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 +1005 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: gemma
|
|
| 4 |
base_model: google/gemma-2-9b
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: hp_ablations_gemma_epoch4_dcftv1.2
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_gemma_epoch4_dcftv1.2
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.6798
|
| 21 |
|
|
|
|
| 4 |
base_model: google/gemma-2-9b
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: hp_ablations_gemma_epoch4_dcftv1.2
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_gemma_epoch4_dcftv1.2
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b) on the mlfoundations-dev/oh-dcft-v1.2_no-curation_gpt-4o-mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6798
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.9985035540591096,
|
| 3 |
+
"eval_loss": 0.6797953248023987,
|
| 4 |
+
"eval_runtime": 269.0155,
|
| 5 |
+
"eval_samples_per_second": 33.459,
|
| 6 |
+
"eval_steps_per_second": 0.524,
|
| 7 |
+
"total_flos": 5090631865073664.0,
|
| 8 |
+
"train_loss": 0.5299089724075294,
|
| 9 |
+
"train_runtime": 62096.6704,
|
| 10 |
+
"train_samples_per_second": 11.016,
|
| 11 |
+
"train_steps_per_second": 0.022
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.9985035540591096,
|
| 3 |
+
"eval_loss": 0.6797953248023987,
|
| 4 |
+
"eval_runtime": 269.0155,
|
| 5 |
+
"eval_samples_per_second": 33.459,
|
| 6 |
+
"eval_steps_per_second": 0.524
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.9985035540591096,
|
| 3 |
+
"total_flos": 5090631865073664.0,
|
| 4 |
+
"train_loss": 0.5299089724075294,
|
| 5 |
+
"train_runtime": 62096.6704,
|
| 6 |
+
"train_samples_per_second": 11.016,
|
| 7 |
+
"train_steps_per_second": 0.022
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1005 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.9985035540591096,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1336,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.029928918817807706,
|
| 13 |
+
"grad_norm": 1.3197502899687559,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.7729,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05985783763561541,
|
| 20 |
+
"grad_norm": 0.7957246856225982,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.694,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08978675645342311,
|
| 27 |
+
"grad_norm": 0.7023919119728642,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6692,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.11971567527123082,
|
| 34 |
+
"grad_norm": 0.6852471491957217,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6652,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.14964459408903855,
|
| 41 |
+
"grad_norm": 0.7441284057804172,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6567,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17957351290684623,
|
| 48 |
+
"grad_norm": 0.6737131800519109,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6489,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.20950243172465394,
|
| 55 |
+
"grad_norm": 0.8645698704938743,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6458,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.23943135054246165,
|
| 62 |
+
"grad_norm": 0.6824403788440216,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6472,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.26936026936026936,
|
| 69 |
+
"grad_norm": 0.8355879527708924,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6382,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2992891881780771,
|
| 76 |
+
"grad_norm": 0.6566317269166482,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6394,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3292181069958848,
|
| 83 |
+
"grad_norm": 0.7025002610859795,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6352,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.35914702581369246,
|
| 90 |
+
"grad_norm": 0.7294514273893201,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6341,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3890759446315002,
|
| 97 |
+
"grad_norm": 0.7204998726570041,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6342,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4190048634493079,
|
| 104 |
+
"grad_norm": 0.9245929000779519,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6279,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.4489337822671156,
|
| 111 |
+
"grad_norm": 0.8312008040431372,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6298,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4788627010849233,
|
| 118 |
+
"grad_norm": 0.6941447661619787,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6287,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.508791619902731,
