End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +885 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: llama3-1_8b_oh_dcft_v1_wo_camel_biology_gpt-4o-mini
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama3-1_8b_oh_dcft_v1_wo_camel_biology_gpt-4o-mini
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.5472
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: llama3-1_8b_oh_dcft_v1_wo_camel_biology_gpt-4o-mini
|
|
|
|
| 16 |
|
| 17 |
# llama3-1_8b_oh_dcft_v1_wo_camel_biology_gpt-4o-mini
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/OH_DCFT_v1_wo_camel_biology_gpt-4o-mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5472
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5471954345703125,
|
| 4 |
+
"eval_runtime": 35.9766,
|
| 5 |
+
"eval_samples_per_second": 291.967,
|
| 6 |
+
"eval_steps_per_second": 1.167,
|
| 7 |
+
"total_flos": 5.524377893797154e+19,
|
| 8 |
+
"train_loss": 0.5494362928928473,
|
| 9 |
+
"train_runtime": 8709.0296,
|
| 10 |
+
"train_samples_per_second": 68.748,
|
| 11 |
+
"train_steps_per_second": 0.134
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5471954345703125,
|
| 4 |
+
"eval_runtime": 35.9766,
|
| 5 |
+
"eval_samples_per_second": 291.967,
|
| 6 |
+
"eval_steps_per_second": 1.167
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 5.524377893797154e+19,
|
| 4 |
+
"train_loss": 0.5494362928928473,
|
| 5 |
+
"train_runtime": 8709.0296,
|
| 6 |
+
"train_samples_per_second": 68.748,
|
| 7 |
+
"train_steps_per_second": 0.134
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,885 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1170,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02564102564102564,
|
| 13 |
+
"grad_norm": 3.406068801879883,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.9168,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05128205128205128,
|
| 20 |
+
"grad_norm": 3.5034339427948,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7688,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07692307692307693,
|
| 27 |
+
"grad_norm": 1.0085859298706055,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.729,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10256410256410256,
|
| 34 |
+
"grad_norm": 0.8697875142097473,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.7004,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1282051282051282,
|
| 41 |
+
"grad_norm": 0.9098314642906189,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6853,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15384615384615385,
|
| 48 |
+
"grad_norm": 1.240673542022705,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6668,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1794871794871795,
|
| 55 |
+
"grad_norm": 1.3245731592178345,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.654,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.20512820512820512,
|
| 62 |
+
"grad_norm": 0.6942853331565857,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6427,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.23076923076923078,
|
| 69 |
+
"grad_norm": 0.7867761254310608,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6425,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2564102564102564,
|
| 76 |
+
"grad_norm": 0.7657337784767151,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6305,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.28205128205128205,
|
| 83 |
+
"grad_norm": 0.8531173467636108,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6308,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3076923076923077,
|
| 90 |
+
"grad_norm": 0.6474210023880005,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6227,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3333333333333333,
|
| 97 |
+
"grad_norm": 0.6232497692108154,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6293,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.358974358974359,
|
| 104 |
+
"grad_norm": 0.6542862057685852,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6217,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.38461538461538464,
|
| 111 |
+
"grad_norm": 0.9984610080718994,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6213,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.41025641025641024,
|
| 118 |
+
"grad_norm": 0.7436656355857849,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.614,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4358974358974359,
|
| 125 |
+
"grad_norm": 0.7107124328613281,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.6057,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.46153846153846156,
|
| 132 |
+
"grad_norm": 0.8069647550582886,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6134,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.48717948717948717,
|
| 139 |
+
