334395e-05,
|
| 411 |
+
"loss": 3.2526,
|
| 412 |
+
"num_input_tokens_seen": 92992,
|
| 413 |
+
"step": 235
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.92,
|
| 417 |
+
"grad_norm": 2.0955724716186523,
|
| 418 |
+
"learning_rate": 4.874384422830167e-05,
|
| 419 |
+
"loss": 2.791,
|
| 420 |
+
"num_input_tokens_seen": 94880,
|
| 421 |
+
"step": 240
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.96,
|
| 425 |
+
"grad_norm": 1.2840852737426758,
|
| 426 |
+
"learning_rate": 4.863227639378124e-05,
|
| 427 |
+
"loss": 2.2217,
|
| 428 |
+
"num_input_tokens_seen": 96704,
|
| 429 |
+
"step": 245
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.0,
|
| 433 |
+
"grad_norm": 1.7399126291275024,
|
| 434 |
+
"learning_rate": 4.851610140015304e-05,
|
| 435 |
+
"loss": 1.6788,
|
| 436 |
+
"num_input_tokens_seen": 98240,
|
| 437 |
+
"step": 250
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 2.016,
|
| 441 |
+
"eval_loss": 1.980229139328003,
|
| 442 |
+
"eval_runtime": 0.8532,
|
| 443 |
+
"eval_samples_per_second": 65.632,
|
| 444 |
+
"eval_steps_per_second": 16.408,
|
| 445 |
+
"num_input_tokens_seen": 99136,
|
| 446 |
+
"step": 252
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.04,
|
| 450 |
+
"grad_norm": 1.2386776208877563,
|
| 451 |
+
"learning_rate": 4.839534189596228e-05,
|
| 452 |
+
"loss": 2.0365,
|
| 453 |
+
"num_input_tokens_seen": 100224,
|
| 454 |
+
"step": 255
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 2.08,
|
| 458 |
+
"grad_norm": 1.1297599077224731,
|
| 459 |
+
"learning_rate": 4.8270021423513554e-05,
|
| 460 |
+
"loss": 1.4581,
|
| 461 |
+
"num_input_tokens_seen": 101920,
|
| 462 |
+
"step": 260
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.12,
|
| 466 |
+
"grad_norm": 1.090529203414917,
|
| 467 |
+
"learning_rate": 4.8140164414281306e-05,
|
| 468 |
+
"loss": 1.2941,
|
| 469 |
+
"num_input_tokens_seen": 103808,
|
| 470 |
+
"step": 265
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 2.16,
|
| 474 |
+
"grad_norm": 1.0772465467453003,
|
| 475 |
+
"learning_rate": 4.800579618414676e-05,
|
| 476 |
+
"loss": 1.4016,
|
| 477 |
+
"num_input_tokens_seen": 105920,
|
| 478 |
+
"step": 270
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 2.2,
|
| 482 |
+
"grad_norm": 1.179438591003418,
|
| 483 |
+
"learning_rate": 4.7866942928462625e-05,
|
| 484 |
+
"loss": 1.5631,
|
| 485 |
+
"num_input_tokens_seen": 108160,
|
| 486 |
+
"step": 275
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 2.24,
|
| 490 |
+
"grad_norm": 1.1663882732391357,
|
| 491 |
+
"learning_rate": 4.772363171694622e-05,
|
| 492 |
+
"loss": 0.856,
|
| 493 |
+
"num_input_tokens_seen": 109920,
|
| 494 |
+
"step": 280
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.2800000000000002,
|
| 498 |
+
"grad_norm": 1.3647843599319458,
|
| 499 |
+
"learning_rate": 4.7575890488402185e-05,
|
| 500 |
+
"loss": 1.1216,
|
| 501 |
+
"num_input_tokens_seen": 111904,
|
| 502 |
+
"step": 285
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 2.32,
|
| 506 |
+
"grad_norm": 1.4680769443511963,
|
| 507 |
+
"learning_rate": 4.742374804527575e-05,
|
| 508 |
+
"loss": 0.776,
|
| 509 |
+
"num_input_tokens_seen": 113632,
|
| 510 |
+
"step": 290
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.36,
|
| 514 |
+
"grad_norm": 0.9575800895690918,
|
| 515 |
+
"learning_rate": 4.7267234048037664e-05,
|
| 516 |
+
"loss": 1.0421,
|
| 517 |
+
"num_input_tokens_seen": 115616,
|
| 518 |
+
"step": 295
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.4,
|
| 522 |
+
"grad_norm": 0.7263527512550354,
|
| 523 |
+
"learning_rate": 4.710637900940181e-05,
|
| 524 |
+
"loss": 0.7994,
|
| 525 |
+
"num_input_tokens_seen": 117472,
|
| 526 |
+
"step": 300
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.44,
|
| 530 |
+
"grad_norm": 1.1039828062057495,
|
| 531 |
+
"learning_rate": 4.694121428837668e-05,
|
| 532 |
+
"loss": 1.0609,
|
| 533 |
+
"num_input_tokens_seen": 119616,
|
| 534 |
+
"step": 305
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.48,
|
| 538 |
+
"grad_norm": 0.9919302463531494,
|
| 539 |
+
"learning_rate": 4.6771772084151885e-05,
|
| 540 |
+
"loss": 0.8039,
|
| 541 |
+
"num_input_tokens_seen": 121568,
|
| 542 |
+
"step": 310
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.52,
|
| 546 |
+
"grad_norm": 1.072923183441162,
|
| 547 |
+
"learning_rate": 4.659808542982088e-05,
|
| 548 |
+
"loss": 1.0818,
|
| 549 |
+
"num_input_tokens_seen": 123904,
|
| 550 |
+
"step": 315
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.52,
|
| 554 |
+
"eval_loss": 0.7967647314071655,
|
| 555 |
+
"eval_runtime": 0.8522,
|
| 556 |
+
"eval_samples_per_second": 65.715,
|
| 557 |
+
"eval_steps_per_second": 16.429,
|
| 558 |
+
"num_input_tokens_seen": 123904,
|
| 559 |
+
"step": 315
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.56,
|
| 563 |
+
"grad_norm": 0.6072083711624146,
|
| 564 |
+
"learning_rate": 4.642018818594107e-05,
|
| 565 |
+
"loss": 0.5327,
|
| 566 |
+
"num_input_tokens_seen": 125696,
|
| 567 |
+
"step": 320
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.6,
|
| 571 |
+
"grad_norm": 0.9770075678825378,
|
| 572 |
+
"learning_rate": 4.6238115033932636e-05,
|
| 573 |
+
"loss": 0.6088,
|
| 574 |
+
"num_input_tokens_seen": 127488,
|
| 575 |
+
"step": 325
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.64,
|
| 579 |
+
"grad_norm": 1.5782830715179443,
|
| 580 |
+
"learning_rate": 4.605190146931731e-05,
|
| 581 |
+
"loss": 0.7891,
|
| 582 |
+
"num_input_tokens_seen": 129632,
|
| 583 |
+
"step": 330
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.68,
|
| 587 |
+
"grad_norm": 1.686546802520752,
|
| 588 |
+
"learning_rate": 4.586158379479848e-05,
|
| 589 |
+
"loss": 0.7432,
|
| 590 |
+
"num_input_tokens_seen": 131680,
|
| 591 |
+
"step": 335
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.7199999999999998,
|
| 595 |
+
"grad_norm": 0.8748849630355835,
|
| 596 |
+
"learning_rate": 4.566719911318389e-05,
|
| 597 |
+
"loss": 0.5091,
|
| 598 |
+
"num_input_tokens_seen": 133472,
|
| 599 |
+
"step": 340
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.76,
|
| 603 |
+
"grad_norm": 0.690437912940979,
|
| 604 |
+
"learning_rate": 4.5468785320152365e-05,
|
| 605 |
+
"loss": 0.5338,
|
| 606 |
+
"num_input_tokens_seen": 135200,
|
| 607 |
+
"step": 345
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 2.8,
|
| 611 |
+
"grad_norm": 1.1224329471588135,
|
| 612 |
+
"learning_rate": 4.5266381096866e-05,
|
| 613 |
+
"loss": 0.9588,
|
| 614 |
+
"num_input_tokens_seen": 137536,
|
| 615 |
+
"step": 350
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.84,
|
| 619 |
+
"grad_norm": 0.998359203338623,
|
| 620 |
+
"learning_rate": 4.5060025902429174e-05,
|
| 621 |
+
"loss": 0.7235,
|
| 622 |
+
"num_input_tokens_seen": 139744,
|
| 623 |
+
"step": 355
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 2.88,
|
| 627 |
+
"grad_norm": 1.1624242067337036,
|
| 628 |
+
"learning_rate": 4.484975996619589e-05,
|
| 629 |
+
"loss": 0.703,
|
| 630 |
+
"num_input_tokens_seen": 141760,
|
| 631 |
+
"step": 360
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.92,
|
| 635 |
+
"grad_norm": 1.1226049661636353,
|
| 636 |
+
"learning_rate": 4.4635624279927044e-05,
|
| 637 |
+
"loss": 0.6324,
|
| 638 |
+
"num_input_tokens_seen": 143872,
|
| 639 |
+
"step": 365
