End of training
Browse files- README.md +3 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1263 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -3,6 +3,7 @@ license: llama3
|
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
- llama-factory
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 8 |
model-index:
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama-3-86-lora-pretrain_v2
|
| 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: 2.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 9 |
model-index:
|
|
|
|
| 16 |
|
| 17 |
# llama-3-86-lora-pretrain_v2
|
| 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 sm_artile dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.2353
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9964796996010326,
|
| 3 |
+
"eval_loss": 2.235288619995117,
|
| 4 |
+
"eval_runtime": 115.517,
|
| 5 |
+
"eval_samples_per_second": 8.198,
|
| 6 |
+
"eval_steps_per_second": 4.103,
|
| 7 |
+
"perplexity": 9.349179823068619,
|
| 8 |
+
"total_flos": 1.1780062420402176e+18,
|
| 9 |
+
"train_loss": 2.428264606566656,
|
| 10 |
+
"train_runtime": 11538.2228,
|
| 11 |
+
"train_samples_per_second": 2.216,
|
| 12 |
+
"train_steps_per_second": 0.138
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9964796996010326,
|
| 3 |
+
"eval_loss": 2.235288619995117,
|
| 4 |
+
"eval_runtime": 115.517,
|
| 5 |
+
"eval_samples_per_second": 8.198,
|
| 6 |
+
"eval_steps_per_second": 4.103,
|
| 7 |
+
"perplexity": 9.349179823068619
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9964796996010326,
|
| 3 |
+
"total_flos": 1.1780062420402176e+18,
|
| 4 |
+
"train_loss": 2.428264606566656,
|
| 5 |
+
"train_runtime": 11538.2228,
|
| 6 |
+
"train_samples_per_second": 2.216,
|
| 7 |
+
"train_steps_per_second": 0.138
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1263 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9964796996010326,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1596,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.018774935461159353,
|
| 13 |
+
"grad_norm": 0.3424220085144043,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 3.1063,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03754987092231871,
|
| 20 |
+
"grad_norm": 0.3918570578098297,
|
| 21 |
+
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 3.0935,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05632480638347806,
|
| 27 |
+
"grad_norm": 0.42540064454078674,
|
| 28 |
+
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 3.0475,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07509974184463741,
|
| 34 |
+
"grad_norm": 0.47563326358795166,
|
| 35 |
+
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 3.0609,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09387467730579677,
|
| 41 |
+
"grad_norm": 0.5615227818489075,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 3.026,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.11264961276695612,
|
| 48 |
+
"grad_norm": 0.5656551122665405,
|
| 49 |
+
"learning_rate": 1.2e-05,
|
| 50 |
+
"loss": 3.0134,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.13142454822811547,
|
| 55 |
+
"grad_norm": 0.5487608313560486,
|
| 56 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 57 |
+
"loss": 2.9549,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.15019948368927483,
|
| 62 |
+
"grad_norm": 0.48524507880210876,
|
| 63 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 64 |
+
"loss": 2.9301,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16897441915043418,
|
| 69 |
+
"grad_norm": 0.5230005383491516,
|
| 70 |
+
"learning_rate": 1.8e-05,
|
| 71 |
+
"loss": 2.9109,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.18774935461159353,
|
| 76 |
+
"grad_norm": 0.38439810276031494,
|
| 77 |
+
"learning_rate": 2e-05,
|
| 78 |
+
"loss": 2.8142,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.18774935461159353,
|
| 83 |
+
"eval_loss": 2.856565237045288,
|
| 84 |
+
"eval_runtime": 115.5533,
|
| 85 |
+
"eval_samples_per_second": 8.195,
|
| 86 |
+
"eval_steps_per_second": 4.102,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.2065242900727529,
|
| 91 |
+
"grad_norm": 0.4284297227859497,
|
| 92 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 93 |
+
"loss": 2.8967,
|
| 94 |
+
"step": 110
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.22529922553391224,
|
| 98 |
+
"grad_norm": 0.41896769404411316,
|
| 99 |
+
"learning_rate": 2.4e-05,
|
| 100 |
+
"loss": 2.8629,
|
| 101 |
+
"step": 120
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2440741609950716,
|
| 105 |
+
"grad_norm": 0.41066381335258484,
|
| 106 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 107 |
+
"loss": 2.8142,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.26284909645623095,
|
| 112 |
+
"grad_norm": 0.44229111075401306,
|
| 113 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 114 |
+
"loss": 2.815,
|
| 115 |
+
"step": 140
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.28162403191739027,
|
| 119 |
+
"grad_norm": 0.43809670209884644,
|
| 120 |
+
"learning_rate": 3e-05,
|
| 121 |
+