"grad_norm": 0.5554560422897339,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6069,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5128205128205128,
|
| 146 |
+
"grad_norm": 0.5725631713867188,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6081,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5384615384615384,
|
| 153 |
+
"grad_norm": 0.7718241214752197,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.6003,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5641025641025641,
|
| 160 |
+
"grad_norm": 0.659430742263794,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6044,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5897435897435898,
|
| 167 |
+
"grad_norm": 0.5641574859619141,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.5988,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6153846153846154,
|
| 174 |
+
"grad_norm": 0.613983154296875,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6079,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6410256410256411,
|
| 181 |
+
"grad_norm": 0.7252273559570312,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.5981,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6666666666666666,
|
| 188 |
+
"grad_norm": 0.6930380463600159,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.5927,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6923076923076923,
|
| 195 |
+
"grad_norm": 0.6013543605804443,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.5942,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.717948717948718,
|
| 202 |
+
"grad_norm": 0.5443430542945862,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.594,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7435897435897436,
|
| 209 |
+
"grad_norm": 0.6630271077156067,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.5926,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7692307692307693,
|
| 216 |
+
"grad_norm": 0.6013346314430237,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.5874,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7948717948717948,
|
| 223 |
+
"grad_norm": 0.5420301556587219,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.5913,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8205128205128205,
|
| 230 |
+
"grad_norm": 0.809611976146698,
|
| 231 |
+
"learning_rate": 5e-06,
|
| 232 |
+
"loss": 0.5917,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8461538461538461,
|
| 237 |
+
"grad_norm": 0.7773135304450989,
|
| 238 |
+
"learning_rate": 5e-06,
|
| 239 |
+
"loss": 0.5836,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8717948717948718,
|
| 244 |
+
"grad_norm": 0.5458276271820068,
|
| 245 |
+
"learning_rate": 5e-06,
|
| 246 |
+
"loss": 0.5794,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8974358974358975,
|
| 251 |
+
"grad_norm": 0.6851891875267029,
|
| 252 |
+
"learning_rate": 5e-06,
|
| 253 |
+
"loss": 0.5902,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.9230769230769231,
|
| 258 |
+
"grad_norm": 0.5757733583450317,
|
| 259 |
+
"learning_rate": 5e-06,
|
| 260 |
+
"loss": 0.5826,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.9487179487179487,
|
| 265 |
+
"grad_norm": 0.5545377135276794,
|
| 266 |
+
"learning_rate": 5e-06,
|
| 267 |
+
"loss": 0.5887,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.9743589743589743,
|
| 272 |
+
"grad_norm": 0.5440164804458618,
|
| 273 |
+
"learning_rate": 5e-06,
|
| 274 |
+
"loss": 0.584,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.0,
|
| 279 |
+
"grad_norm": 0.5356547236442566,
|
| 280 |
+
"learning_rate": 5e-06,
|
| 281 |
+
"loss": 0.5779,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.0,
|
| 286 |
+
"eval_loss": 0.5792508721351624,
|
| 287 |
+
"eval_runtime": 37.2797,
|
| 288 |
+
"eval_samples_per_second": 281.762,
|
| 289 |
+
"eval_steps_per_second": 1.127,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.0256410256410255,
|
| 294 |
+
"grad_norm": 0.6288334131240845,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5453,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.0512820512820513,
|
| 301 |
+
"grad_norm": 0.6558266282081604,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5475,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.0769230769230769,
|
| 308 |
+
"grad_norm": 0.7556455135345459,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.5465,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.1025641025641026,
|
| 315 |
+
"grad_norm": 0.586370050907135,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.5395,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.1282051282051282,
|
| 322 |
+
"grad_norm": 0.5581791996955872,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.5366,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.1538461538461537,
|
| 329 |
+