|
| 125 |
+
"grad_norm": 0.7880713474277835,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.623,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5387205387205387,
|
| 132 |
+
"grad_norm": 0.7199931353143368,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6247,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.5686494575383464,
|
| 139 |
+
"grad_norm": 0.7680737861171968,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6223,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5985783763561542,
|
| 146 |
+
"grad_norm": 0.7601491643468152,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6299,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6285072951739619,
|
| 153 |
+
"grad_norm": 0.76786698349262,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.6236,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.6584362139917695,
|
| 160 |
+
"grad_norm": 0.9029566246000676,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6224,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6883651328095772,
|
| 167 |
+
"grad_norm": 0.7045261038164553,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.6245,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.7182940516273849,
|
| 174 |
+
"grad_norm": 0.6774924026654922,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6234,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.7482229704451927,
|
| 181 |
+
"grad_norm": 0.6011441610841004,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.6201,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.7781518892630004,
|
| 188 |
+
"grad_norm": 0.6589701033868924,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.6188,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.8080808080808081,
|
| 195 |
+
"grad_norm": 0.7793955701511873,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.6263,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.8380097268986157,
|
| 202 |
+
"grad_norm": 0.6801997659823543,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.6164,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.8679386457164235,
|
| 209 |
+
"grad_norm": 0.7863034594758392,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.6133,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.8978675645342312,
|
| 216 |
+
"grad_norm": 0.8097674195506819,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.6145,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.9277964833520389,
|
| 223 |
+
"grad_norm": 0.6976711471967793,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.6132,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.9577254021698466,
|
| 230 |
+
"grad_norm": 0.6845188306806823,
|
| 231 |
+
"learning_rate": 5e-06,
|
| 232 |
+
"loss": 0.615,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.9876543209876543,
|
| 237 |
+
"grad_norm": 0.9820656021369418,
|
| 238 |
+
"learning_rate": 5e-06,
|
| 239 |
+
"loss": 0.6085,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.9996258885147774,
|
| 244 |
+
"eval_loss": 0.6192271709442139,
|
| 245 |
+
"eval_runtime": 268.4518,
|
| 246 |
+
"eval_samples_per_second": 33.529,
|
| 247 |
+
"eval_steps_per_second": 0.525,
|
| 248 |
+
"step": 334
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.017583239805462,
|
| 252 |
+
"grad_norm": 0.8670558349646079,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.6314,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.0475121586232696,
|
| 259 |
+
"grad_norm": 1.0194196930172406,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5525,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.0774410774410774,
|
| 266 |
+
"grad_norm": 0.7571264421758325,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5475,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.1073699962588852,
|
| 273 |
+
"grad_norm": 0.716142300432686,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.5479,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.1372989150766928,
|
| 280 |
+
"grad_norm": 0.7134968159345548,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5483,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.1672278338945006,
|
| 287 |
+
"grad_norm": 0.7093422370162528,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.5497,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.1971567527123081,
|
| 294 |
+
"grad_norm": 0.6758306313904245,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5499,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.227085671530116,
|
| 301 |
+
"grad_norm": 0.6590188596738886,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5523,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.2570145903479237,
|
| 308 |
+
"grad_norm": 0.7115281205352587,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.557,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.2869435091657313,
