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_unnatural_instructions
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama3-1_8b_oh_dcft_v1_wo_unnatural_instructions
|
| 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: 0.5410
|
| 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_unnatural_instructions
|
|
|
|
| 16 |
|
| 17 |
# llama3-1_8b_oh_dcft_v1_wo_unnatural_instructions
|
| 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_unnatural_instructions dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5410
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5409571528434753,
|
| 4 |
+
"eval_runtime": 35.8031,
|
| 5 |
+
"eval_samples_per_second": 295.142,
|
| 6 |
+
"eval_steps_per_second": 1.173,
|
| 7 |
+
"total_flos": 5.566873103382662e+19,
|
| 8 |
+
"train_loss": 0.5357507543102779,
|
| 9 |
+
"train_runtime": 8300.7415,
|
| 10 |
+
"train_samples_per_second": 72.558,
|
| 11 |
+
"train_steps_per_second": 0.142
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5409571528434753,
|
| 4 |
+
"eval_runtime": 35.8031,
|
| 5 |
+
"eval_samples_per_second": 295.142,
|
| 6 |
+
"eval_steps_per_second": 1.173
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 5.566873103382662e+19,
|
| 4 |
+
"train_loss": 0.5357507543102779,
|
| 5 |
+
"train_runtime": 8300.7415,
|
| 6 |
+
"train_samples_per_second": 72.558,
|
| 7 |
+
"train_steps_per_second": 0.142
|
| 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": 1179,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02544529262086514,
|
| 13 |
+
"grad_norm": 3.0892276763916016,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.8937,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05089058524173028,
|
| 20 |
+
"grad_norm": 1.0776044130325317,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7655,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07633587786259542,
|
| 27 |
+
"grad_norm": 2.4046175479888916,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7156,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10178117048346055,
|
| 34 |
+
"grad_norm": 1.0319740772247314,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6884,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1272264631043257,
|
| 41 |
+
"grad_norm": 0.8365700840950012,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6666,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15267175572519084,
|
| 48 |
+
"grad_norm": 1.0443907976150513,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6517,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.178117048346056,
|
| 55 |
+
"grad_norm": 0.6737371683120728,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6476,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2035623409669211,
|
| 62 |
+
"grad_norm": 0.7133864760398865,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6313,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.22900763358778625,
|
| 69 |
+
"grad_norm": 0.873466432094574,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6223,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2544529262086514,
|
| 76 |
+
"grad_norm": 0.8450167179107666,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.6252,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.27989821882951654,
|
| 83 |
+
"grad_norm": 0.5868242383003235,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6153,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3053435114503817,
|
| 90 |
+
"grad_norm": 0.6943073868751526,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6155,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.33078880407124683,
|
| 97 |
+
"grad_norm": 0.568712592124939,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.6086,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.356234096692112,
|
| 104 |
+
"grad_norm": 0.5304349660873413,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6046,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3816793893129771,
|
| 111 |
+
"grad_norm": 0.5914269089698792,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.6014,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4071246819338422,
|
| 118 |
+
"grad_norm": 0.6255616545677185,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6059,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.43256997455470736,
|
| 125 |
+
"grad_norm": 0.6703746318817139,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.595,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4580152671755725,
|
| 132 |
+
"grad_norm": 0.6239282488822937,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.5961,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.48346055979643765,
|
| 139 |
+