"loss": 2.7881,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.30039896737854965,
|
| 126 |
+
"grad_norm": 0.48944246768951416,
|
| 127 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 128 |
+
"loss": 2.7937,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.319173902839709,
|
| 133 |
+
"grad_norm": 0.5136508941650391,
|
| 134 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 135 |
+
"loss": 2.7524,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.33794883830086836,
|
| 140 |
+
"grad_norm": 0.48745179176330566,
|
| 141 |
+
"learning_rate": 3.6e-05,
|
| 142 |
+
"loss": 2.7675,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3567237737620277,
|
| 147 |
+
"grad_norm": 0.4948646128177643,
|
| 148 |
+
"learning_rate": 3.8e-05,
|
| 149 |
+
"loss": 2.731,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.37549870922318707,
|
| 154 |
+
"grad_norm": 0.5109882354736328,
|
| 155 |
+
"learning_rate": 4e-05,
|
| 156 |
+
"loss": 2.7235,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.37549870922318707,
|
| 161 |
+
"eval_loss": 2.6776111125946045,
|
| 162 |
+
"eval_runtime": 115.5701,
|
| 163 |
+
"eval_samples_per_second": 8.194,
|
| 164 |
+
"eval_steps_per_second": 4.101,
|
| 165 |
+
"step": 200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.3942736446843464,
|
| 169 |
+
"grad_norm": 0.5552621483802795,
|
| 170 |
+
"learning_rate": 4.2e-05,
|
| 171 |
+
"loss": 2.7244,
|
| 172 |
+
"step": 210
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.4130485801455058,
|
| 176 |
+
"grad_norm": 0.5443555116653442,
|
| 177 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 178 |
+
"loss": 2.6612,
|
| 179 |
+
"step": 220
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.4318235156066651,
|
| 183 |
+
"grad_norm": 0.6089721918106079,
|
| 184 |
+
"learning_rate": 4.600000000000001e-05,
|
| 185 |
+
"loss": 2.6635,
|
| 186 |
+
"step": 230
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.4505984510678245,
|
| 190 |
+
"grad_norm": 0.6026090383529663,
|
| 191 |
+
"learning_rate": 4.8e-05,
|
| 192 |
+
"loss": 2.6335,
|
| 193 |
+
"step": 240
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.4693733865289838,
|
| 197 |
+
"grad_norm": 0.6256532073020935,
|
| 198 |
+
"learning_rate": 5e-05,
|
| 199 |
+
"loss": 2.6262,
|
| 200 |
+
"step": 250
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.4881483219901432,
|
| 204 |
+
"grad_norm": 0.6782171726226807,
|
| 205 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 206 |
+
"loss": 2.665,
|
| 207 |
+
"step": 260
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.5069232574513025,
|
| 211 |
+
"grad_norm": 0.7474572658538818,
|
| 212 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 213 |
+
"loss": 2.7222,
|
| 214 |
+
"step": 270
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.5256981929124619,
|
| 218 |
+
"grad_norm": 0.6566171646118164,
|
| 219 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 220 |
+
"loss": 2.6626,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.5444731283736212,
|
| 225 |
+
"grad_norm": 0.7244133949279785,
|
| 226 |
+
"learning_rate": 5.8e-05,
|
| 227 |
+
"loss": 2.5429,
|
| 228 |
+
"step": 290
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.5632480638347805,
|
| 232 |
+
"grad_norm": 0.7084689140319824,
|
| 233 |
+
"learning_rate": 6e-05,
|
| 234 |
+
"loss": 2.5809,
|
| 235 |
+
"step": 300
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.5632480638347805,
|
| 239 |
+
"eval_loss": 2.5663962364196777,
|
| 240 |
+
"eval_runtime": 115.4961,
|
| 241 |
+
"eval_samples_per_second": 8.199,
|
| 242 |
+
"eval_steps_per_second": 4.104,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.5820229992959399,
|
| 247 |
+
"grad_norm": 0.7440884709358215,
|
| 248 |
+
"learning_rate": 6.2e-05,
|
| 249 |
+
"loss": 2.6225,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.6007979347570993,
|
| 254 |
+
"grad_norm": 1.0468900203704834,
|
| 255 |
+
"learning_rate": 6.400000000000001e-05,
|
| 256 |
+
"loss": 2.579,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.6195728702182586,
|
| 261 |
+
"grad_norm": 1.0507537126541138,
|
| 262 |
+
"learning_rate": 6.6e-05,
|
| 263 |
+
"loss": 2.6367,
|
| 264 |
+
"step": 330
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.638347805679418,
|
| 268 |
+
"grad_norm": 1.0535833835601807,
|
| 269 |
+
"learning_rate": 6.800000000000001e-05,
|
| 270 |
+
"loss": 2.556,
|
| 271 |
+
"step": 340
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.6571227411405773,
|
| 275 |
+
"grad_norm": 1.5066871643066406,
|
| 276 |
+
"learning_rate": 7e-05,
|
| 277 |
+
"loss": 2.5231,
|
| 278 |
+
"step": 350
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.6758976766017367,
|
| 282 |
+
"grad_norm": 1.1159995794296265,
|
| 283 |
+
"learning_rate": 7.2e-05,
|
| 284 |
+
"loss": 2.4532,
|
| 285 |
+
"step": 360
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.694672612062896,
|
| 289 |
+
"grad_norm": 1.4439431428909302,
|
| 290 |
+
"learning_rate": 7.4e-05,
|
| 291 |
+
"loss": 2.4436,
|
| 292 |
+
"step": 370
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.7134475475240554,
|
| 296 |
+
"grad_norm": 1.1311273574829102,
|
| 297 |
+
"learning_rate": 7.58e-05,
