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 +906 -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: oh_v1.3_evol_instruct_x8
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_v1.3_evol_instruct_x8
|
| 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: 0.8070
|
| 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: oh_v1.3_evol_instruct_x8
|
|
|
|
| 16 |
|
| 17 |
# oh_v1.3_evol_instruct_x8
|
| 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_v1.3_evol_instruct_x8 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.8070
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.8070117235183716,
|
| 4 |
+
"eval_runtime": 273.8922,
|
| 5 |
+
"eval_samples_per_second": 39.348,
|
| 6 |
+
"eval_steps_per_second": 0.617,
|
| 7 |
+
"total_flos": 2009835314872320.0,
|
| 8 |
+
"train_loss": 0.7931587568918864,
|
| 9 |
+
"train_runtime": 40308.7538,
|
| 10 |
+
"train_samples_per_second": 15.238,
|
| 11 |
+
"train_steps_per_second": 0.03
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.8070117235183716,
|
| 4 |
+
"eval_runtime": 273.8922,
|
| 5 |
+
"eval_samples_per_second": 39.348,
|
| 6 |
+
"eval_steps_per_second": 0.617
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2009835314872320.0,
|
| 4 |
+
"train_loss": 0.7931587568918864,
|
| 5 |
+
"train_runtime": 40308.7538,
|
| 6 |
+
"train_samples_per_second": 15.238,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,906 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.025,
|
| 13 |
+
"grad_norm": 6.393901263952026,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.1278,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"grad_norm": 1.705956397184389,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 1.0093,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.075,
|
| 27 |
+
"grad_norm": 1.68971140105189,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.9776,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1,
|
| 34 |
+
"grad_norm": 2.2896479687833042,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.9461,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.125,
|
| 41 |
+
"grad_norm": 2.337387810419128,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.9341,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"grad_norm": 1.882007339408848,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.9133,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.175,
|
| 55 |
+
"grad_norm": 1.0405536021297026,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.9034,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2,
|
| 62 |
+
"grad_norm": 1.1273905617480575,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.8946,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.225,
|
| 69 |
+
"grad_norm": 0.8899956341459144,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.8897,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.25,
|
| 76 |
+
"grad_norm": 0.7569552721791574,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.8838,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.275,
|
| 83 |
+
"grad_norm": 0.775187782225504,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.8804,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3,
|
| 90 |
+
"grad_norm": 0.7058567462838982,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.8742,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.325,
|
| 97 |
+
"grad_norm": 1.2857109201084875,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.8689,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.35,
|
| 104 |
+
"grad_norm": 0.833965423282999,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.8664,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.375,
|
| 111 |
+
"grad_norm": 0.8452787015074426,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.8647,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.4,
|
| 118 |
+
"grad_norm": 0.6905100077591924,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.8614,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.425,
|
| 125 |
+
"grad_norm": 0.6534167059133549,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.8591,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.45,
|
| 132 |
+
"grad_norm": 0.7163070342801734,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.8538,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.475,
|
| 139 |
+
"grad_norm": 0.8379014374017074,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.8519,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5,
|
| 146 |
+
"grad_norm": 0.7058002653208569,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.8519,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.525,
|
| 153 |
+
"grad_norm": 0.7433143437252816,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.8486,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.55,
|
| 160 |
+
"grad_norm": 0.7600103941580408,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.844,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.575,