"grad_norm": 0.5409078598022461,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.5936,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5089058524173028,
|
| 146 |
+
"grad_norm": 0.47156909108161926,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.5871,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5343511450381679,
|
| 153 |
+
"grad_norm": 0.6108508110046387,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.5936,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5597964376590331,
|
| 160 |
+
"grad_norm": 0.510125994682312,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.5804,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5852417302798982,
|
| 167 |
+
"grad_norm": 0.532505214214325,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.583,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6106870229007634,
|
| 174 |
+
"grad_norm": 0.732284426689148,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.5817,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6361323155216285,
|
| 181 |
+
"grad_norm": 0.624311089515686,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.5821,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6615776081424937,
|
| 188 |
+
"grad_norm": 0.6264485120773315,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.5805,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6870229007633588,
|
| 195 |
+
"grad_norm": 0.5334043502807617,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.5818,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.712468193384224,
|
| 202 |
+
"grad_norm": 0.5711916089057922,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.5768,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7379134860050891,
|
| 209 |
+
"grad_norm": 0.4914567172527313,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.5733,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7633587786259542,
|
| 216 |
+
"grad_norm": 0.4886472523212433,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.5827,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7888040712468194,
|
| 223 |
+
"grad_norm": 0.5745688080787659,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.5756,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8142493638676844,
|
| 230 |
+
"grad_norm": 0.6722710132598877,
|
| 231 |
+
"learning_rate": 5e-06,
|
| 232 |
+
"loss": 0.5722,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8396946564885496,
|
| 237 |
+
"grad_norm": 0.5262378454208374,
|
| 238 |
+
"learning_rate": 5e-06,
|
| 239 |
+
"loss": 0.5704,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8651399491094147,
|
| 244 |
+
"grad_norm": 0.5720959305763245,
|
| 245 |
+
"learning_rate": 5e-06,
|
| 246 |
+
"loss": 0.5739,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8905852417302799,
|
| 251 |
+
"grad_norm": 0.525618314743042,
|
| 252 |
+
"learning_rate": 5e-06,
|
| 253 |
+
"loss": 0.5648,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.916030534351145,
|
| 258 |
+
"grad_norm": 0.5551468133926392,
|
| 259 |
+
"learning_rate": 5e-06,
|
| 260 |
+
"loss": 0.5726,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.9414758269720102,
|
| 265 |
+
"grad_norm": 0.5689135193824768,
|
| 266 |
+
"learning_rate": 5e-06,
|
| 267 |
+
"loss": 0.5644,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.9669211195928753,
|
| 272 |
+
"grad_norm": 0.5880911946296692,
|
| 273 |
+
"learning_rate": 5e-06,
|
| 274 |
+
"loss": 0.5628,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.9923664122137404,
|
| 279 |
+
"grad_norm": 0.5270939469337463,
|
| 280 |
+
"learning_rate": 5e-06,
|
| 281 |
+
"loss": 0.5649,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.0,
|
| 286 |
+
"eval_loss": 0.5690909624099731,
|
| 287 |
+
"eval_runtime": 35.8687,
|
| 288 |
+
"eval_samples_per_second": 294.602,
|
| 289 |
+
"eval_steps_per_second": 1.171,
|
| 290 |
+
"step": 393
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.0178117048346056,
|
| 294 |
+
"grad_norm": 0.6128649115562439,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5446,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.0432569974554706,
|
| 301 |
+
"grad_norm": 0.6446624398231506,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5287,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.0687022900763359,
|
| 308 |
+
"grad_norm": 0.6626042127609253,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.528,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.094147582697201,
|
| 315 |
+
"grad_norm": 0.6851757168769836,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.5275,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.1195928753180662,
|
| 322 |
+
"grad_norm": 0.5607665181159973,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.5317,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.1450381679389312,