|
| 298 |
+
"loss": 2.4088,
|
| 299 |
+
"step": 380
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.7322224829852148,
|
| 303 |
+
"grad_norm": 0.9161147475242615,
|
| 304 |
+
"learning_rate": 7.780000000000001e-05,
|
| 305 |
+
"loss": 2.5574,
|
| 306 |
+
"step": 390
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.7509974184463741,
|
| 310 |
+
"grad_norm": 1.2713381052017212,
|
| 311 |
+
"learning_rate": 7.98e-05,
|
| 312 |
+
"loss": 2.3971,
|
| 313 |
+
"step": 400
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.7509974184463741,
|
| 317 |
+
"eval_loss": 2.4457809925079346,
|
| 318 |
+
"eval_runtime": 115.5729,
|
| 319 |
+
"eval_samples_per_second": 8.194,
|
| 320 |
+
"eval_steps_per_second": 4.101,
|
| 321 |
+
"step": 400
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.7697723539075334,
|
| 325 |
+
"grad_norm": 1.4076571464538574,
|
| 326 |
+
"learning_rate": 8.18e-05,
|
| 327 |
+
"loss": 2.5646,
|
| 328 |
+
"step": 410
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.7885472893686928,
|
| 332 |
+
"grad_norm": 1.0278912782669067,
|
| 333 |
+
"learning_rate": 8.38e-05,
|
| 334 |
+
"loss": 2.5097,
|
| 335 |
+
"step": 420
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.8073222248298522,
|
| 339 |
+
"grad_norm": 0.9688978791236877,
|
| 340 |
+
"learning_rate": 8.58e-05,
|
| 341 |
+
"loss": 2.4623,
|
| 342 |
+
"step": 430
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.8260971602910115,
|
| 346 |
+
"grad_norm": 1.3491131067276,
|
| 347 |
+
"learning_rate": 8.78e-05,
|
| 348 |
+
"loss": 2.508,
|
| 349 |
+
"step": 440
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.8448720957521708,
|
| 353 |
+
"grad_norm": 0.8669747710227966,
|
| 354 |
+
"learning_rate": 8.98e-05,
|
| 355 |
+
"loss": 2.518,
|
| 356 |
+
"step": 450
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.8636470312133302,
|
| 360 |
+
"grad_norm": 3.474540948867798,
|
| 361 |
+
"learning_rate": 9.180000000000001e-05,
|
| 362 |
+
"loss": 2.4558,
|
| 363 |
+
"step": 460
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.8824219666744896,
|
| 367 |
+
"grad_norm": 0.9799988865852356,
|
| 368 |
+
"learning_rate": 9.38e-05,
|
| 369 |
+
"loss": 2.3909,
|
| 370 |
+
"step": 470
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.901196902135649,
|
| 374 |
+
"grad_norm": 1.605360507965088,
|
| 375 |
+
"learning_rate": 9.58e-05,
|
| 376 |
+
"loss": 2.3931,
|
| 377 |
+
"step": 480
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.9199718375968082,
|
| 381 |
+
"grad_norm": 1.0437480211257935,
|
| 382 |
+
"learning_rate": 9.78e-05,
|
| 383 |
+
"loss": 2.3893,
|
| 384 |
+
"step": 490
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.9387467730579676,
|
| 388 |
+
"grad_norm": 0.7387896180152893,
|
| 389 |
+
"learning_rate": 9.98e-05,
|
| 390 |
+
"loss": 2.4147,
|
| 391 |
+
"step": 500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.9387467730579676,
|
| 395 |
+
"eval_loss": 2.381179094314575,
|
| 396 |
+
"eval_runtime": 115.6216,
|
| 397 |
+
"eval_samples_per_second": 8.191,
|
| 398 |
+
"eval_steps_per_second": 4.1,
|
| 399 |
+
"step": 500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.957521708519127,
|
| 403 |
+
"grad_norm": 1.0030367374420166,
|
| 404 |
+
"learning_rate": 9.998336282524578e-05,
|
| 405 |
+
"loss": 2.4192,
|
| 406 |
+
"step": 510
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.9762966439802864,
|
| 410 |
+
"grad_norm": 2.564517021179199,
|
| 411 |
+
"learning_rate": 9.992586581827853e-05,
|
| 412 |
+
"loss": 2.4279,
|
| 413 |
+
"step": 520
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.9950715794414456,
|
| 417 |
+
"grad_norm": 0.8353458642959595,
|
| 418 |
+
"learning_rate": 9.983904607709364e-05,
|
| 419 |
+
"loss": 2.4838,
|
| 420 |
+
"step": 530
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.013846514902605,
|
| 424 |
+
"grad_norm": 0.7976542115211487,
|
| 425 |
+
"learning_rate": 9.970368310819e-05,
|
| 426 |
+
"loss": 2.4366,
|
| 427 |
+
"step": 540
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.0326214503637643,
|
| 431 |
+
"grad_norm": 1.2872307300567627,
|
| 432 |
+
"learning_rate": 9.952748467735829e-05,
|
| 433 |
+
"loss": 2.3284,
|
| 434 |
+
"step": 550
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.0513963858249238,
|
| 438 |
+
"grad_norm": 1.1133558750152588,
|
| 439 |
+
"learning_rate": 9.931059554538611e-05,
|
| 440 |
+
"loss": 2.3192,
|
| 441 |
+
"step": 560
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.070171321286083,
|
| 445 |
+
"grad_norm": 1.0255626440048218,
|
| 446 |
+
"learning_rate": 9.905319390365365e-05,
|
| 447 |
+
"loss": 2.3409,
|
| 448 |
+
"step": 570
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.0889462567472425,
|
| 452 |
+
"grad_norm": 1.0734648704528809,
|
| 453 |
+
"learning_rate": 9.875549122773535e-05,
|
| 454 |
+
"loss": 2.4633,
|
| 455 |
+
"step": 580
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.1077211922084018,
|
| 459 |
+
"grad_norm": 0.9561296105384827,
|
| 460 |
+
"learning_rate": 9.841773210365646e-05,
|
| 461 |
+
"loss": 2.3144,
|
| 462 |
+
"step": 590
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.126496127669561,