|
| 315 |
+
"grad_norm": 0.6651956769462775,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.5521,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.316872427983539,
|
| 322 |
+
"grad_norm": 0.686904033471436,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.5542,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.3468013468013469,
|
| 329 |
+
"grad_norm": 0.7052326227629313,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.5473,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.3767302656191545,
|
| 336 |
+
"grad_norm": 0.6603203892732427,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.558,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.4066591844369623,
|
| 343 |
+
"grad_norm": 0.7204930864199384,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.557,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.43658810325477,
|
| 350 |
+
"grad_norm": 0.6582974125304011,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.5598,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.4665170220725776,
|
| 357 |
+
"grad_norm": 0.653408089340934,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.5589,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.4964459408903854,
|
| 364 |
+
"grad_norm": 0.7722703692356943,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.5549,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.5263748597081932,
|
| 371 |
+
"grad_norm": 0.6410463952946445,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.5571,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.5563037785260008,
|
| 378 |
+
"grad_norm": 0.6788292489082296,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.5567,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.5862326973438083,
|
| 385 |
+
"grad_norm": 0.7347495173956178,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.5557,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.6161616161616161,
|
| 392 |
+
"grad_norm": 0.7489697605253047,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.56,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.646090534979424,
|
| 399 |
+
"grad_norm": 0.6649594456868578,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.5596,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.6760194537972315,
|
| 406 |
+
"grad_norm": 0.6944801894329058,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.5499,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.7059483726150393,
|
| 413 |
+
"grad_norm": 0.7516636245416078,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.5519,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.735877291432847,
|
| 420 |
+
"grad_norm": 0.834145985540098,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.5525,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.7658062102506547,
|
| 427 |
+
"grad_norm": 0.6940488546001392,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.5561,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.7957351290684624,
|
| 434 |
+
"grad_norm": 0.6996951151429136,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.5568,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.8256640478862702,
|
| 441 |
+
"grad_norm": 0.6321044767548653,
|
| 442 |
+
"learning_rate": 5e-06,
|
| 443 |
+
"loss": 0.5543,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.8555929667040778,
|
| 448 |
+
"grad_norm": 0.6380400908901183,
|
| 449 |
+
"learning_rate": 5e-06,
|
| 450 |
+
"loss": 0.5521,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.8855218855218854,
|
| 455 |
+
"grad_norm": 0.6726516418071744,
|
| 456 |
+
"learning_rate": 5e-06,
|
| 457 |
+
"loss": 0.5536,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.9154508043396934,
|
| 462 |
+
"grad_norm": 0.6952484734366503,
|
| 463 |
+
"learning_rate": 5e-06,
|
| 464 |
+
"loss": 0.5556,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.945379723157501,
|
| 469 |
+
"grad_norm": 0.6339074390401458,
|
| 470 |
+
"learning_rate": 5e-06,
|
| 471 |
+
"loss": 0.554,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.9753086419753085,
|
| 476 |
+
"grad_norm": 0.8335015680516275,
|
| 477 |
+
"learning_rate": 5e-06,
|
| 478 |
+
"loss": 0.5595,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.9992517770295548,
|
| 483 |
+
"eval_loss": 0.615513026714325,
|
| 484 |
+
"eval_runtime": 267.9582,
|
| 485 |
+
"eval_samples_per_second": 33.591,
|
| 486 |
+
"eval_steps_per_second": 0.526,
|
| 487 |
+
"step": 668
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.0052375607931165,
|
| 491 |
+
"grad_norm": 1.095470898789856,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.5957,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.035166479610924,
|
| 498 |
+
