End of training
Browse files- README.md +3 -2
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1120 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -1,9 +1,10 @@
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
|
|
|
| 3 |
tags:
|
| 4 |
- llama-factory
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
-
base_model: Llama-3.1-70B-Instruct
|
| 7 |
model-index:
|
| 8 |
- name: Llama-3.1-70B-Instruct-fine-llamafactory-lora
|
| 9 |
results: []
|
|
@@ -14,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# Llama-3.1-70B-Instruct-fine-llamafactory-lora
|
| 16 |
|
| 17 |
-
This model
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
+
base_model: Llama-3.1-70B-Instruct
|
| 4 |
tags:
|
| 5 |
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
- generated_from_trainer
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: Llama-3.1-70B-Instruct-fine-llamafactory-lora
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# Llama-3.1-70B-Instruct-fine-llamafactory-lora
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [Llama-3.1-70B-Instruct](https://huggingface.co/Llama-3.1-70B-Instruct) on the sample_1 dataset.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 145121340096512.0,
|
| 4 |
+
"train_loss": 0.2980368115104638,
|
| 5 |
+
"train_runtime": 2252.7177,
|
| 6 |
+
"train_samples_per_second": 0.272,
|
| 7 |
+
"train_steps_per_second": 0.068
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 145121340096512.0,
|
| 4 |
+
"train_loss": 0.2980368115104638,
|
| 5 |
+
"train_runtime": 2252.7177,
|
| 6 |
+
"train_samples_per_second": 0.272,
|
| 7 |
+
"train_steps_per_second": 0.068
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1120 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 154,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.012987012987012988,
|
| 13 |
+
"grad_norm": 0.18787174345625757,
|
| 14 |
+
"learning_rate": 5e-05,
|
| 15 |
+
"loss": 0.8483,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.025974025974025976,
|
| 20 |
+
"grad_norm": 0.1881876180971173,
|
| 21 |
+
"learning_rate": 4.999472998758978e-05,
|
| 22 |
+
"loss": 0.8274,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03896103896103896,
|
| 27 |
+
"grad_norm": 0.2484116599817621,
|
| 28 |
+
"learning_rate": 4.99789221722016e-05,
|
| 29 |
+
"loss": 0.9658,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05194805194805195,
|
| 34 |
+
"grad_norm": 0.2687834599860072,
|
| 35 |
+
"learning_rate": 4.995258321842611e-05,
|
| 36 |
+
"loss": 0.9686,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06493506493506493,
|
| 41 |
+
"grad_norm": 0.14459778376590973,
|
| 42 |
+
"learning_rate": 4.991572423079236e-05,
|
| 43 |
+
"loss": 0.8246,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07792207792207792,
|
| 48 |
+
"grad_norm": 0.12789380689263963,
|
| 49 |
+
"learning_rate": 4.986836074908616e-05,
|
| 50 |
+
"loss": 0.8481,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09090909090909091,
|
| 55 |
+
"grad_norm": 0.1241999521840537,
|
| 56 |
+
"learning_rate": 4.98105127417984e-05,
|
| 57 |
+
"loss": 0.8287,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1038961038961039,
|
| 62 |
+
"grad_norm": 0.12124312827410104,
|
| 63 |
+
"learning_rate": 4.974220459770639e-05,
|
| 64 |
+
"loss": 0.7255,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11688311688311688,
|
| 69 |
+
"grad_norm": 0.10261829314704707,
|
| 70 |
+
"learning_rate": 4.966346511559149e-05,
|
| 71 |
+
"loss": 0.8029,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.12987012987012986,
|
| 76 |
+
"grad_norm": 0.11018416242062518,
|
| 77 |
+
"learning_rate": 4.957432749209755e-05,
|
| 78 |
+
"loss": 0.7758,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14285714285714285,
|
| 83 |
+
"grad_norm": 0.15464498064215176,
|
| 84 |
+
"learning_rate": 4.9474829307735115e-05,
|
| 85 |
+
"loss": 0.738,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15584415584415584,
|
| 90 |
+
"grad_norm": 0.128517721396635,
|
| 91 |
+
"learning_rate": 4.9365012511037514e-05,
|
| 92 |
+
"loss": 0.7488,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16883116883116883,
|