|
| 466 |
+
"grad_norm": 0.8402953743934631,
|
| 467 |
+
"learning_rate": 9.804019402694626e-05,
|
| 468 |
+
"loss": 2.3987,
|
| 469 |
+
"step": 600
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.126496127669561,
|
| 473 |
+
"eval_loss": 2.343639850616455,
|
| 474 |
+
"eval_runtime": 115.6739,
|
| 475 |
+
"eval_samples_per_second": 8.187,
|
| 476 |
+
"eval_steps_per_second": 4.098,
|
| 477 |
+
"step": 600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.1452710631307206,
|
| 481 |
+
"grad_norm": 0.9632971882820129,
|
| 482 |
+
"learning_rate": 9.762318717465388e-05,
|
| 483 |
+
"loss": 2.4487,
|
| 484 |
+
"step": 610
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 1.1640459985918798,
|
| 488 |
+
"grad_norm": 0.996286153793335,
|
| 489 |
+
"learning_rate": 9.716705415051361e-05,
|
| 490 |
+
"loss": 2.5009,
|
| 491 |
+
"step": 620
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.1828209340530391,
|
| 495 |
+
"grad_norm": 0.9334232211112976,
|
| 496 |
+
"learning_rate": 9.667216970346914e-05,
|
| 497 |
+
"loss": 2.2991,
|
| 498 |
+
"step": 630
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.2015958695141986,
|
| 502 |
+
"grad_norm": 1.014267086982727,
|
| 503 |
+
"learning_rate": 9.613894041978795e-05,
|
| 504 |
+
"loss": 2.4963,
|
| 505 |
+
"step": 640
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.2203708049753579,
|
| 509 |
+
"grad_norm": 1.3219252824783325,
|
| 510 |
+
"learning_rate": 9.556780438901898e-05,
|
| 511 |
+
"loss": 2.4339,
|
| 512 |
+
"step": 650
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.2391457404365172,
|
| 516 |
+
"grad_norm": 0.7801995873451233,
|
| 517 |
+
"learning_rate": 9.495923084406772e-05,
|
| 518 |
+
"loss": 2.3548,
|
| 519 |
+
"step": 660
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.2579206758976766,
|
| 523 |
+
"grad_norm": 1.7439610958099365,
|
| 524 |
+
"learning_rate": 9.431371977568483e-05,
|
| 525 |
+
"loss": 2.4373,
|
| 526 |
+
"step": 670
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.276695611358836,
|
| 530 |
+
"grad_norm": 1.080696702003479,
|
| 531 |
+
"learning_rate": 9.363180152168447e-05,
|
| 532 |
+
"loss": 2.1908,
|
| 533 |
+
"step": 680
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.2954705468199954,
|
| 537 |
+
"grad_norm": 1.151071548461914,
|
| 538 |
+
"learning_rate": 9.291403633123046e-05,
|
| 539 |
+
"loss": 2.5268,
|
| 540 |
+
"step": 690
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.3142454822811547,
|
| 544 |
+
"grad_norm": 0.8707749843597412,
|
| 545 |
+
"learning_rate": 9.21610139045477e-05,
|
| 546 |
+
"loss": 2.3,
|
| 547 |
+
"step": 700
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.3142454822811547,
|
| 551 |
+
"eval_loss": 2.3193163871765137,
|
| 552 |
+
"eval_runtime": 115.797,
|
| 553 |
+
"eval_samples_per_second": 8.178,
|
| 554 |
+
"eval_steps_per_second": 4.093,
|
| 555 |
+
"step": 700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.333020417742314,
|
| 559 |
+
"grad_norm": 1.0154194831848145,
|
| 560 |
+
"learning_rate": 9.13733529084374e-05,
|
| 561 |
+
"loss": 2.3169,
|
| 562 |
+
"step": 710
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.3517953532034734,
|
| 566 |
+
"grad_norm": 1.2447744607925415,
|
| 567 |
+
"learning_rate": 9.055170046799385e-05,
|
| 568 |
+
"loss": 2.2876,
|
| 569 |
+
"step": 720
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.3705702886646327,
|
| 573 |
+
"grad_norm": 1.545040249824524,
|
| 574 |
+
"learning_rate": 8.969673163494064e-05,
|
| 575 |
+
"loss": 2.3832,
|
| 576 |
+
"step": 730
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.3893452241257922,
|
| 580 |
+
"grad_norm": 0.9809775352478027,
|
| 581 |
+
"learning_rate": 8.880914883302278e-05,
|
| 582 |
+
"loss": 2.4271,
|
| 583 |
+
"step": 740
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.4081201595869515,
|
| 587 |
+
"grad_norm": 0.8140939474105835,
|
| 588 |
+
"learning_rate": 8.788968128091083e-05,
|
| 589 |
+
"loss": 2.426,
|
| 590 |
+
"step": 750
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.4268950950481107,
|
| 594 |
+
"grad_norm": 0.9135342240333557,
|
| 595 |
+
"learning_rate": 8.69390843930906e-05,
|
| 596 |
+
"loss": 2.3396,
|
| 597 |
+
"step": 760
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.44567003050927,
|
| 601 |
+
"grad_norm": 0.7756408452987671,
|
| 602 |
+
"learning_rate": 8.595813915923113e-05,
|
| 603 |
+
"loss": 2.3238,
|
| 604 |
+
"step": 770
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.4644449659704295,
|
| 608 |
+
"grad_norm": 0.8543408513069153,
|
| 609 |
+
"learning_rate": 8.494765150254062e-05,
|
| 610 |
+
"loss": 2.509,
|
| 611 |
+
"step": 780
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.4832199014315888,
|
| 615 |
+
"grad_norm": 0.8993504643440247,
|
| 616 |
+
"learning_rate": 8.390845161763756e-05,
|
| 617 |
+
"loss": 2.1994,
|
| 618 |
+
"step": 790
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.5019948368927483,
|
| 622 |
+
"grad_norm": 1.2262345552444458,
|
| 623 |
+
"learning_rate": 8.284139328848082e-05,
|
| 624 |
+
"loss": 2.3219,
|
| 625 |
+