"grad_norm": 0.5495139956474304,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.5487,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.1794871794871795,
|
| 336 |
+
"grad_norm": 0.6099910736083984,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.5397,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.205128205128205,
|
| 343 |
+
"grad_norm": 0.5565390586853027,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.5376,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.2307692307692308,
|
| 350 |
+
"grad_norm": 0.6461251378059387,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.534,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.2564102564102564,
|
| 357 |
+
"grad_norm": 0.6213774681091309,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.5366,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.282051282051282,
|
| 364 |
+
"grad_norm": 0.5515546798706055,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.5381,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.3076923076923077,
|
| 371 |
+
"grad_norm": 0.6342268586158752,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.536,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.3333333333333333,
|
| 378 |
+
"grad_norm": 0.5595047473907471,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.5396,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.358974358974359,
|
| 385 |
+
"grad_norm": 0.5788885951042175,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.5371,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.3846153846153846,
|
| 392 |
+
"grad_norm": 0.5231652855873108,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.5357,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.4102564102564101,
|
| 399 |
+
"grad_norm": 0.7670683860778809,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.5376,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.435897435897436,
|
| 406 |
+
"grad_norm": 0.5181464552879333,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.5321,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.4615384615384617,
|
| 413 |
+
"grad_norm": 0.5286003351211548,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.5361,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.4871794871794872,
|
| 420 |
+
"grad_norm": 0.5364812016487122,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.5391,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.5128205128205128,
|
| 427 |
+
"grad_norm": 0.6188532710075378,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.5336,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.5384615384615383,
|
| 434 |
+
"grad_norm": 0.4892689883708954,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.5352,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.564102564102564,
|
| 441 |
+
"grad_norm": 0.5404101014137268,
|
| 442 |
+
"learning_rate": 5e-06,
|
| 443 |
+
"loss": 0.5321,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.5897435897435899,
|
| 448 |
+
"grad_norm": 0.5570105314254761,
|
| 449 |
+
"learning_rate": 5e-06,
|
| 450 |
+
"loss": 0.5341,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.6153846153846154,
|
| 455 |
+
"grad_norm": 0.6178380846977234,
|
| 456 |
+
"learning_rate": 5e-06,
|
| 457 |
+
"loss": 0.5321,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.641025641025641,
|
| 462 |
+
"grad_norm": 0.8673184514045715,
|
| 463 |
+
"learning_rate": 5e-06,
|
| 464 |
+
"loss": 0.5216,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.6666666666666665,
|
| 469 |
+
"grad_norm": 0.6193057298660278,
|
| 470 |
+
"learning_rate": 5e-06,
|
| 471 |
+
"loss": 0.5312,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.6923076923076923,
|
| 476 |
+
"grad_norm": 0.5737960934638977,
|
| 477 |
+
"learning_rate": 5e-06,
|
| 478 |
+
"loss": 0.5317,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.717948717948718,
|
| 483 |
+
"grad_norm": 0.6953139305114746,
|
| 484 |
+
"learning_rate": 5e-06,
|
| 485 |
+
"loss": 0.5297,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.7435897435897436,
|
| 490 |
+
"grad_norm": 0.5698896050453186,
|
| 491 |
+
"learning_rate": 5e-06,
|
| 492 |
+
"loss": 0.5248,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.7692307692307692,
|
| 497 |
+
"grad_norm": 0.4578630328178406,
|
| 498 |
+
"learning_rate": 5e-06,
|
| 499 |
+
"loss": 0.5251,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.7948717948717947,
|
| 504 |
+
"grad_norm": 0.4654785692691803,
|
| 505 |
+
"learning_rate": 5e-06,
|
| 506 |
+
"loss": 0.5308,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.8205128205128205,
|
| 511 |
+
"grad_norm": 0.500807523727417,
|
| 512 |
+
"learning_rate": 5e-06,
|
| 513 |
+
"loss": 0.5259,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.8461538461538463,
|
| 518 |
+
"grad_norm": 0.5273367762565613,
|
| 519 |
+
"learning_rate": 5e-06,
|
| 520 |
+