| 97 |
+
"grad_norm": 0.16896802409788272,
|
| 98 |
+
"learning_rate": 4.9244923400875245e-05,
|
| 99 |
+
"loss": 0.7788,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.18181818181818182,
|
| 104 |
+
"grad_norm": 0.11228565989870234,
|
| 105 |
+
"learning_rate": 4.911461260693638e-05,
|
| 106 |
+
"loss": 0.6975,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.19480519480519481,
|
| 111 |
+
"grad_norm": 0.11221173687008401,
|
| 112 |
+
"learning_rate": 4.8974135068381036e-05,
|
| 113 |
+
"loss": 0.6526,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2077922077922078,
|
| 118 |
+
"grad_norm": 0.1282885413456516,
|
| 119 |
+
"learning_rate": 4.882355001067892e-05,
|
| 120 |
+
"loss": 0.7019,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.22077922077922077,
|
| 125 |
+
"grad_norm": 0.13827934971027034,
|
| 126 |
+
"learning_rate": 4.8662920920639866e-05,
|
| 127 |
+
"loss": 0.6399,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23376623376623376,
|
| 132 |
+
"grad_norm": 0.14252803580500878,
|
| 133 |
+
"learning_rate": 4.849231551964771e-05,
|
| 134 |
+
"loss": 0.5874,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.24675324675324675,
|
| 139 |
+
"grad_norm": 0.1567378099302403,
|
| 140 |
+
"learning_rate": 4.8311805735108894e-05,
|
| 141 |
+
"loss": 0.5871,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2597402597402597,
|
| 146 |
+
"grad_norm": 0.18892688129002116,
|
| 147 |
+
"learning_rate": 4.81214676701278e-05,
|
| 148 |
+
"loss": 0.5562,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2727272727272727,
|
| 153 |
+
"grad_norm": 0.3374448073807016,
|
| 154 |
+
"learning_rate": 4.792138157142158e-05,
|
| 155 |
+
"loss": 0.5319,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2857142857142857,
|
| 160 |
+
"grad_norm": 0.25712355804452103,
|
| 161 |
+
"learning_rate": 4.7711631795488096e-05,
|
| 162 |
+
"loss": 0.4744,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2987012987012987,
|
| 167 |
+
"grad_norm": 0.24700691510131254,
|
| 168 |
+
"learning_rate": 4.749230677304114e-05,
|
| 169 |
+
"loss": 0.444,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3116883116883117,
|
| 174 |
+
"grad_norm": 0.22489370668384237,
|
| 175 |
+
"learning_rate": 4.726349897172791e-05,
|
| 176 |
+
"loss": 0.3836,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.3246753246753247,
|
| 181 |
+
"grad_norm": 0.19240095528864964,
|
| 182 |
+
"learning_rate": 4.702530485714461e-05,
|
| 183 |
+
"loss": 0.3523,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.33766233766233766,
|
| 188 |
+
"grad_norm": 0.2888062734923711,
|
| 189 |
+
"learning_rate": 4.677782485216644e-05,
|
| 190 |
+
"loss": 0.3246,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.35064935064935066,
|
| 195 |
+
"grad_norm": 0.1578459013655223,
|
| 196 |
+
"learning_rate": 4.6521163294609196e-05,
|
| 197 |
+
"loss": 0.2807,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.36363636363636365,
|
| 202 |
+
"grad_norm": 0.17049112601293237,
|
| 203 |
+
"learning_rate": 4.625542839324036e-05,
|
| 204 |
+
"loss": 0.3049,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.37662337662337664,
|
| 209 |
+
"grad_norm": 0.10623476540468543,
|
| 210 |
+
"learning_rate": 4.598073218215817e-05,
|
| 211 |
+
"loss": 0.2877,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.38961038961038963,
|
| 216 |
+
"grad_norm": 0.08997970383090283,
|
| 217 |
+
"learning_rate": 4.5697190473557946e-05,
|
| 218 |
+
"loss": 0.2773,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4025974025974026,
|
| 223 |
+
"grad_norm": 0.11033930425860525,
|
| 224 |
+
"learning_rate": 4.540492280890555e-05,
|
| 225 |
+
"loss": 0.2758,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4155844155844156,
|
| 230 |
+
"grad_norm": 0.12378835216257449,
|
| 231 |
+
"learning_rate": 4.510405240853854e-05,
|
| 232 |
+
"loss": 0.2389,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.42857142857142855,
|
| 237 |
+
"grad_norm": 0.0810759730541574,
|
| 238 |
+
"learning_rate": 4.4794706119716455e-05,
|
| 239 |
+
"loss": 0.2756,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.44155844155844154,