|
| 329 |
+
"grad_norm": 0.5421668887138367,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.5279,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.1704834605597965,
|
| 336 |
+
"grad_norm": 0.5087480545043945,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.5323,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.1959287531806615,
|
| 343 |
+
"grad_norm": 0.6041454672813416,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.5221,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.2213740458015268,
|
| 350 |
+
"grad_norm": 0.5389483571052551,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.5296,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.2468193384223918,
|
| 357 |
+
"grad_norm": 0.5498439073562622,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.5236,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.272264631043257,
|
| 364 |
+
"grad_norm": 0.534608781337738,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.5274,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.297709923664122,
|
| 371 |
+
"grad_norm": 0.5173637866973877,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.5234,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.3231552162849873,
|
| 378 |
+
"grad_norm": 0.640381932258606,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.5191,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.3486005089058524,
|
| 385 |
+
"grad_norm": 0.636634111404419,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.5274,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.3740458015267176,
|
| 392 |
+
"grad_norm": 0.4825289249420166,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.5271,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.3994910941475827,
|
| 399 |
+
"grad_norm": 0.7268803715705872,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.5239,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.424936386768448,
|
| 406 |
+
"grad_norm": 0.5402505993843079,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.5212,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.450381679389313,
|
| 413 |
+
"grad_norm": 0.5150734186172485,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.5147,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.4758269720101782,
|
| 420 |
+
"grad_norm": 0.5088803768157959,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.5238,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.5012722646310432,
|
| 427 |
+
"grad_norm": 0.5912160277366638,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.5226,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.5267175572519083,
|
| 434 |
+
"grad_norm": 0.5235996842384338,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.5161,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.5521628498727735,
|
| 441 |
+
"grad_norm": 0.6075550317764282,
|
| 442 |
+
"learning_rate": 5e-06,
|
| 443 |
+
"loss": 0.5157,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.5776081424936388,
|
| 448 |
+
"grad_norm": 0.5467493534088135,
|
| 449 |
+
"learning_rate": 5e-06,
|
| 450 |
+
"loss": 0.5173,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.6030534351145038,
|
| 455 |
+
"grad_norm": 0.5767125487327576,
|
| 456 |
+
"learning_rate": 5e-06,
|
| 457 |
+
"loss": 0.518,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.6284987277353689,
|
| 462 |
+
"grad_norm": 0.6012499332427979,
|
| 463 |
+
"learning_rate": 5e-06,
|
| 464 |
+
"loss": 0.5136,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.6539440203562341,
|
| 469 |
+
"grad_norm": 0.5314013361930847,
|
| 470 |
+
"learning_rate": 5e-06,
|
| 471 |
+
"loss": 0.5175,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.6793893129770994,
|
| 476 |
+
"grad_norm": 0.6663753390312195,
|
| 477 |
+
"learning_rate": 5e-06,
|
| 478 |
+
"loss": 0.5211,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.7048346055979644,
|
| 483 |
+
"grad_norm": 0.6120356917381287,
|
| 484 |
+
"learning_rate": 5e-06,
|
| 485 |
+
"loss": 0.5228,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.7302798982188294,
|
| 490 |
+
"grad_norm": 0.6509705781936646,
|
| 491 |
+
"learning_rate": 5e-06,
|
| 492 |
+
"loss": 0.5177,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.7557251908396947,
|
| 497 |
+
"grad_norm": 0.5193184614181519,
|
| 498 |
+
"learning_rate": 5e-06,
|
| 499 |
+
"loss": 0.5147,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.78117048346056,
|
| 504 |
+
"grad_norm": 0.5160722136497498,
|
| 505 |
+
"learning_rate": 5e-06,
|
| 506 |
+
"loss": 0.5167,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.806615776081425,
|
| 511 |
+
"grad_norm": 0.6566229462623596,
|
| 512 |
+
"learning_rate": 5e-06,
|
| 513 |
+
"loss": 0.5116,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.83206106870229,