|
| 167 |
+
"grad_norm": 0.7419761211751795,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.8447,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6,
|
| 174 |
+
"grad_norm": 0.6020153457959969,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.8448,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.625,
|
| 181 |
+
"grad_norm": 0.5834136183612872,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.8415,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.65,
|
| 188 |
+
"grad_norm": 0.6861809869702314,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.8412,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.675,
|
| 195 |
+
"grad_norm": 0.6424298466812792,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.8446,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.7,
|
| 202 |
+
"grad_norm": 0.7062264581938785,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.8371,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.725,
|
| 209 |
+
"grad_norm": 0.7413013325508084,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.8368,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.75,
|
| 216 |
+
"grad_norm": 0.6755467656899127,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.8384,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.775,
|
| 223 |
+
"grad_norm": 0.7186164830975353,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.8346,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8,
|
| 230 |
+
"grad_norm": 0.5707506171843091,
|
| 231 |
+
"learning_rate": 5e-06,
|
| 232 |
+
"loss": 0.8331,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.825,
|
| 237 |
+
"grad_norm": 0.9112862077459049,
|
| 238 |
+
"learning_rate": 5e-06,
|
| 239 |
+
"loss": 0.8324,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.85,
|
| 244 |
+
"grad_norm": 0.6828679795547313,
|
| 245 |
+
"learning_rate": 5e-06,
|
| 246 |
+
"loss": 0.8354,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.875,
|
| 251 |
+
"grad_norm": 0.6927842206009195,
|
| 252 |
+
"learning_rate": 5e-06,
|
| 253 |
+
"loss": 0.829,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.9,
|
| 258 |
+
"grad_norm": 0.6008408676809543,
|
| 259 |
+
"learning_rate": 5e-06,
|
| 260 |
+
"loss": 0.8307,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.925,
|
| 265 |
+
"grad_norm": 0.6206381764947855,
|
| 266 |
+
"learning_rate": 5e-06,
|
| 267 |
+
"loss": 0.8267,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.95,
|
| 272 |
+
"grad_norm": 0.5454612469710348,
|
| 273 |
+
"learning_rate": 5e-06,
|
| 274 |
+
"loss": 0.8247,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.975,
|
| 279 |
+
"grad_norm": 0.5220893940655968,
|
| 280 |
+
"learning_rate": 5e-06,
|
| 281 |
+
"loss": 0.8256,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.0,
|
| 286 |
+
"grad_norm": 0.5672238225320995,
|
| 287 |
+
"learning_rate": 5e-06,
|
| 288 |
+
"loss": 0.8272,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0,
|
| 293 |
+
"eval_loss": 0.8262304067611694,
|
| 294 |
+
"eval_runtime": 283.5953,
|
| 295 |
+
"eval_samples_per_second": 38.001,
|
| 296 |
+
"eval_steps_per_second": 0.596,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.025,
|
| 301 |
+
"grad_norm": 1.1102286013620262,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.7842,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.05,
|
| 308 |
+
"grad_norm": 0.7607811738450544,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.7849,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.075,
|
| 315 |
+
"grad_norm": 0.7653482131592415,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.7849,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.1,
|
| 322 |
+
"grad_norm": 0.705031839749241,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.779,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.125,
|
| 329 |
+
"grad_norm": 0.5370550513294627,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.7831,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.15,
|
| 336 |
+
"grad_norm": 0.5298114725731297,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.7818,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.175,
|
| 343 |
+
"grad_norm": 0.6734194289163318,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.7772,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.2,
|
| 350 |
+
"grad_norm": 0.5897636893907213,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.7813,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.225,
|
| 357 |
+
"grad_norm": 0.604266980297563,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.7836,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.25,
|
| 364 |
+
"grad_norm": 0.6714899816376678,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.7801,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.275,
|
| 371 |
+