"grad_norm": 0.7925440628175368,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.4851,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.0650953984287317,
|
| 505 |
+
"grad_norm": 0.7600969395946293,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.4799,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.0950243172465393,
|
| 512 |
+
"grad_norm": 0.8685890982294241,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.4845,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.1249532360643473,
|
| 519 |
+
"grad_norm": 0.7159762779954674,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.4862,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.154882154882155,
|
| 526 |
+
"grad_norm": 0.7850425626912287,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.4882,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.1848110736999624,
|
| 533 |
+
"grad_norm": 0.7829173560959974,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.4894,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.2147399925177704,
|
| 540 |
+
"grad_norm": 0.7053202412118417,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.4898,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.244668911335578,
|
| 547 |
+
"grad_norm": 0.7275797577145928,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.4917,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.2745978301533856,
|
| 554 |
+
"grad_norm": 0.684143630508004,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.4878,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.3045267489711936,
|
| 561 |
+
"grad_norm": 0.778690697436679,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.4936,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.334455667789001,
|
| 568 |
+
"grad_norm": 0.6973756438711023,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.4885,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.3643845866068087,
|
| 575 |
+
"grad_norm": 0.7512378015475496,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.4902,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.3943135054246163,
|
| 582 |
+
"grad_norm": 0.6954041240036626,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.4947,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.4242424242424243,
|
| 589 |
+
"grad_norm": 0.7661445266388807,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.4995,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.454171343060232,
|
| 596 |
+
"grad_norm": 0.7288724567709918,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.4979,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.48410026187804,
|
| 603 |
+
"grad_norm": 0.7507674417043292,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.4968,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.5140291806958475,
|
| 610 |
+
"grad_norm": 0.6886877322873068,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.4951,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.543958099513655,
|
| 617 |
+
"grad_norm": 0.710314562589874,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.498,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.5738870183314626,
|
| 624 |
+
"grad_norm": 0.6994762876301733,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.4959,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.6038159371492706,
|
| 631 |
+
"grad_norm": 0.7582356365854407,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.4938,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.633744855967078,
|
| 638 |
+
"grad_norm": 0.731935619090177,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.4921,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.6636737747848858,
|
| 645 |
+
"grad_norm": 0.74782144362319,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.4977,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.6936026936026938,
|
| 652 |
+
"grad_norm": 0.6942188030457457,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.5011,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.7235316124205013,
|
| 659 |
+
"grad_norm": 0.6881327401867374,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.4947,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.753460531238309,
|
| 666 |
+
"grad_norm": 0.6856202337817314,
|
| 667 |
+
"learning_rate": 5e-06,
|
| 668 |
+
"loss": 0.4959,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.7833894500561165,
|
| 673 |
+
"grad_norm": 0.7141040450174527,
|
| 674 |
+
"learning_rate": 5e-06,
|
| 675 |
+
"loss": 0.5016,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.8133183688739245,
|
| 680 |
+
"grad_norm": 0.6857610208401852,
|
| 681 |
+
"learning_rate": 5e-06,
|
| 682 |
+
"loss": 0.4985,
|
| 683 |
+
"step": 940
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.843247287691732,
|
| 687 |
+
"grad_norm": 0.6698180625003869,
|
| 688 |