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.96,
|
| 643 |
+
"grad_norm": 0.6500603556632996,
|
| 644 |
+
"learning_rate": 4.441766058979898e-05,
|
| 645 |
+
"loss": 0.5377,
|
| 646 |
+
"num_input_tokens_seen": 145856,
|
| 647 |
+
"step": 370
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 3.0,
|
| 651 |
+
"grad_norm": 1.9589425325393677,
|
| 652 |
+
"learning_rate": 4.4195911388264946e-05,
|
| 653 |
+
"loss": 0.8394,
|
| 654 |
+
"num_input_tokens_seen": 147648,
|
| 655 |
+
"step": 375
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 3.024,
|
| 659 |
+
"eval_loss": 0.5601035952568054,
|
| 660 |
+
"eval_runtime": 0.8544,
|
| 661 |
+
"eval_samples_per_second": 65.545,
|
| 662 |
+
"eval_steps_per_second": 16.386,
|
| 663 |
+
"num_input_tokens_seen": 148736,
|
| 664 |
+
"step": 378
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 3.04,
|
| 668 |
+
"grad_norm": 1.4394139051437378,
|
| 669 |
+
"learning_rate": 4.3970419905771145e-05,
|
| 670 |
+
"loss": 0.4311,
|
| 671 |
+
"num_input_tokens_seen": 149472,
|
| 672 |
+
"step": 380
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 3.08,
|
| 676 |
+
"grad_norm": 0.8933354020118713,
|
| 677 |
+
"learning_rate": 4.374123010232888e-05,
|
| 678 |
+
"loss": 0.6238,
|
| 679 |
+
"num_input_tokens_seen": 151552,
|
| 680 |
+
"step": 385
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 3.12,
|
| 684 |
+
"grad_norm": 0.8065023422241211,
|
| 685 |
+
"learning_rate": 4.350838665894446e-05,
|
| 686 |
+
"loss": 0.5635,
|
| 687 |
+
"num_input_tokens_seen": 153568,
|
| 688 |
+
"step": 390
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 3.16,
|
| 692 |
+
"grad_norm": 0.6302071213722229,
|
| 693 |
+
"learning_rate": 4.3271934968908514e-05,
|
| 694 |
+
"loss": 0.4754,
|
| 695 |
+
"num_input_tokens_seen": 155616,
|
| 696 |
+
"step": 395
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 3.2,
|
| 700 |
+
"grad_norm": 0.5202668309211731,
|
| 701 |
+
"learning_rate": 4.303192112894652e-05,
|
| 702 |
+
"loss": 0.6315,
|
| 703 |
+
"num_input_tokens_seen": 157728,
|
| 704 |
+
"step": 400
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.24,
|
| 708 |
+
"grad_norm": 0.9103642702102661,
|
| 709 |
+
"learning_rate": 4.278839193023214e-05,
|
| 710 |
+
"loss": 0.4821,
|
| 711 |
+
"num_input_tokens_seen": 159488,
|
| 712 |
+
"step": 405
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 3.2800000000000002,
|
| 716 |
+
"grad_norm": 0.6605204939842224,
|
| 717 |
+
"learning_rate": 4.254139484926519e-05,
|
| 718 |
+
"loss": 0.5577,
|
| 719 |
+
"num_input_tokens_seen": 161600,
|
| 720 |
+
"step": 410
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 3.32,
|
| 724 |
+
"grad_norm": 0.8729458451271057,
|
| 725 |
+
"learning_rate": 4.2290978038616e-05,
|
| 726 |
+
"loss": 0.5065,
|
| 727 |
+
"num_input_tokens_seen": 163712,
|
| 728 |
+
"step": 415
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 3.36,
|
| 732 |
+
"grad_norm": 0.7586544752120972,
|
| 733 |
+
"learning_rate": 4.2037190317538e-05,
|
| 734 |
+
"loss": 0.4285,
|
| 735 |
+
"num_input_tokens_seen": 165536,
|
| 736 |
+
"step": 420
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 3.4,
|
| 740 |
+
"grad_norm": 1.765929102897644,
|
| 741 |
+
"learning_rate": 4.178008116245024e-05,
|
| 742 |
+
"loss": 0.5842,
|
| 743 |
+
"num_input_tokens_seen": 167872,
|
| 744 |
+
"step": 425
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 3.44,
|
| 748 |
+
"grad_norm": 0.9758070111274719,
|
| 749 |
+
"learning_rate": 4.1519700697291944e-05,
|
| 750 |
+
"loss": 0.6883,
|
| 751 |
+
"num_input_tokens_seen": 170112,
|
| 752 |
+
"step": 430
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 3.48,
|
| 756 |
+
"grad_norm": 1.6534411907196045,
|
| 757 |
+
"learning_rate": 4.125609968375072e-05,
|
| 758 |
+
"loss": 0.4686,
|
| 759 |
+
"num_input_tokens_seen": 172000,
|
| 760 |
+
"step": 435
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 3.52,
|
| 764 |
+
"grad_norm": 0.6654028296470642,
|
| 765 |
+
"learning_rate": 4.098932951136645e-05,
|
| 766 |
+
"loss": 0.5184,
|
| 767 |
+
"num_input_tokens_seen": 174016,
|
| 768 |
+
"step": 440
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.528,
|
| 772 |
+
"eval_loss": 0.4781542420387268,
|
| 773 |
+
"eval_runtime": 0.8555,
|
| 774 |
+
"eval_samples_per_second": 65.46,
|
| 775 |
+
"eval_steps_per_second": 16.365,
|
| 776 |
+
"num_input_tokens_seen": 174432,
|
| 777 |
+
"step": 441
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 3.56,
|
| 781 |
+
"grad_norm": 0.7983071208000183,
|
| 782 |
+
"learning_rate": 4.071944218751282e-05,
|
| 783 |
+
"loss": 0.4466,
|
| 784 |
+
"num_input_tokens_seen": 175776,
|
| 785 |
+
"step": 445
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 3.6,
|
| 789 |
+
"grad_norm": 0.9768652319908142,
|
| 790 |
+
"learning_rate": 4.044649032725836e-05,
|
| 791 |
+
"loss": 0.5084,
|
| 792 |
+
"num_input_tokens_seen": 177952,
|
| 793 |
+
"step": 450
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 3.64,
|
| 797 |
+
"grad_norm": 1.1632089614868164,
|
| 798 |
+
"learning_rate": 4.017052714310906e-05,
|
| 799 |
+
"loss": 0.3983,
|
| 800 |
+
"num_input_tokens_seen": 179968,
|
| 801 |
+
"step": 455
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 3.68,
|
| 805 |
+
"grad_norm": 1.7327042818069458,
|
| 806 |
+
"learning_rate": 3.989160643463445e-05,
|
| 807 |
+
"loss": 0.4557,
|
| 808 |
+
"num_input_tokens_seen": 181952,
|
| 809 |
+
"step": 460
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 3.7199999999999998,
|
| 813 |
+
"grad_norm": 1.0546882152557373,
|
| 814 |
+
"learning_rate": 3.960978257797931e-05,
|
| 815 |
+
"loss": 0.3364,
|
| 816 |
+
"num_input_tokens_seen": 183680,
|
| 817 |
+
"step": 465
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 3.76,
|
| 821 |
+
"grad_norm": 0.5163419246673584,
|
| 822 |
+
"learning_rate": 3.932511051526289e-05,
|
| 823 |
+
"loss": 0.4238,
|
| 824 |
+
"num_input_tokens_seen": 185632,
|
| 825 |
+
"step": 470
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 3.8,
|
| 829 |
+
"grad_norm": 1.192272663116455,
|
| 830 |
+
"learning_rate": 3.903764574386786e-05,
|
| 831 |
+
"loss": 0.46,
|
| 832 |
+
"num_input_tokens_seen": 187552,
|
| 833 |
+
"step": 475
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.84,
|
| 837 |
+
"grad_norm": 1.180910348892212,
|
| 838 |
+
"learning_rate": 3.8747444305621e-05,
|
| 839 |
+
"loss": 0.4533,
|
| 840 |
+
"num_input_tokens_seen": 189408,
|
| 841 |
+
"step": 480
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 3.88,
|
| 845 |
+
"grad_norm": 0.7053818702697754,
|
| 846 |
+
"learning_rate": 3.8454562775867684e-05,
|
| 847 |
+
"loss": 0.4832,
|
| 848 |
+
"num_input_tokens_seen": 191488,
|
| 849 |
+
"step": 485
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.92,
|
| 853 |
+
"grad_norm": 1.434988260269165,
|
| 854 |
+
"learning_rate": 3.8159058252442446e-05,
|
| 855 |
+
"loss": 0.4162,
|
| 856 |
+
"num_input_tokens_seen": 193312,
|
| 857 |
+
"step": 490
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.96,
|
| 861 |
+
"grad_norm": 0.8368187546730042,
|
| 862 |
+
"learning_rate": 3.786098834453766e-05,
|
| 863 |
+
"loss": 0.4683,
|
| 864 |
+
"num_input_tokens_seen": 195424,
|
| 865 |
+
"step": 495
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 4.0,
|
| 869 |
+
"grad_norm": 1.403336763381958,
|
| 870 |
+