"step": 800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.5019948368927483,
|
| 629 |
+
"eval_loss": 2.295060873031616,
|
| 630 |
+
"eval_runtime": 115.6726,
|
| 631 |
+
"eval_samples_per_second": 8.187,
|
| 632 |
+
"eval_steps_per_second": 4.098,
|
| 633 |
+
"step": 800
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.5207697723539075,
|
| 637 |
+
"grad_norm": 0.8056873679161072,
|
| 638 |
+
"learning_rate": 8.174735318691945e-05,
|
| 639 |
+
"loss": 2.4357,
|
| 640 |
+
"step": 810
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.5395447078150668,
|
| 644 |
+
"grad_norm": 0.9234392642974854,
|
| 645 |
+
"learning_rate": 8.062723015243822e-05,
|
| 646 |
+
"loss": 2.172,
|
| 647 |
+
"step": 820
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.558319643276226,
|
| 651 |
+
"grad_norm": 1.4100919961929321,
|
| 652 |
+
"learning_rate": 7.948194445369064e-05,
|
| 653 |
+
"loss": 2.4166,
|
| 654 |
+
"step": 830
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.5770945787373856,
|
| 658 |
+
"grad_norm": 0.9078715443611145,
|
| 659 |
+
"learning_rate": 7.831243703242637e-05,
|
| 660 |
+
"loss": 2.3765,
|
| 661 |
+
"step": 840
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.595869514198545,
|
| 665 |
+
"grad_norm": 1.0631110668182373,
|
| 666 |
+
"learning_rate": 7.711966873043398e-05,
|
| 667 |
+
"loss": 2.3603,
|
| 668 |
+
"step": 850
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.6146444496597043,
|
| 672 |
+
"grad_norm": 0.8904982805252075,
|
| 673 |
+
"learning_rate": 7.590461950013424e-05,
|
| 674 |
+
"loss": 2.2791,
|
| 675 |
+
"step": 860
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.6334193851208636,
|
| 679 |
+
"grad_norm": 1.1240772008895874,
|
| 680 |
+
"learning_rate": 7.46682875994727e-05,
|
| 681 |
+
"loss": 2.4014,
|
| 682 |
+
"step": 870
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.6521943205820229,
|
| 686 |
+
"grad_norm": 0.9138346314430237,
|
| 687 |
+
"learning_rate": 7.341168877177267e-05,
|
| 688 |
+
"loss": 2.3897,
|
| 689 |
+
"step": 880
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.6709692560431824,
|
| 693 |
+
"grad_norm": 0.8400964140892029,
|
| 694 |
+
"learning_rate": 7.213585541122261e-05,
|
| 695 |
+
"loss": 2.3004,
|
| 696 |
+
"step": 890
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.6897441915043419,
|
| 700 |
+
"grad_norm": 1.1160262823104858,
|
| 701 |
+
"learning_rate": 7.084183571468368e-05,
|
| 702 |
+
"loss": 2.377,
|
| 703 |
+
"step": 900
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.6897441915043419,
|
| 707 |
+
"eval_loss": 2.2762601375579834,
|
| 708 |
+
"eval_runtime": 115.7128,
|
| 709 |
+
"eval_samples_per_second": 8.184,
|
| 710 |
+
"eval_steps_per_second": 4.096,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.7085191269655011,
|
| 715 |
+
"grad_norm": 0.8896321654319763,
|
| 716 |
+
"learning_rate": 6.953069282051396e-05,
|
| 717 |
+
"loss": 2.1619,
|
| 718 |
+
"step": 910
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.7272940624266604,
|
| 722 |
+
"grad_norm": 0.848534107208252,
|
| 723 |
+
"learning_rate": 6.820350393511732e-05,
|
| 724 |
+
"loss": 2.447,
|
| 725 |
+
"step": 920
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.7460689978878197,
|
| 729 |
+
"grad_norm": 0.9254674911499023,
|
| 730 |
+
"learning_rate": 6.686135944793395e-05,
|
| 731 |
+
"loss": 2.2199,
|
| 732 |
+
"step": 930
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.7648439333489792,
|
| 736 |
+
"grad_norm": 0.8377211093902588,
|
| 737 |
+
"learning_rate": 6.550536203560029e-05,
|
| 738 |
+
"loss": 2.3059,
|
| 739 |
+
"step": 940
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.7836188688101384,
|
| 743 |
+
"grad_norm": 1.01736319065094,
|
| 744 |
+
"learning_rate": 6.413662575601391e-05,
|
| 745 |
+
"loss": 2.1205,
|
| 746 |
+
"step": 950
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.802393804271298,
|
| 750 |
+
"grad_norm": 0.7986969947814941,
|
| 751 |
+
"learning_rate": 6.27562751330479e-05,
|
| 752 |
+
"loss": 2.2619,
|
| 753 |
+
"step": 960
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.8211687397324572,
|
| 757 |
+
"grad_norm": 1.054882287979126,
|
| 758 |
+
"learning_rate": 6.136544423266651e-05,
|
| 759 |
+
"loss": 2.1804,
|
| 760 |
+
"step": 970
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.8399436751936165,
|
| 764 |
+
"grad_norm": 0.8551494479179382,
|
| 765 |
+
"learning_rate": 5.9965275731201366e-05,
|
| 766 |
+
"loss": 2.3119,
|
| 767 |
+
"step": 980
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.8587186106547757,
|
| 771 |
+
"grad_norm": 1.810271143913269,
|
| 772 |
+
"learning_rate": 5.8556919976553406e-05,
|
| 773 |
+
"loss": 2.4647,
|
| 774 |
+
"step": 990
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.8774935461159352,
|
| 778 |
+
"grad_norm": 0.7847923636436462,
|
| 779 |
+
"learning_rate": 5.714153404309228e-05,
|
| 780 |
+
"loss": 2.2977,
|
| 781 |
+
"step": 1000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.8774935461159352,
|
| 785 |
+
"eval_loss": 2.262303590774536,
|
| 786 |
+
"eval_runtime": 115.6317,
|
| 787 |
+
"eval_samples_per_second": 8.19,
|
| 788 |
+
"eval_steps_per_second": 4.099,