"loss": 0.5286,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.8717948717948718,
|
| 525 |
+
"grad_norm": 0.49950945377349854,
|
| 526 |
+
"learning_rate": 5e-06,
|
| 527 |
+
"loss": 0.5292,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.8974358974358974,
|
| 532 |
+
"grad_norm": 0.46231916546821594,
|
| 533 |
+
"learning_rate": 5e-06,
|
| 534 |
+
"loss": 0.5243,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.9230769230769231,
|
| 539 |
+
"grad_norm": 0.5422210693359375,
|
| 540 |
+
"learning_rate": 5e-06,
|
| 541 |
+
"loss": 0.5241,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.9487179487179487,
|
| 546 |
+
"grad_norm": 0.569016695022583,
|
| 547 |
+
"learning_rate": 5e-06,
|
| 548 |
+
"loss": 0.5237,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.9743589743589745,
|
| 553 |
+
"grad_norm": 0.5401661396026611,
|
| 554 |
+
"learning_rate": 5e-06,
|
| 555 |
+
"loss": 0.5289,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 2.0,
|
| 560 |
+
"grad_norm": 0.6024136543273926,
|
| 561 |
+
"learning_rate": 5e-06,
|
| 562 |
+
"loss": 0.5262,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 2.0,
|
| 567 |
+
"eval_loss": 0.5533561706542969,
|
| 568 |
+
"eval_runtime": 36.5672,
|
| 569 |
+
"eval_samples_per_second": 287.252,
|
| 570 |
+
"eval_steps_per_second": 1.149,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.0256410256410255,
|
| 575 |
+
"grad_norm": 0.5916836857795715,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.4847,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.051282051282051,
|
| 582 |
+
"grad_norm": 0.6934983730316162,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.4906,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.076923076923077,
|
| 589 |
+
"grad_norm": 0.546357274055481,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.4857,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.1025641025641026,
|
| 596 |
+
"grad_norm": 0.7491461634635925,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.4861,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.128205128205128,
|
| 603 |
+
"grad_norm": 0.5870935320854187,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.4847,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.1538461538461537,
|
| 610 |
+
"grad_norm": 0.6286914348602295,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.4895,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.1794871794871793,
|
| 617 |
+
"grad_norm": 0.515387773513794,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.4897,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.2051282051282053,
|
| 624 |
+
"grad_norm": 0.5456680655479431,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.4861,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.230769230769231,
|
| 631 |
+
"grad_norm": 0.49457597732543945,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.484,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.2564102564102564,
|
| 638 |
+
"grad_norm": 0.5255215167999268,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.4936,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.282051282051282,
|
| 645 |
+
"grad_norm": 0.5122373700141907,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.4915,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.3076923076923075,
|
| 652 |
+
"grad_norm": 0.5085675120353699,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.4894,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.3333333333333335,
|
| 659 |
+
"grad_norm": 0.5925104022026062,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.4873,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.358974358974359,
|
| 666 |
+
"grad_norm": 0.519869327545166,
|
| 667 |
+
"learning_rate": 5e-06,
|
| 668 |
+
"loss": 0.4882,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.3846153846153846,
|
| 673 |
+
"grad_norm": 0.550400972366333,
|
| 674 |
+
"learning_rate": 5e-06,
|
| 675 |
+
"loss": 0.486,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.41025641025641,
|
| 680 |
+
"grad_norm": 0.6560305953025818,
|
| 681 |
+
"learning_rate": 5e-06,
|
| 682 |
+
"loss": 0.4963,
|
| 683 |
+
"step": 940
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.435897435897436,
|
| 687 |
+
"grad_norm": 0.5110371112823486,
|
| 688 |
+
"learning_rate": 5e-06,
|
| 689 |
+
"loss": 0.4855,
|
| 690 |
+
"step": 950
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.4615384615384617,
|
| 694 |
+
"grad_norm": 0.48929715156555176,
|
| 695 |
+
"learning_rate": 5e-06,
|
| 696 |
+
"loss": 0.4859,
|
| 697 |
+
"step": 960
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.4871794871794872,
|
| 701 |
+
"grad_norm": 0.5958628058433533,
|
| 702 |
+
"learning_rate": 5e-06,
|
| 703 |
+
"loss": 0.4884,
|
| 704 |
+
"step": 970
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.5128205128205128,
|
| 708 |
+
"grad_norm": 0.604121208190918,
|