"learning_rate": 3.7560411161472456e-05,
|
| 871 |
+
"loss": 0.3853,
|
| 872 |
+
"num_input_tokens_seen": 197024,
|
| 873 |
+
"step": 500
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 4.032,
|
| 877 |
+
"eval_loss": 0.46134254336357117,
|
| 878 |
+
"eval_runtime": 0.8491,
|
| 879 |
+
"eval_samples_per_second": 65.955,
|
| 880 |
+
"eval_steps_per_second": 16.489,
|
| 881 |
+
"num_input_tokens_seen": 198656,
|
| 882 |
+
"step": 504
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 4.04,
|
| 886 |
+
"grad_norm": 0.6058657169342041,
|
| 887 |
+
"learning_rate": 3.725738530136422e-05,
|
| 888 |
+
"loss": 0.5412,
|
| 889 |
+
"num_input_tokens_seen": 199040,
|
| 890 |
+
"step": 505
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 4.08,
|
| 894 |
+
"grad_norm": 1.2126246690750122,
|
| 895 |
+
"learning_rate": 3.695196983970481e-05,
|
| 896 |
+
"loss": 0.4867,
|
| 897 |
+
"num_input_tokens_seen": 200960,
|
| 898 |
+
"step": 510
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 4.12,
|
| 902 |
+
"grad_norm": 0.5591365098953247,
|
| 903 |
+
"learning_rate": 3.664422431784361e-05,
|
| 904 |
+
"loss": 0.3728,
|
| 905 |
+
"num_input_tokens_seen": 203008,
|
| 906 |
+
"step": 515
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 4.16,
|
| 910 |
+
"grad_norm": 1.040564775466919,
|
| 911 |
+
"learning_rate": 3.633420873137988e-05,
|
| 912 |
+
"loss": 0.378,
|
| 913 |
+
"num_input_tokens_seen": 204672,
|
| 914 |
+
"step": 520
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 4.2,
|
| 918 |
+
"grad_norm": 0.4747677743434906,
|
| 919 |
+
"learning_rate": 3.602198351846647e-05,
|
| 920 |
+
"loss": 0.4497,
|
| 921 |
+
"num_input_tokens_seen": 206784,
|
| 922 |
+
"step": 525
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 4.24,
|
| 926 |
+
"grad_norm": 0.8027376532554626,
|
| 927 |
+
"learning_rate": 3.570760954802726e-05,
|
| 928 |
+
"loss": 0.3772,
|
| 929 |
+
"num_input_tokens_seen": 208672,
|
| 930 |
+
"step": 530
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 4.28,
|
| 934 |
+
"grad_norm": 0.7639246582984924,
|
| 935 |
+
"learning_rate": 3.53911481078907e-05,
|
| 936 |
+
"loss": 0.4375,
|
| 937 |
+
"num_input_tokens_seen": 210752,
|
| 938 |
+
"step": 535
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 4.32,
|
| 942 |
+
"grad_norm": 0.7363210320472717,
|
| 943 |
+
"learning_rate": 3.507266089284157e-05,
|
| 944 |
+
"loss": 0.5944,
|
| 945 |
+
"num_input_tokens_seen": 213472,
|
| 946 |
+
"step": 540
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 4.36,
|
| 950 |
+
"grad_norm": 1.636357307434082,
|
| 951 |
+
"learning_rate": 3.475220999259349e-05,
|
| 952 |
+
"loss": 0.4546,
|
| 953 |
+
"num_input_tokens_seen": 215616,
|
| 954 |
+
"step": 545
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 4.4,
|
| 958 |
+
"grad_norm": 0.7487366795539856,
|
| 959 |
+
"learning_rate": 3.442985787968442e-05,
|
| 960 |
+
"loss": 0.4207,
|
| 961 |
+
"num_input_tokens_seen": 217664,
|
| 962 |
+
"step": 550
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 4.44,
|
| 966 |
+
"grad_norm": 0.5726836323738098,
|
| 967 |
+
"learning_rate": 3.410566739729746e-05,
|
| 968 |
+
"loss": 0.4204,
|
| 969 |
+
"num_input_tokens_seen": 219584,
|
| 970 |
+
"step": 555
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 4.48,
|
| 974 |
+
"grad_norm": 0.7676336169242859,
|
| 975 |
+
"learning_rate": 3.3779701747009504e-05,
|
| 976 |
+
"loss": 0.4381,
|
| 977 |
+
"num_input_tokens_seen": 221504,
|
| 978 |
+
"step": 560
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 4.52,
|
| 982 |
+
"grad_norm": 0.8204706907272339,
|
| 983 |
+
"learning_rate": 3.3452024476469934e-05,
|
| 984 |
+
"loss": 0.4549,
|
| 985 |
+
"num_input_tokens_seen": 223424,
|
| 986 |
+
"step": 565
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 4.536,
|
| 990 |
+
"eval_loss": 0.43884095549583435,
|
| 991 |
+
"eval_runtime": 0.8563,
|
| 992 |
+
"eval_samples_per_second": 65.4,
|
| 993 |
+
"eval_steps_per_second": 16.35,
|
| 994 |
+
"num_input_tokens_seen": 224032,
|
| 995 |
+
"step": 567
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 4.5600000000000005,
|
| 999 |
+
"grad_norm": 1.2742042541503906,
|
| 1000 |
+
"learning_rate": 3.312269946701191e-05,
|
| 1001 |
+
"loss": 0.4388,
|
| 1002 |
+
"num_input_tokens_seen": 225216,
|
| 1003 |
+
"step": 570
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 4.6,
|
| 1007 |
+
"grad_norm": 1.1421749591827393,
|
| 1008 |
+
"learning_rate": 3.279179092119855e-05,
|
| 1009 |
+
"loss": 0.3681,
|
| 1010 |
+
"num_input_tokens_seen": 227008,
|
| 1011 |
+
"step": 575
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 4.64,
|
| 1015 |
+
"grad_norm": 1.359496831893921,
|
| 1016 |
+
"learning_rate": 3.245936335030651e-05,
|
| 1017 |
+
"loss": 0.4424,
|
| 1018 |
+
"num_input_tokens_seen": 228736,
|
| 1019 |
+
"step": 580
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 4.68,
|
| 1023 |
+
"grad_norm": 0.7598556280136108,
|
| 1024 |
+
"learning_rate": 3.21254815617494e-05,
|
| 1025 |
+
"loss": 0.3985,
|
| 1026 |
+
"num_input_tokens_seen": 230240,
|
| 1027 |
+
"step": 585
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 4.72,
|
| 1031 |
+
"grad_norm": 0.4738464057445526,
|
| 1032 |
+
"learning_rate": 3.179021064644347e-05,
|
| 1033 |
+
"loss": 0.428,
|
| 1034 |
+
"num_input_tokens_seen": 232192,
|
| 1035 |
+
"step": 590
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 4.76,
|
| 1039 |
+
"grad_norm": 0.8662070035934448,
|
| 1040 |
+
"learning_rate": 3.145361596611795e-05,
|
| 1041 |
+
"loss": 0.416,
|
| 1042 |
+
"num_input_tokens_seen": 234368,
|
| 1043 |
+
"step": 595
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 4.8,
|
| 1047 |
+
"grad_norm": 0.7484014630317688,
|
| 1048 |
+
"learning_rate": 3.111576314057268e-05,
|
| 1049 |
+
"loss": 0.395,
|
| 1050 |
+
"num_input_tokens_seen": 236032,
|
| 1051 |
+
"step": 600
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 4.84,
|
| 1055 |
+
"grad_norm": 0.9487155079841614,
|
| 1056 |
+
"learning_rate": 3.0776718034885454e-05,
|
| 1057 |
+
"loss": 0.3691,
|
| 1058 |
+
"num_input_tokens_seen": 237920,
|
| 1059 |
+
"step": 605
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 4.88,
|
| 1063 |
+
"grad_norm": 0.6988774538040161,
|
| 1064 |
+
"learning_rate": 3.0436546746571372e-05,
|
| 1065 |
+
"loss": 0.3724,
|
| 1066 |
+
"num_input_tokens_seen": 239680,
|
| 1067 |
+
"step": 610
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 4.92,
|
| 1071 |
+
"grad_norm": 0.8400317430496216,
|
| 1072 |
+
"learning_rate": 3.0095315592697126e-05,
|
| 1073 |
+
"loss": 0.3814,
|
| 1074 |
+
"num_input_tokens_seen": 241504,
|
| 1075 |
+
"step": 615
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 4.96,
|
| 1079 |
+
"grad_norm": 0.9846633672714233,
|
| 1080 |
+
"learning_rate": 2.9753091096952255e-05,
|
| 1081 |
+
"loss": 0.4676,
|
| 1082 |
+
"num_input_tokens_seen": 243584,
|
| 1083 |
+
"step": 620
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 5.0,
|
| 1087 |
+
"grad_norm": 0.6307002902030945,
|
| 1088 |
+
"learning_rate": 2.9409939976680313e-05,
|
| 1089 |
+
"loss": 0.4232,
|
| 1090 |
+
"num_input_tokens_seen": 245472,
|
| 1091 |
+
"step": 625
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 5.04,
|
| 1095 |
+
"grad_norm": 1.2824044227600098,
|
| 1096 |
+
"learning_rate": 2.9065929129872094e-05,
|