|
| 789 |
+
"step": 1000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.8962684815770947,
|
| 793 |
+
"grad_norm": 0.8150405883789062,
|
| 794 |
+
"learning_rate": 5.5720280781029166e-05,
|
| 795 |
+
"loss": 2.2873,
|
| 796 |
+
"step": 1010
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.915043417038254,
|
| 800 |
+
"grad_norm": 0.7990459203720093,
|
| 801 |
+
"learning_rate": 5.429432786104446e-05,
|
| 802 |
+
"loss": 2.3096,
|
| 803 |
+
"step": 1020
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 1.9338183524994133,
|
| 807 |
+
"grad_norm": 0.9583683013916016,
|
| 808 |
+
"learning_rate": 5.286484681495499e-05,
|
| 809 |
+
"loss": 2.3133,
|
| 810 |
+
"step": 1030
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 1.9525932879605725,
|
| 814 |
+
"grad_norm": 0.8717370629310608,
|
| 815 |
+
"learning_rate": 5.1433012073209085e-05,
|
| 816 |
+
"loss": 2.3946,
|
| 817 |
+
"step": 1040
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.971368223421732,
|
| 821 |
+
"grad_norm": 0.8073247671127319,
|
| 822 |
+
"learning_rate": 5e-05,
|
| 823 |
+
"loss": 2.1941,
|
| 824 |
+
"step": 1050
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.9901431588828915,
|
| 828 |
+
"grad_norm": 1.0495169162750244,
|
| 829 |
+
"learning_rate": 4.856698792679094e-05,
|
| 830 |
+
"loss": 2.2465,
|
| 831 |
+
"step": 1060
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.0089180943440508,
|
| 835 |
+
"grad_norm": 0.8290483355522156,
|
| 836 |
+
"learning_rate": 4.7135153185045014e-05,
|
| 837 |
+
"loss": 2.2935,
|
| 838 |
+
"step": 1070
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.02769302980521,
|
| 842 |
+
"grad_norm": 0.8408583998680115,
|
| 843 |
+
"learning_rate": 4.570567213895555e-05,
|
| 844 |
+
"loss": 2.2642,
|
| 845 |
+
"step": 1080
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.0464679652663693,
|
| 849 |
+
"grad_norm": 1.1089783906936646,
|
| 850 |
+
"learning_rate": 4.427971921897085e-05,
|
| 851 |
+
"loss": 2.3274,
|
| 852 |
+
"step": 1090
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 2.0652429007275286,
|
| 856 |
+
"grad_norm": 1.12042236328125,
|
| 857 |
+
"learning_rate": 4.2858465956907724e-05,
|
| 858 |
+
"loss": 2.269,
|
| 859 |
+
"step": 1100
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 2.0652429007275286,
|
| 863 |
+
"eval_loss": 2.252450704574585,
|
| 864 |
+
"eval_runtime": 115.7991,
|
| 865 |
+
"eval_samples_per_second": 8.178,
|
| 866 |
+
"eval_steps_per_second": 4.093,
|
| 867 |
+
"step": 1100
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.0840178361886883,
|
| 871 |
+
"grad_norm": 0.9814222455024719,
|
| 872 |
+
"learning_rate": 4.14430800234466e-05,
|
| 873 |
+
"loss": 2.1596,
|
| 874 |
+
"step": 1110
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.1027927716498476,
|
| 878 |
+
"grad_norm": 0.8501948118209839,
|
| 879 |
+
"learning_rate": 4.003472426879866e-05,
|
| 880 |
+
"loss": 2.3033,
|
| 881 |
+
"step": 1120
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.121567707111007,
|
| 885 |
+
"grad_norm": 0.8352840542793274,
|
| 886 |
+
"learning_rate": 3.863455576733349e-05,
|
| 887 |
+
"loss": 2.2893,
|
| 888 |
+
"step": 1130
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.140342642572166,
|
| 892 |
+
"grad_norm": 0.8014605045318604,
|
| 893 |
+
"learning_rate": 3.724372486695211e-05,
|
| 894 |
+
"loss": 2.2274,
|
| 895 |
+
"step": 1140
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.1591175780333254,
|
| 899 |
+
"grad_norm": 0.9311611652374268,
|
| 900 |
+
"learning_rate": 3.5863374243986094e-05,
|
| 901 |
+
"loss": 2.0715,
|
| 902 |
+
"step": 1150
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.177892513494485,
|
| 906 |
+
"grad_norm": 0.8880860209465027,
|
| 907 |
+
"learning_rate": 3.4494637964399724e-05,
|
| 908 |
+
"loss": 2.2517,
|
| 909 |
+
"step": 1160
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.1966674489556444,
|
| 913 |
+
"grad_norm": 0.9553248882293701,
|
| 914 |
+
"learning_rate": 3.313864055206607e-05,
|
| 915 |
+
"loss": 2.3358,
|
| 916 |
+
"step": 1170
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 2.2154423844168036,
|
| 920 |
+
"grad_norm": 1.015997052192688,
|
| 921 |
+
"learning_rate": 3.179649606488267e-05,
|
| 922 |
+
"loss": 2.2711,
|
| 923 |
+
"step": 1180
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.234217319877963,
|
| 927 |
+
"grad_norm": 0.8522908091545105,
|
| 928 |
+
"learning_rate": 3.046930717948604e-05,
|
| 929 |
+
"loss": 2.2483,
|
| 930 |
+
"step": 1190
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 2.252992255339122,
|
| 934 |
+
"grad_norm": 0.844487190246582,
|
| 935 |
+
"learning_rate": 2.9158164285316353e-05,
|
| 936 |
+
"loss": 2.2305,
|
| 937 |
+
"step": 1200
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.252992255339122,
|
| 941 |
+
"eval_loss": 2.2441632747650146,
|
| 942 |
+
"eval_runtime": 115.7498,
|
| 943 |
+
"eval_samples_per_second": 8.181,
|
| 944 |
+
"eval_steps_per_second": 4.095,
|
| 945 |
+
"step": 1200
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 2.2717671908002814,
|
| 949 |
+
"grad_norm": 0.87571120262146,
|
| 950 |
+