|
| 244 |
+
"grad_norm": 0.09746550952897874,
|
| 245 |
+
"learning_rate": 4.447701436314176e-05,
|
| 246 |
+
"loss": 0.2836,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.45454545454545453,
|
| 251 |
+
"grad_norm": 0.08835165902502037,
|
| 252 |
+
"learning_rate": 4.415111107797445e-05,
|
| 253 |
+
"loss": 0.2363,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4675324675324675,
|
| 258 |
+
"grad_norm": 0.09871638815546385,
|
| 259 |
+
"learning_rate": 4.381713366536311e-05,
|
| 260 |
+
"loss": 0.261,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4805194805194805,
|
| 265 |
+
"grad_norm": 0.09771703372866393,
|
| 266 |
+
"learning_rate": 4.347522293051648e-05,
|
| 267 |
+
"loss": 0.2328,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4935064935064935,
|
| 272 |
+
"grad_norm": 0.08303169266120812,
|
| 273 |
+
"learning_rate": 4.312552302333982e-05,
|
| 274 |
+
"loss": 0.2669,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5064935064935064,
|
| 279 |
+
"grad_norm": 0.09110609687207077,
|
| 280 |
+
"learning_rate": 4.276818137766118e-05,
|
| 281 |
+
"loss": 0.2493,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5194805194805194,
|
| 286 |
+
"grad_norm": 0.0857887848090492,
|
| 287 |
+
"learning_rate": 4.2403348649073174e-05,
|
| 288 |
+
"loss": 0.256,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5324675324675324,
|
| 293 |
+
"grad_norm": 0.07992881618393588,
|
| 294 |
+
"learning_rate": 4.203117865141635e-05,
|
| 295 |
+
"loss": 0.2504,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5454545454545454,
|
| 300 |
+
"grad_norm": 0.10190426019208573,
|
| 301 |
+
"learning_rate": 4.1651828291931264e-05,
|
| 302 |
+
"loss": 0.2561,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5584415584415584,
|
| 307 |
+
"grad_norm": 0.09480915169746995,
|
| 308 |
+
"learning_rate": 4.126545750510605e-05,
|
| 309 |
+
"loss": 0.2571,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5714285714285714,
|
| 314 |
+
"grad_norm": 0.09362782945360802,
|
| 315 |
+
"learning_rate": 4.0872229185248075e-05,
|
| 316 |
+
"loss": 0.2472,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5844155844155844,
|
| 321 |
+
"grad_norm": 0.07394172741628195,
|
| 322 |
+
"learning_rate": 4.047230911780737e-05,
|
| 323 |
+
"loss": 0.2208,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5974025974025974,
|
| 328 |
+
"grad_norm": 0.0728323124890986,
|
| 329 |
+
"learning_rate": 4.0065865909481417e-05,
|
| 330 |
+
"loss": 0.2572,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6103896103896104,
|
| 335 |
+
"grad_norm": 0.10380156711543853,
|
| 336 |
+
"learning_rate": 3.965307091713037e-05,
|
| 337 |
+
"loss": 0.218,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6233766233766234,
|
| 342 |
+
"grad_norm": 0.09455312570599705,
|
| 343 |
+
"learning_rate": 3.923409817553284e-05,
|
| 344 |
+
"loss": 0.234,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6363636363636364,
|
| 349 |
+
"grad_norm": 0.08887812243253841,
|
| 350 |
+
"learning_rate": 3.880912432401265e-05,
|
| 351 |
+
"loss": 0.2037,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6493506493506493,
|
| 356 |
+
"grad_norm": 0.07668088360221971,
|
| 357 |
+
"learning_rate": 3.837832853196751e-05,
|
| 358 |
+
"loss": 0.242,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6623376623376623,
|
| 363 |
+
"grad_norm": 0.07759668294077805,
|
| 364 |
+
"learning_rate": 3.794189242333106e-05,
|
| 365 |
+
"loss": 0.2439,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6753246753246753,
|
| 370 |
+
"grad_norm": 0.09233954207922085,
|
| 371 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 372 |
+
"loss": 0.2081,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6883116883116883,
|
| 377 |
+
"grad_norm": 0.08183875046921972,
|
| 378 |
+
"learning_rate": 3.705283756425872e-05,
|
| 379 |
+
"loss": 0.2424,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7012987012987013,
|
| 384 |
+
"grad_norm": 0.10315585570133078,
|
| 385 |
+
"learning_rate": 3.6600593640234086e-05,
|
| 386 |
+
"loss": 0.2468,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7142857142857143,
|
| 391 |