|
| 518 |
+
"grad_norm": 0.4844520688056946,
|
| 519 |
+
"learning_rate": 5e-06,
|
| 520 |
+
"loss": 0.5125,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.8575063613231553,
|
| 525 |
+
"grad_norm": 0.5609481334686279,
|
| 526 |
+
"learning_rate": 5e-06,
|
| 527 |
+
"loss": 0.5134,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.8829516539440203,
|
| 532 |
+
"grad_norm": 0.4838588535785675,
|
| 533 |
+
"learning_rate": 5e-06,
|
| 534 |
+
"loss": 0.5111,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.9083969465648853,
|
| 539 |
+
"grad_norm": 0.5089361667633057,
|
| 540 |
+
"learning_rate": 5e-06,
|
| 541 |
+
"loss": 0.5147,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.9338422391857506,
|
| 546 |
+
"grad_norm": 0.4935343861579895,
|
| 547 |
+
"learning_rate": 5e-06,
|
| 548 |
+
"loss": 0.5092,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.9592875318066159,
|
| 553 |
+
"grad_norm": 0.5261263251304626,
|
| 554 |
+
"learning_rate": 5e-06,
|
| 555 |
+
"loss": 0.5092,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.984732824427481,
|
| 560 |
+
"grad_norm": 0.5327720642089844,
|
| 561 |
+
"learning_rate": 5e-06,
|
| 562 |
+
"loss": 0.5093,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 2.0,
|
| 567 |
+
"eval_loss": 0.5446056127548218,
|
| 568 |
+
"eval_runtime": 35.403,
|
| 569 |
+
"eval_samples_per_second": 298.478,
|
| 570 |
+
"eval_steps_per_second": 1.186,
|
| 571 |
+
"step": 786
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.010178117048346,
|
| 575 |
+
"grad_norm": 0.7140676975250244,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.4964,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.035623409669211,
|
| 582 |
+
"grad_norm": 0.5526481866836548,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.4699,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.0610687022900764,
|
| 589 |
+
"grad_norm": 0.8097620010375977,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.4747,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.0865139949109412,
|
| 596 |
+
"grad_norm": 0.5774556994438171,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.4731,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.1119592875318065,
|
| 603 |
+
"grad_norm": 0.7799936532974243,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.481,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.1374045801526718,
|
| 610 |
+
"grad_norm": 0.7324852347373962,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.472,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.162849872773537,
|
| 617 |
+
"grad_norm": 0.6810862421989441,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.4788,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.188295165394402,
|
| 624 |
+
"grad_norm": 0.6535518169403076,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.4787,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.213740458015267,
|
| 631 |
+
"grad_norm": 0.5570102334022522,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.4785,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.2391857506361323,
|
| 638 |
+
"grad_norm": 0.5471156239509583,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.4817,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.2646310432569976,
|
| 645 |
+
"grad_norm": 0.5434932112693787,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.4732,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.2900763358778624,
|
| 652 |
+
"grad_norm": 0.5234562754631042,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.4742,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.3155216284987277,
|
| 659 |
+
"grad_norm": 0.5137863159179688,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.4725,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.340966921119593,
|
| 666 |
+
"grad_norm": 0.5251743197441101,
|
| 667 |
+
"learning_rate": 5e-06,
|
| 668 |
+
"loss": 0.4793,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.366412213740458,
|
| 673 |
+
"grad_norm": 0.5063709020614624,
|
| 674 |
+
"learning_rate": 5e-06,
|
| 675 |
+
"loss": 0.4735,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.391857506361323,
|
| 680 |
+
"grad_norm": 0.5757549405097961,
|
| 681 |
+
"learning_rate": 5e-06,
|
| 682 |
+
"loss": 0.4728,
|
| 683 |
+
"step": 940
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.4173027989821882,
|
| 687 |
+
"grad_norm": 0.48364320397377014,
|
| 688 |
+
"learning_rate": 5e-06,
|
| 689 |
+
"loss": 0.4785,
|
| 690 |
+
"step": 950
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.4427480916030535,
|
| 694 |
+
"grad_norm": 0.5600482225418091,
|
| 695 |
+
"learning_rate": 5e-06,
|
| 696 |
+
"loss": 0.4748,
|
| 697 |
+
"step": 960