| 709 |
+
"learning_rate": 5e-06,
|
| 710 |
+
"loss": 0.4858,
|
| 711 |
+
"step": 980
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.5384615384615383,
|
| 715 |
+
"grad_norm": 0.5479825735092163,
|
| 716 |
+
"learning_rate": 5e-06,
|
| 717 |
+
"loss": 0.4864,
|
| 718 |
+
"step": 990
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.564102564102564,
|
| 722 |
+
"grad_norm": 0.5345059633255005,
|
| 723 |
+
"learning_rate": 5e-06,
|
| 724 |
+
"loss": 0.4859,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.58974358974359,
|
| 729 |
+
"grad_norm": 0.519960880279541,
|
| 730 |
+
"learning_rate": 5e-06,
|
| 731 |
+
"loss": 0.4867,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.6153846153846154,
|
| 736 |
+
"grad_norm": 0.5771586298942566,
|
| 737 |
+
"learning_rate": 5e-06,
|
| 738 |
+
"loss": 0.4921,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.641025641025641,
|
| 743 |
+
"grad_norm": 0.5097467303276062,
|
| 744 |
+
"learning_rate": 5e-06,
|
| 745 |
+
"loss": 0.4898,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.6666666666666665,
|
| 750 |
+
"grad_norm": 0.5330426096916199,
|
| 751 |
+
"learning_rate": 5e-06,
|
| 752 |
+
"loss": 0.487,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.6923076923076925,
|
| 757 |
+
"grad_norm": 0.6157095432281494,
|
| 758 |
+
"learning_rate": 5e-06,
|
| 759 |
+
"loss": 0.4896,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.717948717948718,
|
| 764 |
+
"grad_norm": 0.5218502283096313,
|
| 765 |
+
"learning_rate": 5e-06,
|
| 766 |
+
"loss": 0.4883,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.7435897435897436,
|
| 771 |
+
"grad_norm": 0.620424747467041,
|
| 772 |
+
"learning_rate": 5e-06,
|
| 773 |
+
"loss": 0.4895,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.769230769230769,
|
| 778 |
+
"grad_norm": 0.5703057646751404,
|
| 779 |
+
"learning_rate": 5e-06,
|
| 780 |
+
"loss": 0.4854,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.7948717948717947,
|
| 785 |
+
"grad_norm": 0.49105432629585266,
|
| 786 |
+
"learning_rate": 5e-06,
|
| 787 |
+
"loss": 0.4929,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.8205128205128203,
|
| 792 |
+
"grad_norm": 0.5956032872200012,
|
| 793 |
+
"learning_rate": 5e-06,
|
| 794 |
+
"loss": 0.4847,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.8461538461538463,
|
| 799 |
+
"grad_norm": 0.6310620307922363,
|
| 800 |
+
"learning_rate": 5e-06,
|
| 801 |
+
"loss": 0.4824,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.871794871794872,
|
| 806 |
+
"grad_norm": 0.5469298958778381,
|
| 807 |
+
"learning_rate": 5e-06,
|
| 808 |
+
"loss": 0.4843,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.8974358974358974,
|
| 813 |
+
"grad_norm": 0.6324126720428467,
|
| 814 |
+
"learning_rate": 5e-06,
|
| 815 |
+
"loss": 0.4822,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.9230769230769234,
|
| 820 |
+
"grad_norm": 0.5963416695594788,
|
| 821 |
+
"learning_rate": 5e-06,
|
| 822 |
+
"loss": 0.4856,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.948717948717949,
|
| 827 |
+
"grad_norm": 0.5227168202400208,
|
| 828 |
+
"learning_rate": 5e-06,
|
| 829 |
+
"loss": 0.4888,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.9743589743589745,
|
| 834 |
+
"grad_norm": 0.4714514911174774,
|
| 835 |
+
"learning_rate": 5e-06,
|
| 836 |
+
"loss": 0.4913,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.0,
|
| 841 |
+
"grad_norm": 0.5580222010612488,
|
| 842 |
+
"learning_rate": 5e-06,
|
| 843 |
+
"loss": 0.4851,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 3.0,
|
| 848 |
+
"eval_loss": 0.5471954345703125,
|
| 849 |
+
"eval_runtime": 35.7738,
|
| 850 |
+
"eval_samples_per_second": 293.622,
|
| 851 |
+
"eval_steps_per_second": 1.174,
|
| 852 |
+
"step": 1170
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 3.0,
|
| 856 |
+
"step": 1170,
|
| 857 |
+
"total_flos": 5.524377893797154e+19,
|
| 858 |
+
"train_loss": 0.5494362928928473,
|
| 859 |
+
"train_runtime": 8709.0296,
|
| 860 |
+
"train_samples_per_second": 68.748,
|
| 861 |
+
"train_steps_per_second": 0.134
|
| 862 |
+
}
|
| 863 |
+
],
|
| 864 |
+
"logging_steps": 10,
|
| 865 |
+
"max_steps": 1170,
|
| 866 |
+
"num_input_tokens_seen": 0,
|
| 867 |
+
"num_train_epochs": 3,
|
| 868 |
+
"save_steps": 500,
|
| 869 |
+
"stateful_callbacks": {
|
| 870 |
+
"TrainerControl": {
|
| 871 |
+
"args": {
|
| 872 |
+
"should_epoch_stop": false,
|
| 873 |
+
"should_evaluate": false,
|
| 874 |
+
"should_log": false,
|
| 875 |
+
"should_save": true,
|
| 876 |
+
"should_training_stop": true
|
| 877 |
+
},
|
| 878 |
+
"attributes": {}
|
| 879 |
+
}
|
| 880 |
+
},
|
| 881 |
+
"total_flos": 5.524377893797154e+19,
|
| 882 |
+
"train_batch_size": 16,
|
| 883 |
+
"trial_name": null,
|
| 884 |
+
"trial_params": null
|
| 885 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|