+
"grad_norm": 0.08508686423963427,
|
| 392 |
+
"learning_rate": 3.6143458894413465e-05,
|
| 393 |
+
"loss": 0.2218,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7272727272727273,
|
| 398 |
+
"grad_norm": 0.08486843572184524,
|
| 399 |
+
"learning_rate": 3.568162605525953e-05,
|
| 400 |
+
"loss": 0.2322,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7402597402597403,
|
| 405 |
+
"grad_norm": 0.08545739449487202,
|
| 406 |
+
"learning_rate": 3.5215289831955786e-05,
|
| 407 |
+
"loss": 0.241,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7532467532467533,
|
| 412 |
+
"grad_norm": 0.06488948479052951,
|
| 413 |
+
"learning_rate": 3.474464683231698e-05,
|
| 414 |
+
"loss": 0.2249,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7662337662337663,
|
| 419 |
+
"grad_norm": 0.0904558318938033,
|
| 420 |
+
"learning_rate": 3.426989547989902e-05,
|
| 421 |
+
"loss": 0.2337,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7792207792207793,
|
| 426 |
+
"grad_norm": 0.09019149052874986,
|
| 427 |
+
"learning_rate": 3.379123593034342e-05,
|
| 428 |
+
"loss": 0.2112,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7922077922077922,
|
| 433 |
+
"grad_norm": 0.08328318607208472,
|
| 434 |
+
"learning_rate": 3.330886998699149e-05,
|
| 435 |
+
"loss": 0.2007,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.8051948051948052,
|
| 440 |
+
"grad_norm": 0.08263878807320083,
|
| 441 |
+
"learning_rate": 3.282300101580386e-05,
|
| 442 |
+
"loss": 0.2259,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8181818181818182,
|
| 447 |
+
"grad_norm": 0.07957332172491985,
|
| 448 |
+
"learning_rate": 3.2333833859621153e-05,
|
| 449 |
+
"loss": 0.2251,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8311688311688312,
|
| 454 |
+
"grad_norm": 0.10440656634971286,
|
| 455 |
+
"learning_rate": 3.1841574751802076e-05,
|
| 456 |
+
"loss": 0.2114,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8441558441558441,
|
| 461 |
+
"grad_norm": 0.07647118607084558,
|
| 462 |
+
"learning_rate": 3.13464312292752e-05,
|
| 463 |
+
"loss": 0.2123,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8571428571428571,
|
| 468 |
+
"grad_norm": 0.08658758520644272,
|
| 469 |
+
"learning_rate": 3.084861204504122e-05,
|
| 470 |
+
"loss": 0.1994,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8701298701298701,
|
| 475 |
+
"grad_norm": 0.08858346318492633,
|
| 476 |
+
"learning_rate": 3.0348327080162435e-05,
|
| 477 |
+
"loss": 0.2377,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8831168831168831,
|
| 482 |
+
"grad_norm": 0.09235241797460036,
|
| 483 |
+
"learning_rate": 2.9845787255276753e-05,
|
| 484 |
+
"loss": 0.215,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8961038961038961,
|
| 489 |
+
"grad_norm": 0.0788972148991508,
|
| 490 |
+
"learning_rate": 2.9341204441673266e-05,
|
| 491 |
+
"loss": 0.2112,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.9090909090909091,
|
| 496 |
+
"grad_norm": 0.08960215541982136,
|
| 497 |
+
"learning_rate": 2.8834791371967142e-05,
|
| 498 |
+
"loss": 0.2647,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.922077922077922,
|
| 503 |
+
"grad_norm": 0.08380836318512074,
|
| 504 |
+
"learning_rate": 2.8326761550411345e-05,
|
| 505 |
+
"loss": 0.2431,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.935064935064935,
|
| 510 |
+
"grad_norm": 0.09187039951568829,
|
| 511 |
+
"learning_rate": 2.781732916288303e-05,
|
| 512 |
+
"loss": 0.2479,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.948051948051948,
|
| 517 |
+
"grad_norm": 0.09291540687726117,
|
| 518 |
+
"learning_rate": 2.7306708986582553e-05,
|
| 519 |
+
"loss": 0.2357,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.961038961038961,
|
| 524 |
+
"grad_norm": 0.0832454958287789,
|
| 525 |
+
"learning_rate": 2.679511629948319e-05,
|
| 526 |
+
"loss": 0.236,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.974025974025974,
|
| 531 |
+
"grad_norm": 0.07504819631461856,
|
| 532 |
+
"learning_rate": 2.628276678956974e-05,
|
| 533 |
+
"loss": 0.2288,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.987012987012987,
|
| 538 |
+