| 1097 |
+
"loss": 0.4193,
|
| 1098 |
+
"num_input_tokens_seen": 247424,
|
| 1099 |
+
"step": 630
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 5.04,
|
| 1103 |
+
"eval_loss": 0.42154452204704285,
|
| 1104 |
+
"eval_runtime": 0.8526,
|
| 1105 |
+
"eval_samples_per_second": 65.682,
|
| 1106 |
+
"eval_steps_per_second": 16.42,
|
| 1107 |
+
"num_input_tokens_seen": 247424,
|
| 1108 |
+
"step": 630
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 5.08,
|
| 1112 |
+
"grad_norm": 0.6836050152778625,
|
| 1113 |
+
"learning_rate": 2.8721125622123806e-05,
|
| 1114 |
+
"loss": 0.3778,
|
| 1115 |
+
"num_input_tokens_seen": 249472,
|
| 1116 |
+
"step": 635
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 5.12,
|
| 1120 |
+
"grad_norm": 1.2711609601974487,
|
| 1121 |
+
"learning_rate": 2.8375596673562482e-05,
|
| 1122 |
+
"loss": 0.3189,
|
| 1123 |
+
"num_input_tokens_seen": 251296,
|
| 1124 |
+
"step": 640
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 5.16,
|
| 1128 |
+
"grad_norm": 1.1848688125610352,
|
| 1129 |
+
"learning_rate": 2.8029409645741267e-05,
|
| 1130 |
+
"loss": 0.3988,
|
| 1131 |
+
"num_input_tokens_seen": 253344,
|
| 1132 |
+
"step": 645
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 5.2,
|
| 1136 |
+
"grad_norm": 0.43571653962135315,
|
| 1137 |
+
"learning_rate": 2.7682632028507167e-05,
|
| 1138 |
+
"loss": 0.3687,
|
| 1139 |
+
"num_input_tokens_seen": 255104,
|
| 1140 |
+
"step": 650
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 5.24,
|
| 1144 |
+
"grad_norm": 2.1147029399871826,
|
| 1145 |
+
"learning_rate": 2.733533142684377e-05,
|
| 1146 |
+
"loss": 0.3907,
|
| 1147 |
+
"num_input_tokens_seen": 256832,
|
| 1148 |
+
"step": 655
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 5.28,
|
| 1152 |
+
"grad_norm": 0.5585914850234985,
|
| 1153 |
+
"learning_rate": 2.6987575547691497e-05,
|
| 1154 |
+
"loss": 0.4176,
|
| 1155 |
+
"num_input_tokens_seen": 258720,
|
| 1156 |
+
"step": 660
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 5.32,
|
| 1160 |
+
"grad_norm": 1.013206124305725,
|
| 1161 |
+
"learning_rate": 2.6639432186748043e-05,
|
| 1162 |
+
"loss": 0.4127,
|
| 1163 |
+
"num_input_tokens_seen": 260576,
|
| 1164 |
+
"step": 665
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 5.36,
|
| 1168 |
+
"grad_norm": 0.6060461401939392,
|
| 1169 |
+
"learning_rate": 2.6290969215251416e-05,
|
| 1170 |
+
"loss": 0.3817,
|
| 1171 |
+
"num_input_tokens_seen": 262368,
|
| 1172 |
+
"step": 670
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 5.4,
|
| 1176 |
+
"grad_norm": 0.9579009413719177,
|
| 1177 |
+
"learning_rate": 2.594225456674837e-05,
|
| 1178 |
+
"loss": 0.3826,
|
| 1179 |
+
"num_input_tokens_seen": 264320,
|
| 1180 |
+
"step": 675
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 5.44,
|
| 1184 |
+
"grad_norm": 1.1483041048049927,
|
| 1185 |
+
"learning_rate": 2.559335622385055e-05,
|
| 1186 |
+
"loss": 0.3825,
|
| 1187 |
+
"num_input_tokens_seen": 266304,
|
| 1188 |
+
"step": 680
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 5.48,
|
| 1192 |
+
"grad_norm": 0.8167299628257751,
|
| 1193 |
+
"learning_rate": 2.524434220498123e-05,
|
| 1194 |
+
"loss": 0.3944,
|
| 1195 |
+
"num_input_tokens_seen": 268384,
|
| 1196 |
+
"step": 685
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 5.52,
|
| 1200 |
+
"grad_norm": 0.5977395176887512,
|
| 1201 |
+
"learning_rate": 2.4895280551114907e-05,
|
| 1202 |
+
"loss": 0.3691,
|
| 1203 |
+
"num_input_tokens_seen": 270208,
|
| 1204 |
+
"step": 690
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 5.5440000000000005,
|
| 1208 |
+
"eval_loss": 0.4073176383972168,
|
| 1209 |
+
"eval_runtime": 0.8504,
|
| 1210 |
+
"eval_samples_per_second": 65.848,
|
| 1211 |
+
"eval_steps_per_second": 16.462,
|
| 1212 |
+
"num_input_tokens_seen": 271232,
|
| 1213 |
+
"step": 693
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 5.5600000000000005,
|
| 1217 |
+
"grad_norm": 1.8195998668670654,
|
| 1218 |
+
"learning_rate": 2.4546239312512635e-05,
|
| 1219 |
+
"loss": 0.4443,
|
| 1220 |
+
"num_input_tokens_seen": 271840,
|
| 1221 |
+
"step": 695
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 5.6,
|
| 1225 |
+
"grad_norm": 1.045985460281372,
|
| 1226 |
+
"learning_rate": 2.4197286535455464e-05,
|
| 1227 |
+
"loss": 0.502,
|
| 1228 |
+
"num_input_tokens_seen": 273888,
|
| 1229 |
+
"step": 700
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 5.64,
|
| 1233 |
+
"grad_norm": 1.0764570236206055,
|
| 1234 |
+
"learning_rate": 2.384849024897869e-05,
|
| 1235 |
+
"loss": 0.4579,
|
| 1236 |
+
"num_input_tokens_seen": 275904,
|
| 1237 |
+
"step": 705
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 5.68,
|
| 1241 |
+
"grad_norm": 0.2045237421989441,
|
| 1242 |
+
"learning_rate": 2.349991845160949e-05,
|
| 1243 |
+
"loss": 0.4459,
|
| 1244 |
+
"num_input_tokens_seen": 277888,
|
| 1245 |
+
"step": 710
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 5.72,
|
| 1249 |
+
"grad_norm": 1.0971755981445312,
|
| 1250 |
+
"learning_rate": 2.3151639098110377e-05,
|
| 1251 |
+
"loss": 0.4206,
|
| 1252 |
+
"num_input_tokens_seen": 279872,
|
| 1253 |
+
"step": 715
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 5.76,
|
| 1257 |
+
"grad_norm": 0.39166244864463806,
|
| 1258 |
+
"learning_rate": 2.280372008623142e-05,
|
| 1259 |
+
"loss": 0.3705,
|
| 1260 |
+
"num_input_tokens_seen": 281664,
|
| 1261 |
+
"step": 720
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 5.8,
|
| 1265 |
+
"grad_norm": 0.7744928002357483,
|
| 1266 |
+
"learning_rate": 2.2456229243473345e-05,
|
| 1267 |
+
"loss": 0.3991,
|
| 1268 |
+
"num_input_tokens_seen": 283776,
|
| 1269 |
+
"step": 725
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 5.84,
|
| 1273 |
+
"grad_norm": 0.4687577486038208,
|
| 1274 |
+
"learning_rate": 2.2109234313864465e-05,
|
| 1275 |
+
"loss": 0.3788,
|
| 1276 |
+
"num_input_tokens_seen": 285568,
|
| 1277 |
+
"step": 730
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 5.88,
|
| 1281 |
+
"grad_norm": 0.439230352640152,
|
| 1282 |
+
"learning_rate": 2.176280294475383e-05,
|
| 1283 |
+
"loss": 0.3795,
|
| 1284 |
+
"num_input_tokens_seen": 287360,
|
| 1285 |
+
"step": 735
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 5.92,
|
| 1289 |
+
"grad_norm": 0.6617324948310852,
|
| 1290 |
+
"learning_rate": 2.1417002673623264e-05,
|
| 1291 |
+
"loss": 0.4557,
|
| 1292 |
+
"num_input_tokens_seen": 289632,
|
| 1293 |
+
"step": 740
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 5.96,
|
| 1297 |
+
"grad_norm": 0.32485124468803406,
|
| 1298 |
+
"learning_rate": 2.1071900914920816e-05,
|
| 1299 |
+
"loss": 0.4056,
|
| 1300 |
+
"num_input_tokens_seen": 291552,
|
| 1301 |
+
"step": 745
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 6.0,
|
| 1305 |
+
"grad_norm": 1.3294658660888672,
|
| 1306 |
+
"learning_rate": 2.0727564946918087e-05,
|
| 1307 |
+
"loss": 0.4079,
|
| 1308 |
+
"num_input_tokens_seen": 293616,
|
| 1309 |
+
"step": 750
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 6.04,
|
| 1313 |
+
"grad_norm": 0.7842015027999878,
|
| 1314 |
+
"learning_rate": 2.038406189859433e-05,
|
| 1315 |
+
"loss": 0.3746,
|
| 1316 |
+
"num_input_tokens_seen": 295440,
|
| 1317 |
+
"step": 755
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 6.048,
|
| 1321 |
+