"learning_rate": 2.7864144588777403e-05,
|
| 951 |
+
"loss": 2.2648,
|
| 952 |
+
"step": 1210
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 2.290542126261441,
|
| 956 |
+
"grad_norm": 0.8913039565086365,
|
| 957 |
+
"learning_rate": 2.6588311228227347e-05,
|
| 958 |
+
"loss": 2.1812,
|
| 959 |
+
"step": 1220
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 2.3093170617226004,
|
| 963 |
+
"grad_norm": 1.0479601621627808,
|
| 964 |
+
"learning_rate": 2.5331712400527298e-05,
|
| 965 |
+
"loss": 2.2898,
|
| 966 |
+
"step": 1230
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 2.3280919971837597,
|
| 970 |
+
"grad_norm": 0.9554978013038635,
|
| 971 |
+
"learning_rate": 2.4095380499865762e-05,
|
| 972 |
+
"loss": 2.3384,
|
| 973 |
+
"step": 1240
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.346866932644919,
|
| 977 |
+
"grad_norm": 0.8619790077209473,
|
| 978 |
+
"learning_rate": 2.288033126956604e-05,
|
| 979 |
+
"loss": 2.2454,
|
| 980 |
+
"step": 1250
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.3656418681060782,
|
| 984 |
+
"grad_norm": 0.980107843875885,
|
| 985 |
+
"learning_rate": 2.1687562967573645e-05,
|
| 986 |
+
"loss": 2.193,
|
| 987 |
+
"step": 1260
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 2.3844168035672375,
|
| 991 |
+
"grad_norm": 0.935689389705658,
|
| 992 |
+
"learning_rate": 2.0518055546309362e-05,
|
| 993 |
+
"loss": 2.3337,
|
| 994 |
+
"step": 1270
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 2.4031917390283972,
|
| 998 |
+
"grad_norm": 0.8614078164100647,
|
| 999 |
+
"learning_rate": 1.9372769847561788e-05,
|
| 1000 |
+
"loss": 2.1658,
|
| 1001 |
+
"step": 1280
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 2.4219666744895565,
|
| 1005 |
+
"grad_norm": 0.9031429886817932,
|
| 1006 |
+
"learning_rate": 1.8252646813080565e-05,
|
| 1007 |
+
"loss": 2.1985,
|
| 1008 |
+
"step": 1290
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 2.4407416099507158,
|
| 1012 |
+
"grad_norm": 0.9556667804718018,
|
| 1013 |
+
"learning_rate": 1.7158606711519194e-05,
|
| 1014 |
+
"loss": 2.3866,
|
| 1015 |
+
"step": 1300
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 2.4407416099507158,
|
| 1019 |
+
"eval_loss": 2.2396435737609863,
|
| 1020 |
+
"eval_runtime": 115.863,
|
| 1021 |
+
"eval_samples_per_second": 8.173,
|
| 1022 |
+
"eval_steps_per_second": 4.091,
|
| 1023 |
+
"step": 1300
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 2.459516545411875,
|
| 1027 |
+
"grad_norm": 0.9352513551712036,
|
| 1028 |
+
"learning_rate": 1.6091548382362458e-05,
|
| 1029 |
+
"loss": 2.2399,
|
| 1030 |
+
"step": 1310
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 2.4782914808730343,
|
| 1034 |
+
"grad_norm": 1.494532585144043,
|
| 1035 |
+
"learning_rate": 1.50523484974594e-05,
|
| 1036 |
+
"loss": 2.3273,
|
| 1037 |
+
"step": 1320
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.497066416334194,
|
| 1041 |
+
"grad_norm": 0.9380200505256653,
|
| 1042 |
+
"learning_rate": 1.40418608407689e-05,
|
| 1043 |
+
"loss": 2.2071,
|
| 1044 |
+
"step": 1330
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.5158413517953533,
|
| 1048 |
+
"grad_norm": 0.8295271396636963,
|
| 1049 |
+
"learning_rate": 1.3060915606909413e-05,
|
| 1050 |
+
"loss": 2.274,
|
| 1051 |
+
"step": 1340
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 2.5346162872565126,
|
| 1055 |
+
"grad_norm": 0.8793519735336304,
|
| 1056 |
+
"learning_rate": 1.2110318719089158e-05,
|
| 1057 |
+
"loss": 2.3791,
|
| 1058 |
+
"step": 1350
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.553391222717672,
|
| 1062 |
+
"grad_norm": 0.9659551382064819,
|
| 1063 |
+
"learning_rate": 1.1190851166977217e-05,
|
| 1064 |
+
"loss": 2.2612,
|
| 1065 |
+
"step": 1360
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.572166158178831,
|
| 1069 |
+
"grad_norm": 0.9426782727241516,
|
| 1070 |
+
"learning_rate": 1.0303268365059382e-05,
|
| 1071 |
+
"loss": 2.2907,
|
| 1072 |
+
"step": 1370
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 2.590941093639991,
|
| 1076 |
+
"grad_norm": 1.0491048097610474,
|
| 1077 |
+
"learning_rate": 9.448299532006149e-06,
|
| 1078 |
+
"loss": 2.3124,
|
| 1079 |
+
"step": 1380
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.60971602910115,
|
| 1083 |
+
"grad_norm": 0.9064353704452515,
|
| 1084 |
+
"learning_rate": 8.626647091562612e-06,
|
| 1085 |
+
"loss": 2.2214,
|
| 1086 |
+
"step": 1390
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 2.6284909645623094,
|
| 1090 |
+
"grad_norm": 0.7819197773933411,
|
| 1091 |
+
"learning_rate": 7.838986095452311e-06,
|
| 1092 |
+
"loss": 2.3217,
|
| 1093 |
+
"step": 1400
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 2.6284909645623094,
|
| 1097 |
+
"eval_loss": 2.2368621826171875,
|
| 1098 |
+
"eval_runtime": 115.8127,
|
| 1099 |
+
"eval_samples_per_second": 8.177,
|
| 1100 |
+
"eval_steps_per_second": 4.093,
|
| 1101 |
+
"step": 1400
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.6472659000234686,
|
| 1105 |
+
"grad_norm": 0.8173615336418152,
|
| 1106 |
+
"learning_rate": 7.085963668769552e-06,
|
| 1107 |
+
"loss": 2.2193,