"grad_norm": 0.10741006282709406,
|
| 539 |
+
"learning_rate": 2.5769876463904265e-05,
|
| 540 |
+
"loss": 0.2404,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.0,
|
| 545 |
+
"grad_norm": 0.0825957236169846,
|
| 546 |
+
"learning_rate": 2.5256661557557247e-05,
|
| 547 |
+
"loss": 0.2436,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.0129870129870129,
|
| 552 |
+
"grad_norm": 0.07684874681832127,
|
| 553 |
+
"learning_rate": 2.4743338442442755e-05,
|
| 554 |
+
"loss": 0.2244,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.025974025974026,
|
| 559 |
+
"grad_norm": 0.07704336923237068,
|
| 560 |
+
"learning_rate": 2.4230123536095748e-05,
|
| 561 |
+
"loss": 0.2626,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.0389610389610389,
|
| 566 |
+
"grad_norm": 0.07500658424467158,
|
| 567 |
+
"learning_rate": 2.3717233210430256e-05,
|
| 568 |
+
"loss": 0.2192,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.051948051948052,
|
| 573 |
+
"grad_norm": 0.08184169243002529,
|
| 574 |
+
"learning_rate": 2.3204883700516812e-05,
|
| 575 |
+
"loss": 0.2148,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.0649350649350648,
|
| 580 |
+
"grad_norm": 0.08482067089986454,
|
| 581 |
+
"learning_rate": 2.2693291013417453e-05,
|
| 582 |
+
"loss": 0.2113,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.077922077922078,
|
| 587 |
+
"grad_norm": 0.10550670423205909,
|
| 588 |
+
"learning_rate": 2.2182670837116975e-05,
|
| 589 |
+
"loss": 0.2463,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0909090909090908,
|
| 594 |
+
"grad_norm": 0.07848865475761758,
|
| 595 |
+
"learning_rate": 2.1673238449588668e-05,
|
| 596 |
+
"loss": 0.2544,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.103896103896104,
|
| 601 |
+
"grad_norm": 0.08116964479727118,
|
| 602 |
+
"learning_rate": 2.116520862803286e-05,
|
| 603 |
+
"loss": 0.203,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.1168831168831168,
|
| 608 |
+
"grad_norm": 0.0773019598198873,
|
| 609 |
+
"learning_rate": 2.0658795558326743e-05,
|
| 610 |
+
"loss": 0.1896,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.12987012987013,
|
| 615 |
+
"grad_norm": 0.08577851670838929,
|
| 616 |
+
"learning_rate": 2.015421274472325e-05,
|
| 617 |
+
"loss": 0.2132,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.1428571428571428,
|
| 622 |
+
"grad_norm": 0.08126404156555671,
|
| 623 |
+
"learning_rate": 1.965167291983757e-05,
|
| 624 |
+
"loss": 0.2156,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.155844155844156,
|
| 629 |
+
"grad_norm": 0.08196682847954194,
|
| 630 |
+
"learning_rate": 1.9151387954958794e-05,
|
| 631 |
+
"loss": 0.178,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.1688311688311688,
|
| 636 |
+
"grad_norm": 0.07898254614136963,
|
| 637 |
+
"learning_rate": 1.8653568770724806e-05,
|
| 638 |
+
"loss": 0.2158,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1818181818181819,
|
| 643 |
+
"grad_norm": 0.1007894448072893,
|
| 644 |
+
"learning_rate": 1.815842524819793e-05,
|
| 645 |
+
"loss": 0.176,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1948051948051948,
|
| 650 |
+
"grad_norm": 0.08182215036544653,
|
| 651 |
+
"learning_rate": 1.7666166140378852e-05,
|
| 652 |
+
"loss": 0.2048,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.2077922077922079,
|
| 657 |
+
"grad_norm": 0.11560790878431627,
|
| 658 |
+
"learning_rate": 1.7176998984196146e-05,
|
| 659 |
+
"loss": 0.1904,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.2207792207792207,
|
| 664 |
+
"grad_norm": 0.12113066083881273,
|
| 665 |
+
"learning_rate": 1.6691130013008514e-05,
|
| 666 |
+
"loss": 0.1974,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.2337662337662338,
|
| 671 |
+
"grad_norm": 0.08515149977031479,
|
| 672 |
+
"learning_rate": 1.620876406965658e-05,
|
| 673 |
+
"loss": 0.2049,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2467532467532467,
|
| 678 |
+
"grad_norm": 0.09322796776822118,
|
| 679 |
+
"learning_rate": 1.5730104520100982e-05,
|
| 680 |
+
"loss": 0.2225,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.2597402597402598,
|
| 685 |
+