+
"learning_rate": 5e-06,
|
| 689 |
+
"loss": 0.4986,
|
| 690 |
+
"step": 950
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.87317620650954,
|
| 694 |
+
"grad_norm": 0.8039520213911328,
|
| 695 |
+
"learning_rate": 5e-06,
|
| 696 |
+
"loss": 0.502,
|
| 697 |
+
"step": 960
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.9031051253273477,
|
| 701 |
+
"grad_norm": 0.7415409936401505,
|
| 702 |
+
"learning_rate": 5e-06,
|
| 703 |
+
"loss": 0.4959,
|
| 704 |
+
"step": 970
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.9330340441451552,
|
| 708 |
+
"grad_norm": 0.7856625436324756,
|
| 709 |
+
"learning_rate": 5e-06,
|
| 710 |
+
"loss": 0.5023,
|
| 711 |
+
"step": 980
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.962962962962963,
|
| 715 |
+
"grad_norm": 0.761345605606732,
|
| 716 |
+
"learning_rate": 5e-06,
|
| 717 |
+
"loss": 0.5048,
|
| 718 |
+
"step": 990
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.992891881780771,
|
| 722 |
+
"grad_norm": 0.7302412373936236,
|
| 723 |
+
"learning_rate": 5e-06,
|
| 724 |
+
"loss": 0.5047,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.9988776655443323,
|
| 729 |
+
"eval_loss": 0.6354114413261414,
|
| 730 |
+
"eval_runtime": 269.2475,
|
| 731 |
+
"eval_samples_per_second": 33.43,
|
| 732 |
+
"eval_steps_per_second": 0.524,
|
| 733 |
+
"step": 1002
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 3.0228208005985784,
|
| 737 |
+
"grad_norm": 1.1087903273234458,
|
| 738 |
+
"learning_rate": 5e-06,
|
| 739 |
+
"loss": 0.4757,
|
| 740 |
+
"step": 1010
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 3.052749719416386,
|
| 744 |
+
"grad_norm": 0.8684090527265274,
|
| 745 |
+
"learning_rate": 5e-06,
|
| 746 |
+
"loss": 0.4189,
|
| 747 |
+
"step": 1020
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.082678638234194,
|
| 751 |
+
"grad_norm": 0.8005907139036144,
|
| 752 |
+
"learning_rate": 5e-06,
|
| 753 |
+
"loss": 0.42,
|
| 754 |
+
"step": 1030
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 3.1126075570520015,
|
| 758 |
+
"grad_norm": 0.7642262444187367,
|
| 759 |
+
"learning_rate": 5e-06,
|
| 760 |
+
"loss": 0.4184,
|
| 761 |
+
"step": 1040
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.142536475869809,
|
| 765 |
+
"grad_norm": 0.7766668581347643,
|
| 766 |
+
"learning_rate": 5e-06,
|
| 767 |
+
"loss": 0.4173,
|
| 768 |
+
"step": 1050
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.1724653946876167,
|
| 772 |
+
"grad_norm": 0.7661079163483049,
|
| 773 |
+
"learning_rate": 5e-06,
|
| 774 |
+
"loss": 0.426,
|
| 775 |
+
"step": 1060
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 3.2023943135054247,
|
| 779 |
+
"grad_norm": 0.8204327830042761,
|
| 780 |
+
"learning_rate": 5e-06,
|
| 781 |
+
"loss": 0.4192,
|
| 782 |
+
"step": 1070
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 3.2323232323232323,
|
| 786 |
+
"grad_norm": 0.7418029541161775,
|
| 787 |
+
"learning_rate": 5e-06,
|
| 788 |
+
"loss": 0.4248,
|
| 789 |
+
"step": 1080
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.2622521511410403,
|
| 793 |
+
"grad_norm": 0.7688720105233667,
|
| 794 |
+
"learning_rate": 5e-06,
|
| 795 |
+
"loss": 0.4309,
|
| 796 |
+
"step": 1090
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 3.292181069958848,
|
| 800 |
+
"grad_norm": 0.783937291766432,
|
| 801 |
+
"learning_rate": 5e-06,
|
| 802 |
+
"loss": 0.4265,
|
| 803 |
+
"step": 1100
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 3.3221099887766554,
|
| 807 |
+
"grad_norm": 0.7522379239239885,
|
| 808 |
+
"learning_rate": 5e-06,
|
| 809 |
+
"loss": 0.4247,
|
| 810 |
+
"step": 1110
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 3.352038907594463,
|
| 814 |
+
"grad_norm": 0.7694503676828072,
|
| 815 |
+
"learning_rate": 5e-06,
|
| 816 |
+
"loss": 0.4247,
|
| 817 |
+
"step": 1120
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 3.381967826412271,
|
| 821 |
+
"grad_norm": 0.7524030817183468,
|
| 822 |
+
"learning_rate": 5e-06,
|
| 823 |
+
"loss": 0.427,
|
| 824 |
+
"step": 1130
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 3.4118967452300786,
|
| 828 |
+
"grad_norm": 0.7702225573698022,
|
| 829 |
+
"learning_rate": 5e-06,
|
| 830 |
+
"loss": 0.4296,
|
| 831 |
+
"step": 1140
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 3.441825664047886,
|
| 835 |
+
"grad_norm": 0.7621134864427016,
|
| 836 |
+
"learning_rate": 5e-06,
|
| 837 |
+
"loss": 0.4272,
|
| 838 |
+
"step": 1150
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 3.471754582865694,
|
| 842 |
+
"grad_norm": 0.7536787003193455,
|
| 843 |
+
"learning_rate": 5e-06,
|
| 844 |
+
"loss": 0.4298,
|
| 845 |
+
"step": 1160
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.5016835016835017,
|
| 849 |
+
"grad_norm": 0.9135172996253528,