|
| 1108 |
+
"step": 1410
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.666040835484628,
|
| 1112 |
+
"grad_norm": 0.7969537377357483,
|
| 1113 |
+
"learning_rate": 6.36819847831554e-06,
|
| 1114 |
+
"loss": 2.3388,
|
| 1115 |
+
"step": 1420
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.6848157709457876,
|
| 1119 |
+
"grad_norm": 0.8578535318374634,
|
| 1120 |
+
"learning_rate": 5.686280224315188e-06,
|
| 1121 |
+
"loss": 2.3577,
|
| 1122 |
+
"step": 1430
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.703590706406947,
|
| 1126 |
+
"grad_norm": 0.8416107296943665,
|
| 1127 |
+
"learning_rate": 5.040769155932284e-06,
|
| 1128 |
+
"loss": 2.2829,
|
| 1129 |
+
"step": 1440
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.722365641868106,
|
| 1133 |
+
"grad_norm": 0.8079660534858704,
|
| 1134 |
+
"learning_rate": 4.432195610981032e-06,
|
| 1135 |
+
"loss": 2.2675,
|
| 1136 |
+
"step": 1450
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.7411405773292654,
|
| 1140 |
+
"grad_norm": 0.8975071907043457,
|
| 1141 |
+
"learning_rate": 3.861059580212056e-06,
|
| 1142 |
+
"loss": 2.3367,
|
| 1143 |
+
"step": 1460
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.7599155127904247,
|
| 1147 |
+
"grad_norm": 1.0292245149612427,
|
| 1148 |
+
"learning_rate": 3.3278302965308596e-06,
|
| 1149 |
+
"loss": 2.3209,
|
| 1150 |
+
"step": 1470
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.7786904482515844,
|
| 1154 |
+
"grad_norm": 0.7877647280693054,
|
| 1155 |
+
"learning_rate": 2.8329458494863847e-06,
|
| 1156 |
+
"loss": 2.2889,
|
| 1157 |
+
"step": 1480
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.7974653837127432,
|
| 1161 |
+
"grad_norm": 0.8482506275177002,
|
| 1162 |
+
"learning_rate": 2.3768128253461253e-06,
|
| 1163 |
+
"loss": 2.3301,
|
| 1164 |
+
"step": 1490
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.816240319173903,
|
| 1168 |
+
"grad_norm": 0.7603471279144287,
|
| 1169 |
+
"learning_rate": 1.9598059730537466e-06,
|
| 1170 |
+
"loss": 2.2007,
|
| 1171 |
+
"step": 1500
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.816240319173903,
|
| 1175 |
+
"eval_loss": 2.2355196475982666,
|
| 1176 |
+
"eval_runtime": 115.7546,
|
| 1177 |
+
"eval_samples_per_second": 8.181,
|
| 1178 |
+
"eval_steps_per_second": 4.095,
|
| 1179 |
+
"step": 1500
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.835015254635062,
|
| 1183 |
+
"grad_norm": 0.8267006874084473,
|
| 1184 |
+
"learning_rate": 1.5822678963435478e-06,
|
| 1185 |
+
"loss": 2.392,
|
| 1186 |
+
"step": 1510
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 2.8537901900962215,
|
| 1190 |
+
"grad_norm": 0.8282234072685242,
|
| 1191 |
+
"learning_rate": 1.2445087722646575e-06,
|
| 1192 |
+
"loss": 2.2968,
|
| 1193 |
+
"step": 1520
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 2.8725651255573807,
|
| 1197 |
+
"grad_norm": 0.8622906804084778,
|
| 1198 |
+
"learning_rate": 9.468060963463755e-07,
|
| 1199 |
+
"loss": 2.2902,
|
| 1200 |
+
"step": 1530
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 2.89134006101854,
|
| 1204 |
+
"grad_norm": 0.8282898664474487,
|
| 1205 |
+
"learning_rate": 6.894044546138845e-07,
|
| 1206 |
+
"loss": 2.2151,
|
| 1207 |
+
"step": 1540
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 2.9101149964796997,
|
| 1211 |
+
"grad_norm": 0.8747497797012329,
|
| 1212 |
+
"learning_rate": 4.7251532264170895e-07,
|
| 1213 |
+
"loss": 2.188,
|
| 1214 |
+
"step": 1550
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 2.928889931940859,
|
| 1218 |
+
"grad_norm": 0.96438068151474,
|
| 1219 |
+
"learning_rate": 2.9631689180999457e-07,
|
| 1220 |
+
"loss": 2.4261,
|
| 1221 |
+
"step": 1560
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 2.9476648674020183,
|
| 1225 |
+
"grad_norm": 0.8510390520095825,
|
| 1226 |
+
"learning_rate": 1.6095392290635393e-07,
|
| 1227 |
+
"loss": 2.272,
|
| 1228 |
+
"step": 1570
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 2.9664398028631775,
|
| 1232 |
+
"grad_norm": 0.7817071080207825,
|
| 1233 |
+
"learning_rate": 6.653762719355805e-08,
|
| 1234 |
+
"loss": 2.3397,
|
| 1235 |
+
"step": 1580
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 2.985214738324337,
|
| 1239 |
+
"grad_norm": 0.8862946033477783,
|
| 1240 |
+
"learning_rate": 1.3145575040801606e-08,
|
| 1241 |
+
"loss": 2.269,
|
| 1242 |
+
"step": 1590
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 2.9964796996010326,
|
| 1246 |
+
"step": 1596,
|
| 1247 |
+
"total_flos": 1.1780062420402176e+18,
|
| 1248 |
+
"train_loss": 2.428264606566656,
|
| 1249 |
+
"train_runtime": 11538.2228,
|
| 1250 |
+
"train_samples_per_second": 2.216,
|
| 1251 |
+
"train_steps_per_second": 0.138
|
| 1252 |
+
}
|
| 1253 |
+
],
|
| 1254 |
+
"logging_steps": 10,
|
| 1255 |
+
"max_steps": 1596,
|
| 1256 |
+
"num_input_tokens_seen": 0,
|
| 1257 |
+
"num_train_epochs": 3,
|
| 1258 |
+
"save_steps": 500,
|
| 1259 |
+
"total_flos": 1.1780062420402176e+18,
|
| 1260 |
+
"train_batch_size": 1,
|
| 1261 |
+
"trial_name": null,
|
| 1262 |
+
"trial_params": null
|
| 1263 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|