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.4681933842239188,
|
| 701 |
+
"grad_norm": 0.5102260708808899,
|
| 702 |
+
"learning_rate": 5e-06,
|
| 703 |
+
"loss": 0.4752,
|
| 704 |
+
"step": 970
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.4936386768447836,
|
| 708 |
+
"grad_norm": 0.642202615737915,
|
| 709 |
+
"learning_rate": 5e-06,
|
| 710 |
+
"loss": 0.4743,
|
| 711 |
+
"step": 980
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.519083969465649,
|
| 715 |
+
"grad_norm": 0.7704691886901855,
|
| 716 |
+
"learning_rate": 5e-06,
|
| 717 |
+
"loss": 0.4743,
|
| 718 |
+
"step": 990
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.544529262086514,
|
| 722 |
+
"grad_norm": 0.6885623931884766,
|
| 723 |
+
"learning_rate": 5e-06,
|
| 724 |
+
"loss": 0.4742,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.569974554707379,
|
| 729 |
+
"grad_norm": 0.5846525430679321,
|
| 730 |
+
"learning_rate": 5e-06,
|
| 731 |
+
"loss": 0.4734,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.595419847328244,
|
| 736 |
+
"grad_norm": 0.66885906457901,
|
| 737 |
+
"learning_rate": 5e-06,
|
| 738 |
+
"loss": 0.4758,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.6208651399491094,
|
| 743 |
+
"grad_norm": 0.5633515119552612,
|
| 744 |
+
"learning_rate": 5e-06,
|
| 745 |
+
"loss": 0.474,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.6463104325699747,
|
| 750 |
+
"grad_norm": 0.5127110481262207,
|
| 751 |
+
"learning_rate": 5e-06,
|
| 752 |
+
"loss": 0.4792,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.67175572519084,
|
| 757 |
+
"grad_norm": 0.6211147308349609,
|
| 758 |
+
"learning_rate": 5e-06,
|
| 759 |
+
"loss": 0.4721,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.6972010178117047,
|
| 764 |
+
"grad_norm": 0.700775146484375,
|
| 765 |
+
"learning_rate": 5e-06,
|
| 766 |
+
"loss": 0.4731,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.72264631043257,
|
| 771 |
+
"grad_norm": 0.5291451215744019,
|
| 772 |
+
"learning_rate": 5e-06,
|
| 773 |
+
"loss": 0.475,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.7480916030534353,
|
| 778 |
+
"grad_norm": 0.5429026484489441,
|
| 779 |
+
"learning_rate": 5e-06,
|
| 780 |
+
"loss": 0.4822,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.7735368956743,
|
| 785 |
+
"grad_norm": 0.5198021531105042,
|
| 786 |
+
"learning_rate": 5e-06,
|
| 787 |
+
"loss": 0.4738,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.7989821882951653,
|
| 792 |
+
"grad_norm": 0.4582120478153229,
|
| 793 |
+
"learning_rate": 5e-06,
|
| 794 |
+
"loss": 0.4709,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.8244274809160306,
|
| 799 |
+
"grad_norm": 0.47587454319000244,
|
| 800 |
+
"learning_rate": 5e-06,
|
| 801 |
+
"loss": 0.4692,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.849872773536896,
|
| 806 |
+
"grad_norm": 0.5221173167228699,
|
| 807 |
+
"learning_rate": 5e-06,
|
| 808 |
+
"loss": 0.4767,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.875318066157761,
|
| 813 |
+
"grad_norm": 0.5617064833641052,
|
| 814 |
+
"learning_rate": 5e-06,
|
| 815 |
+
"loss": 0.478,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.900763358778626,
|
| 820 |
+
"grad_norm": 0.5362435579299927,
|
| 821 |
+
"learning_rate": 5e-06,
|
| 822 |
+
"loss": 0.4745,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.926208651399491,
|
| 827 |
+
"grad_norm": 0.5266700387001038,
|
| 828 |
+
"learning_rate": 5e-06,
|
| 829 |
+
"loss": 0.4725,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.9516539440203564,
|
| 834 |
+
"grad_norm": 0.5272775292396545,
|
| 835 |
+
"learning_rate": 5e-06,
|
| 836 |
+
"loss": 0.4744,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.9770992366412212,
|
| 841 |
+
"grad_norm": 0.6244516372680664,
|
| 842 |
+
"learning_rate": 5e-06,
|
| 843 |
+
"loss": 0.4766,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 3.0,
|
| 848 |
+
"eval_loss": 0.5409571528434753,
|
| 849 |
+
"eval_runtime": 35.9061,
|
| 850 |
+
"eval_samples_per_second": 294.296,
|
| 851 |
+
"eval_steps_per_second": 1.17,
|
| 852 |
+
"step": 1179
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 3.0,
|
| 856 |
+
"step": 1179,
|
| 857 |
+
"total_flos": 5.566873103382662e+19,
|
| 858 |
+
"train_loss": 0.5357507543102779,
|
| 859 |
+
"train_runtime": 8300.7415,
|
| 860 |
+
"train_samples_per_second": 72.558,
|
| 861 |
+
"train_steps_per_second": 0.142
|
| 862 |
+
}
|
| 863 |
+
],
|
| 864 |
+
"logging_steps": 10,
|
| 865 |
+
"max_steps": 1179,
|
| 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": 5.566873103382662e+19,
|
| 882 |
+
"train_batch_size": 16,
|
| 883 |
+
"trial_name": null,
|
| 884 |
+
"trial_params": null
|
| 885 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|