"eval_loss": 0.40050727128982544,
|
| 1322 |
+
"eval_runtime": 0.8608,
|
| 1323 |
+
"eval_samples_per_second": 65.059,
|
| 1324 |
+
"eval_steps_per_second": 16.265,
|
| 1325 |
+
"num_input_tokens_seen": 295728,
|
| 1326 |
+
"step": 756
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 6.08,
|
| 1330 |
+
"grad_norm": 0.7505316734313965,
|
| 1331 |
+
"learning_rate": 2.004145873654942e-05,
|
| 1332 |
+
"loss": 0.4006,
|
| 1333 |
+
"num_input_tokens_seen": 297360,
|
| 1334 |
+
"step": 760
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 6.12,
|
| 1338 |
+
"grad_norm": 0.787607729434967,
|
| 1339 |
+
"learning_rate": 1.969982225194864e-05,
|
| 1340 |
+
"loss": 0.3928,
|
| 1341 |
+
"num_input_tokens_seen": 299312,
|
| 1342 |
+
"step": 765
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 6.16,
|
| 1346 |
+
"grad_norm": 0.8527250289916992,
|
| 1347 |
+
"learning_rate": 1.9359219047501565e-05,
|
| 1348 |
+
"loss": 0.3839,
|
| 1349 |
+
"num_input_tokens_seen": 301488,
|
| 1350 |
+
"step": 770
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 6.2,
|
| 1354 |
+
"grad_norm": 0.8097345232963562,
|
| 1355 |
+
"learning_rate": 1.9019715524477767e-05,
|
| 1356 |
+
"loss": 0.4211,
|
| 1357 |
+
"num_input_tokens_seen": 303696,
|
| 1358 |
+
"step": 775
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 6.24,
|
| 1362 |
+
"grad_norm": 1.303476095199585,
|
| 1363 |
+
"learning_rate": 1.868137786976177e-05,
|
| 1364 |
+
"loss": 0.3858,
|
| 1365 |
+
"num_input_tokens_seen": 305360,
|
| 1366 |
+
"step": 780
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 6.28,
|
| 1370 |
+
"grad_norm": 0.7829755544662476,
|
| 1371 |
+
"learning_rate": 1.8344272042949724e-05,
|
| 1372 |
+
"loss": 0.3779,
|
| 1373 |
+
"num_input_tokens_seen": 307408,
|
| 1374 |
+
"step": 785
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 6.32,
|
| 1378 |
+
"grad_norm": 0.659096360206604,
|
| 1379 |
+
"learning_rate": 1.800846376349051e-05,
|
| 1380 |
+
"loss": 0.4222,
|
| 1381 |
+
"num_input_tokens_seen": 309232,
|
| 1382 |
+
"step": 790
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 6.36,
|
| 1386 |
+
"grad_norm": 0.6200412511825562,
|
| 1387 |
+
"learning_rate": 1.767401849787357e-05,
|
| 1388 |
+
"loss": 0.3675,
|
| 1389 |
+
"num_input_tokens_seen": 311184,
|
| 1390 |
+
"step": 795
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 6.4,
|
| 1394 |
+
"grad_norm": 0.5834780335426331,
|
| 1395 |
+
"learning_rate": 1.73410014468661e-05,
|
| 1396 |
+
"loss": 0.3727,
|
| 1397 |
+
"num_input_tokens_seen": 313072,
|
| 1398 |
+
"step": 800
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 6.44,
|
| 1402 |
+
"grad_norm": 1.6976076364517212,
|
| 1403 |
+
"learning_rate": 1.7009477532802054e-05,
|
| 1404 |
+
"loss": 0.3823,
|
| 1405 |
+
"num_input_tokens_seen": 314832,
|
| 1406 |
+
"step": 805
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 6.48,
|
| 1410 |
+
"grad_norm": 1.1492793560028076,
|
| 1411 |
+
"learning_rate": 1.6679511386925337e-05,
|
| 1412 |
+
"loss": 0.3356,
|
| 1413 |
+
"num_input_tokens_seen": 316560,
|
| 1414 |
+
"step": 810
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 6.52,
|
| 1418 |
+
"grad_norm": 0.6304459571838379,
|
| 1419 |
+
"learning_rate": 1.635116733678988e-05,
|
| 1420 |
+
"loss": 0.427,
|
| 1421 |
+
"num_input_tokens_seen": 318960,
|
| 1422 |
+
"step": 815
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 6.552,
|
| 1426 |
+
"eval_loss": 0.40054336190223694,
|
| 1427 |
+
"eval_runtime": 0.8527,
|
| 1428 |
+
"eval_samples_per_second": 65.674,
|
| 1429 |
+
"eval_steps_per_second": 16.418,
|
| 1430 |
+
"num_input_tokens_seen": 320464,
|
| 1431 |
+
"step": 819
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 6.5600000000000005,
|
| 1435 |
+
"grad_norm": 0.825368344783783,
|
| 1436 |
+
"learning_rate": 1.6024509393718844e-05,
|
| 1437 |
+
"loss": 0.3747,
|
| 1438 |
+
"num_input_tokens_seen": 320880,
|
| 1439 |
+
"step": 820
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 6.6,
|
| 1443 |
+
"grad_norm": 0.5931684970855713,
|
| 1444 |
+
"learning_rate": 1.5699601240325474e-05,
|
| 1445 |
+
"loss": 0.3936,
|
| 1446 |
+
"num_input_tokens_seen": 323184,
|
| 1447 |
+
"step": 825
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 6.64,
|
| 1451 |
+
"grad_norm": 0.8659843802452087,
|
| 1452 |
+
"learning_rate": 1.5376506218098015e-05,
|
| 1453 |
+
"loss": 0.371,
|
| 1454 |
+
"num_input_tokens_seen": 325168,
|
| 1455 |
+
"step": 830
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 6.68,
|
| 1459 |
+
"grad_norm": 0.6564781069755554,
|
| 1460 |
+
"learning_rate": 1.505528731505126e-05,
|
| 1461 |
+
"loss": 0.3795,
|
| 1462 |
+
"num_input_tokens_seen": 326992,
|
| 1463 |
+
"step": 835
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 6.72,
|
| 1467 |
+
"grad_norm": 0.9912712574005127,
|
| 1468 |
+
"learning_rate": 1.4736007153446801e-05,
|
| 1469 |
+
"loss": 0.4107,
|
| 1470 |
+
"num_input_tokens_seen": 329104,
|
| 1471 |
+
"step": 840
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 6.76,
|
| 1475 |
+
"grad_norm": 0.7603425979614258,
|
| 1476 |
+
"learning_rate": 1.4418727977584774e-05,
|
| 1477 |
+
"loss": 0.3653,
|
| 1478 |
+
"num_input_tokens_seen": 331088,
|
| 1479 |
+
"step": 845
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 6.8,
|
| 1483 |
+
"grad_norm": 0.5329681038856506,
|
| 1484 |
+
"learning_rate": 1.4103511641669152e-05,
|
| 1485 |
+
"loss": 0.3939,
|
| 1486 |
+
"num_input_tokens_seen": 333008,
|
| 1487 |
+
"step": 850
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 6.84,
|
| 1491 |
+
"grad_norm": 0.5622363686561584,
|
| 1492 |
+
"learning_rate": 1.3790419597749199e-05,
|
| 1493 |
+
"loss": 0.3725,
|
| 1494 |
+
"num_input_tokens_seen": 335024,
|
| 1495 |
+
"step": 855
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 6.88,
|
| 1499 |
+
"grad_norm": 0.9950863718986511,
|
| 1500 |
+
"learning_rate": 1.3479512883739232e-05,
|
| 1501 |
+
"loss": 0.4179,
|
| 1502 |
+
"num_input_tokens_seen": 337104,
|
| 1503 |
+
"step": 860
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 6.92,
|
| 1507 |
+
"grad_norm": 0.7129436135292053,
|
| 1508 |
+
"learning_rate": 1.3170852111519175e-05,
|
| 1509 |
+
"loss": 0.3773,
|
| 1510 |
+
"num_input_tokens_seen": 338960,
|
| 1511 |
+
"step": 865
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 6.96,
|
| 1515 |
+
"grad_norm": 1.585489273071289,
|
| 1516 |
+
"learning_rate": 1.2864497455118152e-05,
|
| 1517 |
+
"loss": 0.3702,
|
| 1518 |
+
"num_input_tokens_seen": 340848,
|
| 1519 |
+
"step": 870
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 7.0,
|
| 1523 |
+
"grad_norm": 1.2564765214920044,
|
| 1524 |
+
"learning_rate": 1.2560508638983437e-05,
|
| 1525 |
+
"loss": 0.3594,
|
| 1526 |
+
"num_input_tokens_seen": 343040,
|
| 1527 |
+
"step": 875
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 7.04,
|
| 1531 |
+
"grad_norm": 0.6992194056510925,
|
| 1532 |
+
"learning_rate": 1.2258944926337057e-05,
|
| 1533 |
+
"loss": 0.3347,
|
| 1534 |
+
"num_input_tokens_seen": 345056,
|
| 1535 |
+
"step": 880
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 7.056,
|
| 1539 |
+
"eval_loss": 0.41065576672554016,
|
| 1540 |
+
"eval_runtime": 0.8558,
|
| 1541 |
+
"eval_samples_per_second": 65.435,
|
| 1542 |
+
"eval_steps_per_second": 16.359,
|
| 1543 |
+
"num_input_tokens_seen": 345856,
|