"grad_norm": 0.0898952962414004,
|
| 686 |
+
"learning_rate": 1.5255353167683017e-05,
|
| 687 |
+
"loss": 0.2139,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.2727272727272727,
|
| 692 |
+
"grad_norm": 0.08430600938698268,
|
| 693 |
+
"learning_rate": 1.4784710168044213e-05,
|
| 694 |
+
"loss": 0.2254,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.2857142857142856,
|
| 699 |
+
"grad_norm": 0.08721397526200095,
|
| 700 |
+
"learning_rate": 1.4318373944740484e-05,
|
| 701 |
+
"loss": 0.1948,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.2987012987012987,
|
| 706 |
+
"grad_norm": 0.09193805251870187,
|
| 707 |
+
"learning_rate": 1.3856541105586545e-05,
|
| 708 |
+
"loss": 0.189,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.3116883116883118,
|
| 713 |
+
"grad_norm": 0.09260064723369027,
|
| 714 |
+
"learning_rate": 1.339940635976592e-05,
|
| 715 |
+
"loss": 0.1853,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.3246753246753247,
|
| 720 |
+
"grad_norm": 0.09161658350392674,
|
| 721 |
+
"learning_rate": 1.2947162435741278e-05,
|
| 722 |
+
"loss": 0.2351,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.3376623376623376,
|
| 727 |
+
"grad_norm": 0.07939421486765358,
|
| 728 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 729 |
+
"loss": 0.2199,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.3506493506493507,
|
| 734 |
+
"grad_norm": 0.08988123365679614,
|
| 735 |
+
"learning_rate": 1.205810757666894e-05,
|
| 736 |
+
"loss": 0.2203,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.3636363636363638,
|
| 741 |
+
"grad_norm": 0.08790912980331905,
|
| 742 |
+
"learning_rate": 1.1621671468032493e-05,
|
| 743 |
+
"loss": 0.1891,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.3766233766233766,
|
| 748 |
+
"grad_norm": 0.07960570450390624,
|
| 749 |
+
"learning_rate": 1.1190875675987356e-05,
|
| 750 |
+
"loss": 0.2464,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.3896103896103895,
|
| 755 |
+
"grad_norm": 0.08768976314977797,
|
| 756 |
+
"learning_rate": 1.0765901824467167e-05,
|
| 757 |
+
"loss": 0.213,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.4025974025974026,
|
| 762 |
+
"grad_norm": 0.0841723141024622,
|
| 763 |
+
"learning_rate": 1.0346929082869641e-05,
|
| 764 |
+
"loss": 0.2121,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.4155844155844157,
|
| 769 |
+
"grad_norm": 0.08969311043319497,
|
| 770 |
+
"learning_rate": 9.934134090518593e-06,
|
| 771 |
+
"loss": 0.1684,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.4285714285714286,
|
| 776 |
+
"grad_norm": 0.10925176253438569,
|
| 777 |
+
"learning_rate": 9.527690882192636e-06,
|
| 778 |
+
"loss": 0.2108,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.4415584415584415,
|
| 783 |
+
"grad_norm": 0.10044980805756751,
|
| 784 |
+
"learning_rate": 9.127770814751933e-06,
|
| 785 |
+
"loss": 0.2142,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.4545454545454546,
|
| 790 |
+
"grad_norm": 0.10075986222908401,
|
| 791 |
+
"learning_rate": 8.734542494893955e-06,
|
| 792 |
+
"loss": 0.2239,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.4675324675324675,
|
| 797 |
+
"grad_norm": 0.14624227257627387,
|
| 798 |
+
"learning_rate": 8.348171708068747e-06,
|
| 799 |
+
"loss": 0.1907,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.4805194805194806,
|
| 804 |
+
"grad_norm": 0.08167017231165358,
|
| 805 |
+
"learning_rate": 7.968821348583644e-06,
|
| 806 |
+
"loss": 0.2178,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.4935064935064934,
|
| 811 |
+
"grad_norm": 0.1167316559372597,
|
| 812 |
+
"learning_rate": 7.5966513509268365e-06,
|
| 813 |
+
"loss": 0.1998,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.5064935064935066,
|
| 818 |
+
"grad_norm": 0.08396562767447123,
|
| 819 |
+
"learning_rate": 7.231818622338823e-06,
|
| 820 |
+
"loss": 0.213,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.5194805194805194,
|
| 825 |
+
"grad_norm": 0.08943222467451503,
|
| 826 |
+
"learning_rate": 6.8744769766601854e-06,
|
| 827 |
+
"loss": 0.2085,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.5324675324675323,
|
| 832 |
+