|
| 850 |
+
"learning_rate": 5e-06,
|
| 851 |
+
"loss": 0.4299,
|
| 852 |
+
"step": 1170
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 3.5316124205013093,
|
| 856 |
+
"grad_norm": 0.7993783814738175,
|
| 857 |
+
"learning_rate": 5e-06,
|
| 858 |
+
"loss": 0.4299,
|
| 859 |
+
"step": 1180
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 3.561541339319117,
|
| 863 |
+
"grad_norm": 0.7454852579268769,
|
| 864 |
+
"learning_rate": 5e-06,
|
| 865 |
+
"loss": 0.436,
|
| 866 |
+
"step": 1190
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 3.591470258136925,
|
| 870 |
+
"grad_norm": 0.7472361919942885,
|
| 871 |
+
"learning_rate": 5e-06,
|
| 872 |
+
"loss": 0.4307,
|
| 873 |
+
"step": 1200
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 3.6213991769547325,
|
| 877 |
+
"grad_norm": 0.7709040310826254,
|
| 878 |
+
"learning_rate": 5e-06,
|
| 879 |
+
"loss": 0.435,
|
| 880 |
+
"step": 1210
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 3.6513280957725405,
|
| 884 |
+
"grad_norm": 0.9162628098304457,
|
| 885 |
+
"learning_rate": 5e-06,
|
| 886 |
+
"loss": 0.4366,
|
| 887 |
+
"step": 1220
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 3.681257014590348,
|
| 891 |
+
"grad_norm": 0.748944264236295,
|
| 892 |
+
"learning_rate": 5e-06,
|
| 893 |
+
"loss": 0.4328,
|
| 894 |
+
"step": 1230
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 3.7111859334081556,
|
| 898 |
+
"grad_norm": 0.7943321429757056,
|
| 899 |
+
"learning_rate": 5e-06,
|
| 900 |
+
"loss": 0.4423,
|
| 901 |
+
"step": 1240
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 3.741114852225963,
|
| 905 |
+
"grad_norm": 0.7732334849903745,
|
| 906 |
+
"learning_rate": 5e-06,
|
| 907 |
+
"loss": 0.4348,
|
| 908 |
+
"step": 1250
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 3.771043771043771,
|
| 912 |
+
"grad_norm": 0.9404932255120924,
|
| 913 |
+
"learning_rate": 5e-06,
|
| 914 |
+
"loss": 0.4341,
|
| 915 |
+
"step": 1260
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 3.8009726898615788,
|
| 919 |
+
"grad_norm": 0.836108708076803,
|
| 920 |
+
"learning_rate": 5e-06,
|
| 921 |
+
"loss": 0.4417,
|
| 922 |
+
"step": 1270
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 3.8309016086793863,
|
| 926 |
+
"grad_norm": 0.708913428348208,
|
| 927 |
+
"learning_rate": 5e-06,
|
| 928 |
+
"loss": 0.4385,
|
| 929 |
+
"step": 1280
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 3.8608305274971944,
|
| 933 |
+
"grad_norm": 0.7741293535971421,
|
| 934 |
+
"learning_rate": 5e-06,
|
| 935 |
+
"loss": 0.4389,
|
| 936 |
+
"step": 1290
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 3.890759446315002,
|
| 940 |
+
"grad_norm": 0.7445598509368653,
|
| 941 |
+
"learning_rate": 5e-06,
|
| 942 |
+
"loss": 0.4371,
|
| 943 |
+
"step": 1300
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 3.9206883651328095,
|
| 947 |
+
"grad_norm": 0.9179953662967183,
|
| 948 |
+
"learning_rate": 5e-06,
|
| 949 |
+
"loss": 0.4394,
|
| 950 |
+
"step": 1310
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 3.950617283950617,
|
| 954 |
+
"grad_norm": 0.7659677690542608,
|
| 955 |
+
"learning_rate": 5e-06,
|
| 956 |
+
"loss": 0.4351,
|
| 957 |
+
"step": 1320
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 3.980546202768425,
|
| 961 |
+
"grad_norm": 0.827226367347478,
|
| 962 |
+
"learning_rate": 5e-06,
|
| 963 |
+
"loss": 0.4386,
|
| 964 |
+
"step": 1330
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 3.9985035540591096,
|
| 968 |
+
"eval_loss": 0.6797953248023987,
|
| 969 |
+
"eval_runtime": 268.7833,
|
| 970 |
+
"eval_samples_per_second": 33.488,
|
| 971 |
+
"eval_steps_per_second": 0.525,
|
| 972 |
+
"step": 1336
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 3.9985035540591096,
|
| 976 |
+
"step": 1336,
|
| 977 |
+
"total_flos": 5090631865073664.0,
|
| 978 |
+
"train_loss": 0.5299089724075294,
|
| 979 |
+
"train_runtime": 62096.6704,
|
| 980 |
+
"train_samples_per_second": 11.016,
|
| 981 |
+
"train_steps_per_second": 0.022
|
| 982 |
+
}
|
| 983 |
+
],
|
| 984 |
+
"logging_steps": 10,
|
| 985 |
+
"max_steps": 1336,
|
| 986 |
+
"num_input_tokens_seen": 0,
|
| 987 |
+
"num_train_epochs": 4,
|
| 988 |
+
"save_steps": 500,
|
| 989 |
+
"stateful_callbacks": {
|
| 990 |
+
"TrainerControl": {
|
| 991 |
+
"args": {
|
| 992 |
+
"should_epoch_stop": false,
|
| 993 |
+
"should_evaluate": false,
|
| 994 |
+
"should_log": false,
|
| 995 |
+
"should_save": true,
|
| 996 |
+
"should_training_stop": true
|
| 997 |
+
},
|
| 998 |
+
"attributes": {}
|
| 999 |
+
}
|
| 1000 |
+
},
|
| 1001 |
+
"total_flos": 5090631865073664.0,
|
| 1002 |
+
"train_batch_size": 4,
|
| 1003 |
+
"trial_name": null,
|
| 1004 |
+
"trial_params": null
|
| 1005 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|