| 1544 |
+
"step": 882
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 7.08,
|
| 1548 |
+
"grad_norm": 0.7301174402236938,
|
| 1549 |
+
"learning_rate": 1.1959865107622307e-05,
|
| 1550 |
+
"loss": 0.3597,
|
| 1551 |
+
"num_input_tokens_seen": 347232,
|
| 1552 |
+
"step": 885
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 7.12,
|
| 1556 |
+
"grad_norm": 1.776132345199585,
|
| 1557 |
+
"learning_rate": 1.1663327489042435e-05,
|
| 1558 |
+
"loss": 0.4383,
|
| 1559 |
+
"num_input_tokens_seen": 349504,
|
| 1560 |
+
"step": 890
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 7.16,
|
| 1564 |
+
"grad_norm": 0.7216264605522156,
|
| 1565 |
+
"learning_rate": 1.1369389881193749e-05,
|
| 1566 |
+
"loss": 0.4164,
|
| 1567 |
+
"num_input_tokens_seen": 351296,
|
| 1568 |
+
"step": 895
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 7.2,
|
| 1572 |
+
"grad_norm": 1.4257539510726929,
|
| 1573 |
+
"learning_rate": 1.107810958779531e-05,
|
| 1574 |
+
"loss": 0.3858,
|
| 1575 |
+
"num_input_tokens_seen": 353248,
|
| 1576 |
+
"step": 900
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 7.24,
|
| 1580 |
+
"grad_norm": 1.8053635358810425,
|
| 1581 |
+
"learning_rate": 1.0789543394517435e-05,
|
| 1582 |
+
"loss": 0.4069,
|
| 1583 |
+
"num_input_tokens_seen": 355232,
|
| 1584 |
+
"step": 905
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 7.28,
|
| 1588 |
+
"grad_norm": 0.6325796246528625,
|
| 1589 |
+
"learning_rate": 1.050374755791127e-05,
|
| 1590 |
+
"loss": 0.3591,
|
| 1591 |
+
"num_input_tokens_seen": 357376,
|
| 1592 |
+
"step": 910
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 7.32,
|
| 1596 |
+
"grad_norm": 1.1207585334777832,
|
| 1597 |
+
"learning_rate": 1.022077779444145e-05,
|
| 1598 |
+
"loss": 0.3539,
|
| 1599 |
+
"num_input_tokens_seen": 359232,
|
| 1600 |
+
"step": 915
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 7.36,
|
| 1604 |
+
"grad_norm": 1.4070286750793457,
|
| 1605 |
+
"learning_rate": 9.94068926962404e-06,
|
| 1606 |
+
"loss": 0.3527,
|
| 1607 |
+
"num_input_tokens_seen": 361024,
|
| 1608 |
+
"step": 920
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 7.4,
|
| 1612 |
+
"grad_norm": 0.627964973449707,
|
| 1613 |
+
"learning_rate": 9.663536587271902e-06,
|
| 1614 |
+
"loss": 0.376,
|
| 1615 |
+
"num_input_tokens_seen": 363200,
|
| 1616 |
+
"step": 925
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 7.44,
|
| 1620 |
+
"grad_norm": 0.6566188335418701,
|
| 1621 |
+
"learning_rate": 9.389373778849612e-06,
|
| 1622 |
+
"loss": 0.3517,
|
| 1623 |
+
"num_input_tokens_seen": 365152,
|
| 1624 |
+
"step": 930
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 7.48,
|
| 1628 |
+
"grad_norm": 0.6392218470573425,
|
| 1629 |
+
"learning_rate": 9.11825429293989e-06,
|
| 1630 |
+
"loss": 0.4175,
|
| 1631 |
+
"num_input_tokens_seen": 367328,
|
| 1632 |
+
"step": 935
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 7.52,
|
| 1636 |
+
"grad_norm": 0.5262883901596069,
|
| 1637 |
+
"learning_rate": 8.850230984823735e-06,
|
| 1638 |
+
"loss": 0.358,
|
| 1639 |
+
"num_input_tokens_seen": 369248,
|
| 1640 |
+
"step": 940
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 7.5600000000000005,
|
| 1644 |
+
"grad_norm": 0.9610680937767029,
|
| 1645 |
+
"learning_rate": 8.585356106176094e-06,
|
| 1646 |
+
"loss": 0.331,
|
| 1647 |
+
"num_input_tokens_seen": 371040,
|
| 1648 |
+
"step": 945
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 7.5600000000000005,
|
| 1652 |
+
"eval_loss": 0.4088888466358185,
|
| 1653 |
+
"eval_runtime": 0.8481,
|
| 1654 |
+
"eval_samples_per_second": 66.028,
|
| 1655 |
+
"eval_steps_per_second": 16.507,
|
| 1656 |
+
"num_input_tokens_seen": 371040,
|
| 1657 |
+
"step": 945
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 7.6,
|
| 1661 |
+
"grad_norm": 0.7290608882904053,
|
| 1662 |
+
"learning_rate": 8.323681294879394e-06,
|
| 1663 |
+
"loss": 0.4139,
|
| 1664 |
+
"num_input_tokens_seen": 372928,
|
| 1665 |
+
"step": 950
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 7.64,
|
| 1669 |
+
"grad_norm": 0.532922089099884,
|
| 1670 |
+
"learning_rate": 8.06525756495657e-06,
|
| 1671 |
+
"loss": 0.4047,
|
| 1672 |
+
"num_input_tokens_seen": 374816,
|
| 1673 |
+
"step": 955
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 7.68,
|
| 1677 |
+
"grad_norm": 0.77403724193573,
|
| 1678 |
+
"learning_rate": 7.810135296625818e-06,
|
| 1679 |
+
"loss": 0.3748,
|
| 1680 |
+
"num_input_tokens_seen": 376704,
|
| 1681 |
+
"step": 960
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 7.72,
|
| 1685 |
+
"grad_norm": 0.5218887329101562,
|
| 1686 |
+
"learning_rate": 7.558364226478842e-06,
|
| 1687 |
+
"loss": 0.3622,
|
| 1688 |
+
"num_input_tokens_seen": 378624,
|
| 1689 |
+
"step": 965
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 7.76,
|
| 1693 |
+
"grad_norm": 0.5533618927001953,
|
| 1694 |
+
"learning_rate": 7.309993437784624e-06,
|
| 1695 |
+
"loss": 0.3991,
|
| 1696 |
+
"num_input_tokens_seen": 380480,
|
| 1697 |
+
"step": 970
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 7.8,
|
| 1701 |
+
"grad_norm": 0.6465098261833191,
|
| 1702 |
+
"learning_rate": 7.065071350920538e-06,
|
| 1703 |
+
"loss": 0.3731,
|
| 1704 |
+
"num_input_tokens_seen": 382144,
|
| 1705 |
+
"step": 975
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 7.84,
|
| 1709 |
+
"grad_norm": 0.6888776421546936,
|
| 1710 |
+
"learning_rate": 6.823645713932708e-06,
|
| 1711 |
+
"loss": 0.4033,
|
| 1712 |
+
"num_input_tokens_seen": 384320,
|
| 1713 |
+
"step": 980
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 7.88,
|
| 1717 |
+
"grad_norm": 0.6575106382369995,
|
| 1718 |
+
"learning_rate": 6.58576359322742e-06,
|
| 1719 |
+
"loss": 0.4272,
|
| 1720 |
+
"num_input_tokens_seen": 386656,
|
| 1721 |
+
"step": 985
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 7.92,
|
| 1725 |
+
"grad_norm": 0.4115797281265259,
|
| 1726 |
+
"learning_rate": 6.3514713643954475e-06,
|
| 1727 |
+
"loss": 0.3612,
|
| 1728 |
+
"num_input_tokens_seen": 388672,
|
| 1729 |
+
"step": 990
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 7.96,
|
| 1733 |
+
"grad_norm": 1.37602698802948,
|
| 1734 |
+
"learning_rate": 6.120814703171024e-06,
|
| 1735 |
+
"loss": 0.3805,
|
| 1736 |
+
"num_input_tokens_seen": 390400,
|
| 1737 |
+
"step": 995
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 8.0,
|
| 1741 |
+
"grad_norm": 0.3773898780345917,
|
| 1742 |
+
"learning_rate": 5.893838576527275e-06,
|
| 1743 |
+
"loss": 0.3604,
|
| 1744 |
+
"num_input_tokens_seen": 392080,
|
| 1745 |
+
"step": 1000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 8.04,
|
| 1749 |
+
"grad_norm": 0.7060356140136719,
|
| 1750 |
+
"learning_rate": 5.6705872339098186e-06,
|
| 1751 |
+
"loss": 0.4144,
|
| 1752 |
+
"num_input_tokens_seen": 394160,
|
| 1753 |
+
"step": 1005
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 8.064,
|
| 1757 |
+
"eval_loss": 0.38490504026412964,
|
| 1758 |
+
"eval_runtime": 0.8524,
|
| 1759 |
+
"eval_samples_per_second": 65.693,
|
| 1760 |
+
"eval_steps_per_second": 16.423,
|
| 1761 |
+
"num_input_tokens_seen": 395216,
|
| 1762 |
+
"step": 1008
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 8.08,
|
| 1766 |
+
"grad_norm": 0.7593961954116821,
|
| 1767 |
+
"learning_rate": 5.451104198610249e-06,
|
| 1768 |
+
"loss": 0.3568,