"grad_norm": 0.08594290696448413,
|
| 833 |
+
"learning_rate": 6.524777069483526e-06,
|
| 834 |
+
"loss": 0.2005,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.5454545454545454,
|
| 839 |
+
"grad_norm": 0.09141957886942967,
|
| 840 |
+
"learning_rate": 6.182866334636889e-06,
|
| 841 |
+
"loss": 0.185,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.5584415584415585,
|
| 846 |
+
"grad_norm": 0.09996431130975963,
|
| 847 |
+
"learning_rate": 5.848888922025553e-06,
|
| 848 |
+
"loss": 0.2306,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.5714285714285714,
|
| 853 |
+
"grad_norm": 0.09454859006099142,
|
| 854 |
+
"learning_rate": 5.522985636858239e-06,
|
| 855 |
+
"loss": 0.2195,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.5844155844155843,
|
| 860 |
+
"grad_norm": 0.08951517611083147,
|
| 861 |
+
"learning_rate": 5.205293880283552e-06,
|
| 862 |
+
"loss": 0.1824,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.5974025974025974,
|
| 867 |
+
"grad_norm": 0.08973290394393159,
|
| 868 |
+
"learning_rate": 4.8959475914614554e-06,
|
| 869 |
+
"loss": 0.2031,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.6103896103896105,
|
| 874 |
+
"grad_norm": 0.0972420458274638,
|
| 875 |
+
"learning_rate": 4.5950771910944605e-06,
|
| 876 |
+
"loss": 0.218,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.6233766233766234,
|
| 881 |
+
"grad_norm": 0.1028775215887483,
|
| 882 |
+
"learning_rate": 4.3028095264420535e-06,
|
| 883 |
+
"loss": 0.2295,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.6363636363636362,
|
| 888 |
+
"grad_norm": 0.09422285979262787,
|
| 889 |
+
"learning_rate": 4.019267817841835e-06,
|
| 890 |
+
"loss": 0.231,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.6493506493506493,
|
| 895 |
+
"grad_norm": 0.08245701472584765,
|
| 896 |
+
"learning_rate": 3.7445716067596503e-06,
|
| 897 |
+
"loss": 0.1968,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.6623376623376624,
|
| 902 |
+
"grad_norm": 0.0900955932960198,
|
| 903 |
+
"learning_rate": 3.478836705390809e-06,
|
| 904 |
+
"loss": 0.2201,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.6753246753246753,
|
| 909 |
+
"grad_norm": 0.08630235476963846,
|
| 910 |
+
"learning_rate": 3.222175147833556e-06,
|
| 911 |
+
"loss": 0.2172,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.6883116883116882,
|
| 916 |
+
"grad_norm": 0.09086270485409138,
|
| 917 |
+
"learning_rate": 2.974695142855388e-06,
|
| 918 |
+
"loss": 0.2116,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.7012987012987013,
|
| 923 |
+
"grad_norm": 0.09162946812983505,
|
| 924 |
+
"learning_rate": 2.7365010282720952e-06,
|
| 925 |
+
"loss": 0.198,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.7142857142857144,
|
| 930 |
+
"grad_norm": 0.08702063300963413,
|
| 931 |
+
"learning_rate": 2.507693226958871e-06,
|
| 932 |
+
"loss": 0.2129,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.7272727272727273,
|
| 937 |
+
"grad_norm": 0.0881788504739707,
|
| 938 |
+
"learning_rate": 2.2883682045119063e-06,
|
| 939 |
+
"loss": 0.167,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.7402597402597402,
|
| 944 |
+
"grad_norm": 0.11644782656509362,
|
| 945 |
+
"learning_rate": 2.0786184285784297e-06,
|
| 946 |
+
"loss": 0.2054,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.7532467532467533,
|
| 951 |
+
"grad_norm": 0.1766008455564786,
|
| 952 |
+
"learning_rate": 1.8785323298722097e-06,
|
| 953 |
+
"loss": 0.2186,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.7662337662337664,
|
| 958 |
+
"grad_norm": 0.09795182977403628,
|
| 959 |
+
"learning_rate": 1.6881942648911076e-06,
|
| 960 |
+
"loss": 0.1994,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.7792207792207793,
|
| 965 |
+
"grad_norm": 0.12180290940767717,
|
| 966 |
+
"learning_rate": 1.5076844803522922e-06,
|
| 967 |
+
"loss": 0.2012,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.7922077922077921,
|
| 972 |
+
"grad_norm": 0.094696794518402,
|
| 973 |
+
"learning_rate": 1.3370790793601373e-06,
|
| 974 |
+
"loss": 0.2245,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.8051948051948052,
|
| 979 |