|
| 1769 |
+
"num_input_tokens_seen": 395888,
|
| 1770 |
+
"step": 1010
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 8.12,
|
| 1774 |
+
"grad_norm": 1.2459512948989868,
|
| 1775 |
+
"learning_rate": 5.235432259281175e-06,
|
| 1776 |
+
"loss": 0.3882,
|
| 1777 |
+
"num_input_tokens_seen": 398032,
|
| 1778 |
+
"step": 1015
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 8.16,
|
| 1782 |
+
"grad_norm": 1.0449175834655762,
|
| 1783 |
+
"learning_rate": 5.023613461594512e-06,
|
| 1784 |
+
"loss": 0.3798,
|
| 1785 |
+
"num_input_tokens_seen": 399856,
|
| 1786 |
+
"step": 1020
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 8.2,
|
| 1790 |
+
"grad_norm": 0.7643663287162781,
|
| 1791 |
+
"learning_rate": 4.8156891000445406e-06,
|
| 1792 |
+
"loss": 0.3718,
|
| 1793 |
+
"num_input_tokens_seen": 401616,
|
| 1794 |
+
"step": 1025
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 8.24,
|
| 1798 |
+
"grad_norm": 0.7826510071754456,
|
| 1799 |
+
"learning_rate": 4.6116997098975465e-06,
|
| 1800 |
+
"loss": 0.3458,
|
| 1801 |
+
"num_input_tokens_seen": 403568,
|
| 1802 |
+
"step": 1030
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 8.28,
|
| 1806 |
+
"grad_norm": 0.6359825134277344,
|
| 1807 |
+
"learning_rate": 4.411685059289314e-06,
|
| 1808 |
+
"loss": 0.3725,
|
| 1809 |
+
"num_input_tokens_seen": 405712,
|
| 1810 |
+
"step": 1035
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 8.32,
|
| 1814 |
+
"grad_norm": 0.4394562244415283,
|
| 1815 |
+
"learning_rate": 4.215684141472292e-06,
|
| 1816 |
+
"loss": 0.3745,
|
| 1817 |
+
"num_input_tokens_seen": 407888,
|
| 1818 |
+
"step": 1040
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 8.36,
|
| 1822 |
+
"grad_norm": 0.6596431732177734,
|
| 1823 |
+
"learning_rate": 4.023735167213752e-06,
|
| 1824 |
+
"loss": 0.3587,
|
| 1825 |
+
"num_input_tokens_seen": 409712,
|
| 1826 |
+
"step": 1045
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 8.4,
|
| 1830 |
+
"grad_norm": 0.9594627618789673,
|
| 1831 |
+
"learning_rate": 3.835875557346552e-06,
|
| 1832 |
+
"loss": 0.3633,
|
| 1833 |
+
"num_input_tokens_seen": 411504,
|
| 1834 |
+
"step": 1050
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 8.44,
|
| 1838 |
+
"grad_norm": 0.5970214009284973,
|
| 1839 |
+
"learning_rate": 3.6521419354738738e-06,
|
| 1840 |
+
"loss": 0.368,
|
| 1841 |
+
"num_input_tokens_seen": 413168,
|
| 1842 |
+
"step": 1055
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 8.48,
|
| 1846 |
+
"grad_norm": 0.4645940959453583,
|
| 1847 |
+
"learning_rate": 3.4725701208293435e-06,
|
| 1848 |
+
"loss": 0.3561,
|
| 1849 |
+
"num_input_tokens_seen": 414960,
|
| 1850 |
+
"step": 1060
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 8.52,
|
| 1854 |
+
"grad_norm": 0.6897419691085815,
|
| 1855 |
+
"learning_rate": 3.297195121294022e-06,
|
| 1856 |
+
"loss": 0.3644,
|
| 1857 |
+
"num_input_tokens_seen": 416880,
|
| 1858 |
+
"step": 1065
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 8.56,
|
| 1862 |
+
"grad_norm": 0.8192169666290283,
|
| 1863 |
+
"learning_rate": 3.126051126571561e-06,
|
| 1864 |
+
"loss": 0.3779,
|
| 1865 |
+
"num_input_tokens_seen": 418768,
|
| 1866 |
+
"step": 1070
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 8.568,
|
| 1870 |
+
"eval_loss": 0.38693496584892273,
|
| 1871 |
+
"eval_runtime": 0.8505,
|
| 1872 |
+
"eval_samples_per_second": 65.843,
|
| 1873 |
+
"eval_steps_per_second": 16.461,
|
| 1874 |
+
"num_input_tokens_seen": 419184,
|
| 1875 |
+
"step": 1071
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 8.6,
|
| 1879 |
+
"grad_norm": 0.9851546287536621,
|
| 1880 |
+
"learning_rate": 2.9591715015228284e-06,
|
| 1881 |
+
"loss": 0.4053,
|
| 1882 |
+
"num_input_tokens_seen": 421008,
|
| 1883 |
+
"step": 1075
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 8.64,
|
| 1887 |
+
"grad_norm": 0.5261719226837158,
|
| 1888 |
+
"learning_rate": 2.7965887796613884e-06,
|
| 1889 |
+
"loss": 0.3589,
|
| 1890 |
+
"num_input_tokens_seen": 422864,
|
| 1891 |
+
"step": 1080
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 8.68,
|
| 1895 |
+
"grad_norm": 0.7083816528320312,
|
| 1896 |
+
"learning_rate": 2.6383346568110062e-06,
|
| 1897 |
+
"loss": 0.3792,
|
| 1898 |
+
"num_input_tokens_seen": 424976,
|
| 1899 |
+
"step": 1085
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 8.72,
|
| 1903 |
+
"grad_norm": 0.8124401569366455,
|
| 1904 |
+
"learning_rate": 2.4844399849264928e-06,
|
| 1905 |
+
"loss": 0.3899,
|
| 1906 |
+
"num_input_tokens_seen": 427120,
|
| 1907 |
+
"step": 1090
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 8.76,
|
| 1911 |
+
"grad_norm": 0.7263258695602417,
|
| 1912 |
+
"learning_rate": 2.3349347660790582e-06,
|
| 1913 |
+
"loss": 0.4017,
|
| 1914 |
+
"num_input_tokens_seen": 429264,
|
| 1915 |
+
"step": 1095
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 8.8,
|
| 1919 |
+
"grad_norm": 0.6860739588737488,
|
| 1920 |
+
"learning_rate": 2.189848146607348e-06,
|
| 1921 |
+
"loss": 0.3718,
|
| 1922 |
+
"num_input_tokens_seen": 431056,
|
| 1923 |
+
"step": 1100
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 8.84,
|
| 1927 |
+
"grad_norm": 0.49945640563964844,
|
| 1928 |
+
"learning_rate": 2.0492084114352965e-06,
|
| 1929 |
+
"loss": 0.3644,
|
| 1930 |
+
"num_input_tokens_seen": 433136,
|
| 1931 |
+
"step": 1105
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 8.88,
|
| 1935 |
+
"grad_norm": 0.8145077228546143,
|
| 1936 |
+
"learning_rate": 1.913042978557944e-06,
|
| 1937 |
+
"loss": 0.3859,
|
| 1938 |
+
"num_input_tokens_seen": 435216,
|
| 1939 |
+
"step": 1110
|
| 1940 |
+
},
|
| 1941 |
+
{
|
| 1942 |
+
"epoch": 8.92,
|
| 1943 |
+
"grad_norm": 1.1362378597259521,
|
| 1944 |
+
"learning_rate": 1.7813783936962258e-06,
|
| 1945 |
+
"loss": 0.3521,
|
| 1946 |
+
"num_input_tokens_seen": 437040,
|
| 1947 |
+
"step": 1115
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 8.96,
|
| 1951 |
+
"grad_norm": 1.012280821800232,
|
| 1952 |
+
"learning_rate": 1.654240325121831e-06,
|
| 1953 |
+
"loss": 0.3873,
|
| 1954 |
+
"num_input_tokens_seen": 439088,
|
| 1955 |
+
"step": 1120
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 9.0,
|
| 1959 |
+
"grad_norm": 1.0121548175811768,
|
| 1960 |
+
"learning_rate": 1.5316535586531483e-06,
|
| 1961 |
+
"loss": 0.4015,
|
| 1962 |
+
"num_input_tokens_seen": 440848,
|
| 1963 |
+
"step": 1125
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 9.04,
|
| 1967 |
+
"grad_norm": 0.9955313205718994,
|
| 1968 |
+
"learning_rate": 1.4136419928231892e-06,
|
| 1969 |
+
"loss": 0.3714,
|
| 1970 |
+
"num_input_tokens_seen": 442864,
|
| 1971 |
+
"step": 1130
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 9.072,
|
| 1975 |
+
"eval_loss": 0.38989585638046265,
|
| 1976 |
+
"eval_runtime": 0.8537,
|
| 1977 |
+
"eval_samples_per_second": 65.599,
|
| 1978 |
+
"eval_steps_per_second": 16.4,
|
| 1979 |
+
"num_input_tokens_seen": 444560,
|
| 1980 |
+
"step": 1134
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 9.08,
|
| 1984 |
+
"grad_norm": 0.6308397054672241,
|
| 1985 |
+
"learning_rate": 1.3002286342205462e-06,
|
| 1986 |
+
"loss": 0.3985,
|
| 1987 |
+
"num_input_tokens_seen": 445040,
|
| 1988 |
+
"step": 1135
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 9.12,
|
| 1992 |
+
"grad_norm": 0.438340961933136,
|
| 1993 |
+
"learning_rate": 1.1914355930041837e-06,
|
| 1994 |
+
"loss": 0.3583,
|
| 1995 |
+
"num_input_tokens_seen": 446864,
|
| 1996 |
+
"step": 1140
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 9.16,
|
| 2000 |
+
"grad_norm": 0.80867600440979,
|
| 2001 |
+
"learning_rate": 1.087284078593051e-06,
|
| 2002 |
+
"loss": 0.3824,
|
| 2003 |
+
"num_input_tokens_seen": 449008,
|
| 2004 |
+
"step": 1145
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 9.2,
|
| 2008 |
+
"grad_norm": 1.2045706510543823,
|
| 2009 |
+
"learning_rate": 9.877943955312552e-07,
|
| 2010 |
+
"loss": 0.3471,
|
| 2011 |
+
"num_input_tokens_seen": 450832,
|
| 2012 |
+
"step": 1150
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 9.24,
|
| 2016 |
+
"grad_norm": 1.406896948814392,
|
| 2017 |
+
"learning_rate": 8.929859395296364e-07,
|
| 2018 |
+
"loss": 0.3737,
|
| 2019 |
+
"num_input_tokens_seen": 452784,
|
| 2020 |
+
"step": 1155
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 9.28,
|
| 2024 |
+
"grad_norm": 0.849940836429596,
|
| 2025 |
+
"learning_rate": 8.028771936845342e-07,
|
| 2026 |
+
"loss": 0.3719,
|
| 2027 |
+
"num_input_tokens_seen": 454992,
|
| 2028 |
+
"step": 1160
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 9.32,
|
| 2032 |
+
"grad_norm": 0.889966607093811,
|
| 2033 |
+
"learning_rate": 7.174857248745004e-07,
|
| 2034 |
+
"loss": 0.3571,
|
| 2035 |
+
"num_input_tokens_seen": 456528,
|
| 2036 |
+
"step": 1165
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 9.36,
|
| 2040 |
+
"grad_norm": 1.1385633945465088,
|
| 2041 |
+
"learning_rate": 6.368281803355691e-07,
|
| 2042 |
+
"loss": 0.3921,
|
| 2043 |
+
"num_input_tokens_seen": 458768,
|
| 2044 |
+
"step": 1170
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 9.4,
|
| 2048 |
+
"grad_norm": 0.6559231877326965,
|
| 2049 |
+
"learning_rate": 5.609202844158723e-07,
|
| 2050 |
+
"loss": 0.3935,
|
| 2051 |
+
"num_input_tokens_seen": 460816,
|
| 2052 |
+
"step": 1175
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 9.44,
|
| 2056 |
+
"grad_norm": 0.8425350785255432,
|
| 2057 |
+
"learning_rate": 4.897768355101084e-07,
|
| 2058 |
+
"loss": 0.3657,
|
| 2059 |
+
"num_input_tokens_seen": 462384,
|
| 2060 |
+
"step": 1180
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 9.48,
|
| 2064 |
+
"grad_norm": 0.9185330271720886,
|
| 2065 |
+
"learning_rate": 4.234117031746143e-07,
|
| 2066 |
+
"loss": 0.3923,
|
| 2067 |
+
"num_input_tokens_seen": 464304,
|
| 2068 |
+
"step": 1185
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 9.52,
|
| 2072 |
+
"grad_norm": 0.5886545777320862,
|
| 2073 |
+
"learning_rate": 3.6183782542343057e-07,
|
| 2074 |
+
"loss": 0.3341,
|
| 2075 |
+
"num_input_tokens_seen": 466384,
|
| 2076 |
+
"step": 1190
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 9.56,
|
| 2080 |
+
"grad_norm": 0.7718898057937622,
|
| 2081 |
+
"learning_rate": 3.050672062060278e-07,
|
| 2082 |
+
"loss": 0.3858,
|
| 2083 |
+
"num_input_tokens_seen": 468368,
|
| 2084 |
+
"step": 1195
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 9.576,
|
| 2088 |
+
"eval_loss": 0.3861676752567291,
|
| 2089 |
+
"eval_runtime": 0.8505,
|
| 2090 |
+
"eval_samples_per_second": 65.845,
|
| 2091 |
+
"eval_steps_per_second": 16.461,
|
| 2092 |
+
"num_input_tokens_seen": 469104,
|
| 2093 |
+
"step": 1197
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 9.6,
|
| 2097 |
+
"grad_norm": 1.3115769624710083,
|
| 2098 |
+
"learning_rate": 2.531109130671061e-07,
|
| 2099 |
+
"loss": 0.3761,
|
| 2100 |
+
"num_input_tokens_seen": 470192,
|
| 2101 |
+
"step": 1200
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 9.64,
|
| 2105 |
+
"grad_norm": 0.625157356262207,
|
| 2106 |
+
"learning_rate": 2.0597907498896007e-07,
|
| 2107 |
+
"loss": 0.3878,
|
| 2108 |
+
"num_input_tokens_seen": 472240,
|
| 2109 |
+
"step": 1205
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 9.68,
|
| 2113 |
+
"grad_norm": 1.4318360090255737,
|
| 2114 |
+
"learning_rate": 1.6368088041681108e-07,
|
| 2115 |
+
"loss": 0.3748,
|
| 2116 |
+
"num_input_tokens_seen": 474160,
|
| 2117 |
+
"step": 1210
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 9.72,
|
| 2121 |
+
"grad_norm": 0.62953782081604,
|
| 2122 |
+
"learning_rate": 1.2622457546749567e-07,
|
| 2123 |
+
"loss": 0.4067,
|
| 2124 |
+
"num_input_tokens_seen": 476784,
|
| 2125 |
+
"step": 1215
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 9.76,
|
| 2129 |
+
"grad_norm": 1.3049143552780151,
|
| 2130 |
+
"learning_rate": 9.361746232188495e-08,
|
| 2131 |
+
"loss": 0.3679,
|
| 2132 |
+
"num_input_tokens_seen": 478864,
|
| 2133 |
+
"step": 1220
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 9.8,
|
| 2137 |
+
"grad_norm": 0.5870345234870911,
|
| 2138 |
+
"learning_rate": 6.586589780128716e-08,
|
| 2139 |
+
"loss": 0.3603,
|
| 2140 |
+
"num_input_tokens_seen": 480816,
|
| 2141 |
+
"step": 1225
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 9.84,
|
| 2145 |
+
"grad_norm": 0.5798351764678955,
|
| 2146 |
+
"learning_rate": 4.2975292128200064e-08,
|
| 2147 |
+
"loss": 0.384,
|
| 2148 |
+
"num_input_tokens_seen": 482832,
|
| 2149 |
+
"step": 1230
|
| 2150 |
+
},
|
| 2151 |
+
{
|
| 2152 |
+
"epoch": 9.88,
|
| 2153 |
+
"grad_norm": 0.6793395280838013,
|
| 2154 |
+
"learning_rate": 2.4950107871549167e-08,
|
| 2155 |
+
"loss": 0.3827,
|
| 2156 |
+
"num_input_tokens_seen": 484912,
|
| 2157 |
+
"step": 1235
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 9.92,
|
| 2161 |
+
"grad_norm": 0.9757604002952576,
|
| 2162 |
+
"learning_rate": 1.179385907672248e-08,
|
| 2163 |
+
"loss": 0.3863,
|
| 2164 |
+
"num_input_tokens_seen": 486640,
|
| 2165 |
+
"step": 1240
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 9.96,
|
| 2169 |
+
"grad_norm": 0.7342190742492676,
|
| 2170 |
+
"learning_rate": 3.5091105804907487e-09,
|
| 2171 |
+
"loss": 0.3537,
|
| 2172 |
+
"num_input_tokens_seen": 488336,
|
| 2173 |
+
"step": 1245
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 10.0,
|
| 2177 |
+
"grad_norm": 0.7796341180801392,
|
| 2178 |
+
"learning_rate": 9.747751098521107e-11,
|
| 2179 |
+
"loss": 0.3453,
|
| 2180 |
+
"num_input_tokens_seen": 490000,
|
| 2181 |
+
"step": 1250
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 10.0,
|
| 2185 |
+
"num_input_tokens_seen": 490000,
|
| 2186 |
+
"step": 1250,
|
| 2187 |
+
"total_flos": 2.206447853568e+16,
|
| 2188 |
+
"train_loss": 2.282332957649231,
|
| 2189 |
+
"train_runtime": 172.1756,
|
| 2190 |
+
"train_samples_per_second": 28.924,
|
| 2191 |
+
"train_steps_per_second": 7.26
|
| 2192 |
+
}
|
| 2193 |
+
],
|
| 2194 |
+
"logging_steps": 5,
|
| 2195 |
+
"max_steps": 1250,
|
| 2196 |
+
"num_input_tokens_seen": 490000,
|
| 2197 |
+
"num_train_epochs": 10,
|
| 2198 |
+
"save_steps": 63,
|
| 2199 |
+
"stateful_callbacks": {
|
| 2200 |
+
"TrainerControl": {
|
| 2201 |
+
"args": {
|
| 2202 |
+
"should_epoch_stop": false,
|
| 2203 |
+
"should_evaluate": false,
|
| 2204 |
+
"should_log": false,
|
| 2205 |
+
"should_save": true,
|
| 2206 |
+
"should_training_stop": true
|
| 2207 |
+
},
|
| 2208 |
+
"attributes": {}
|
| 2209 |
+
}
|
| 2210 |
+
},
|
| 2211 |
+
"total_flos": 2.206447853568e+16,
|
| 2212 |
+
"train_batch_size": 4,
|
| 2213 |
+
"trial_name": null,
|
| 2214 |
+
"trial_params": null
|
| 2215 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|