+
"grad_norm": 0.0761787809631694,
|
| 980 |
+
"learning_rate": 1.1764499893210878e-06,
|
| 981 |
+
"loss": 0.1963,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.8181818181818183,
|
| 986 |
+
"grad_norm": 0.09090562786095378,
|
| 987 |
+
"learning_rate": 1.0258649316189722e-06,
|
| 988 |
+
"loss": 0.1877,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.8311688311688312,
|
| 993 |
+
"grad_norm": 0.08173059503885717,
|
| 994 |
+
"learning_rate": 8.85387393063622e-07,
|
| 995 |
+
"loss": 0.2222,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.844155844155844,
|
| 1000 |
+
"grad_norm": 0.08635060155818351,
|
| 1001 |
+
"learning_rate": 7.550765991247654e-07,
|
| 1002 |
+
"loss": 0.2235,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.8571428571428572,
|
| 1007 |
+
"grad_norm": 0.08607284916997383,
|
| 1008 |
+
"learning_rate": 6.349874889624962e-07,
|
| 1009 |
+
"loss": 0.2071,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.87012987012987,
|
| 1014 |
+
"grad_norm": 0.07825829604724413,
|
| 1015 |
+
"learning_rate": 5.25170692264887e-07,
|
| 1016 |
+
"loss": 0.192,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.883116883116883,
|
| 1021 |
+
"grad_norm": 0.08910203901931234,
|
| 1022 |
+
"learning_rate": 4.256725079024554e-07,
|
| 1023 |
+
"loss": 0.2164,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.896103896103896,
|
| 1028 |
+
"grad_norm": 0.08928208281125428,
|
| 1029 |
+
"learning_rate": 3.3653488440851255e-07,
|
| 1030 |
+
"loss": 0.2018,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.9090909090909092,
|
| 1035 |
+
"grad_norm": 0.08495149169392952,
|
| 1036 |
+
"learning_rate": 2.5779540229361745e-07,
|
| 1037 |
+
"loss": 0.1961,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.922077922077922,
|
| 1042 |
+
"grad_norm": 0.08626229151823323,
|
| 1043 |
+
"learning_rate": 1.8948725820160662e-07,
|
| 1044 |
+
"loss": 0.1768,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.935064935064935,
|
| 1049 |
+
"grad_norm": 0.08571927306677737,
|
| 1050 |
+
"learning_rate": 1.3163925091384533e-07,
|
| 1051 |
+
"loss": 0.2232,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.948051948051948,
|
| 1056 |
+
"grad_norm": 0.09343183243370598,
|
| 1057 |
+
"learning_rate": 8.427576920763958e-08,
|
| 1058 |
+
"loss": 0.1687,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.9610389610389611,
|
| 1063 |
+
"grad_norm": 0.0946463454270307,
|
| 1064 |
+
"learning_rate": 4.741678157389739e-08,
|
| 1065 |
+
"loss": 0.187,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.974025974025974,
|
| 1070 |
+
"grad_norm": 0.09513291585193097,
|
| 1071 |
+
"learning_rate": 2.1077827798404726e-08,
|
| 1072 |
+
"loss": 0.2231,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.987012987012987,
|
| 1077 |
+
"grad_norm": 0.085055457153114,
|
| 1078 |
+
"learning_rate": 5.270012410216185e-09,
|
| 1079 |
+
"loss": 0.1792,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.0,
|
| 1084 |
+
"grad_norm": 0.08464289636250304,
|
| 1085 |
+
"learning_rate": 0.0,
|
| 1086 |
+
"loss": 0.1753,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.0,
|
| 1091 |
+
"step": 154,
|
| 1092 |
+
"total_flos": 145121340096512.0,
|
| 1093 |
+
"train_loss": 0.2980368115104638,
|
| 1094 |
+
"train_runtime": 2252.7177,
|
| 1095 |
+
"train_samples_per_second": 0.272,
|
| 1096 |
+
"train_steps_per_second": 0.068
|
| 1097 |
+
}
|
| 1098 |
+
],
|
| 1099 |
+
"logging_steps": 1,
|
| 1100 |
+
"max_steps": 154,
|
| 1101 |
+
"num_input_tokens_seen": 0,
|
| 1102 |
+
"num_train_epochs": 2,
|
| 1103 |
+
"save_steps": 50,
|
| 1104 |
+
"stateful_callbacks": {
|
| 1105 |
+
"TrainerControl": {
|
| 1106 |
+
"args": {
|
| 1107 |
+
"should_epoch_stop": false,
|
| 1108 |
+
"should_evaluate": false,
|
| 1109 |
+
"should_log": false,
|
| 1110 |
+
"should_save": true,
|
| 1111 |
+
"should_training_stop": true
|
| 1112 |
+
},
|
| 1113 |
+
"attributes": {}
|
| 1114 |
+
}
|
| 1115 |
+
},
|
| 1116 |
+
"total_flos": 145121340096512.0,
|
| 1117 |
+
"train_batch_size": 1,
|
| 1118 |
+
"trial_name": null,
|
| 1119 |
+
"trial_params": null
|
| 1120 |
+
}
|
training_loss.png
ADDED
|