"grad_norm": 0.8168625249318959,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.7806,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.3,
|
| 378 |
+
"grad_norm": 0.7932681977775804,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.7776,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.325,
|
| 385 |
+
"grad_norm": 0.8862643278939233,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.7796,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.35,
|
| 392 |
+
"grad_norm": 0.7421035460384174,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.7823,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.375,
|
| 399 |
+
"grad_norm": 0.5962649233174171,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.7783,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.4,
|
| 406 |
+
"grad_norm": 0.9067010653184415,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.7755,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.425,
|
| 413 |
+
"grad_norm": 0.9272486884932043,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.7811,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.45,
|
| 420 |
+
"grad_norm": 0.634573401992294,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.7783,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.475,
|
| 427 |
+
"grad_norm": 0.7647615867543145,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.7789,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.5,
|
| 434 |
+
"grad_norm": 0.7186976301425001,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.7732,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.525,
|
| 441 |
+
"grad_norm": 0.6741640245568876,
|
| 442 |
+
"learning_rate": 5e-06,
|
| 443 |
+
"loss": 0.7803,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.55,
|
| 448 |
+
"grad_norm": 0.7673936498272227,
|
| 449 |
+
"learning_rate": 5e-06,
|
| 450 |
+
"loss": 0.7771,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.575,
|
| 455 |
+
"grad_norm": 0.6762446070662301,
|
| 456 |
+
"learning_rate": 5e-06,
|
| 457 |
+
"loss": 0.7781,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.6,
|
| 462 |
+
"grad_norm": 0.694582981661087,
|
| 463 |
+
"learning_rate": 5e-06,
|
| 464 |
+
"loss": 0.7803,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.625,
|
| 469 |
+
"grad_norm": 0.6911825951192573,
|
| 470 |
+
"learning_rate": 5e-06,
|
| 471 |
+
"loss": 0.7755,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.65,
|
| 476 |
+
"grad_norm": 0.7214035778878081,
|
| 477 |
+
"learning_rate": 5e-06,
|
| 478 |
+
"loss": 0.7784,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.675,
|
| 483 |
+
"grad_norm": 0.565532898038658,
|
| 484 |
+
"learning_rate": 5e-06,
|
| 485 |
+
"loss": 0.7798,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.7,
|
| 490 |
+
"grad_norm": 0.7144730054018653,
|
| 491 |
+
"learning_rate": 5e-06,
|
| 492 |
+
"loss": 0.7779,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.725,
|
| 497 |
+
"grad_norm": 0.68929751507137,
|
| 498 |
+
"learning_rate": 5e-06,
|
| 499 |
+
"loss": 0.7743,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.75,
|
| 504 |
+
"grad_norm": 0.6232088446530882,
|
| 505 |
+
"learning_rate": 5e-06,
|
| 506 |
+
"loss": 0.7764,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.775,
|
| 511 |
+
"grad_norm": 0.5516916387958084,
|
| 512 |
+
"learning_rate": 5e-06,
|
| 513 |
+
"loss": 0.7743,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.8,
|
| 518 |
+
"grad_norm": 0.6556060585205762,
|
| 519 |
+
"learning_rate": 5e-06,
|
| 520 |
+
"loss": 0.7795,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.825,
|
| 525 |
+
"grad_norm": 0.6395664223230667,
|
| 526 |
+
"learning_rate": 5e-06,
|
| 527 |
+
"loss": 0.7727,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.85,
|
| 532 |
+
"grad_norm": 0.7264470243374237,
|
| 533 |
+
"learning_rate": 5e-06,
|
| 534 |
+
"loss": 0.7716,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.875,
|
| 539 |
+
"grad_norm": 0.7348089904512866,
|
| 540 |
+
"learning_rate": 5e-06,
|
| 541 |
+
"loss": 0.7759,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.9,
|
| 546 |
+
"grad_norm": 0.5362704236504767,
|
| 547 |
+
"learning_rate": 5e-06,
|
| 548 |
+
"loss": 0.7725,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.925,
|
| 553 |
+
"grad_norm": 0.6132209869820257,
|
| 554 |
+
"learning_rate": 5e-06,
|
| 555 |
+
"loss": 0.7746,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.95,
|
| 560 |
+
"grad_norm": 0.5674492579544563,
|
| 561 |
+
"learning_rate": 5e-06,
|
| 562 |
+
"loss": 0.7762,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.975,
|
| 567 |
+
"grad_norm": 0.8144733224752085,
|
| 568 |
+
"learning_rate": 5e-06,
|
| 569 |
+
"loss": 0.771,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.0,
|
| 574 |
+
"grad_norm": 0.6438437017098873,
|
| 575 |
+
"learning_rate": 5e-06,
|
| 576 |
+
"loss": 0.7753,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 2.0,
|
| 581 |
+
"eval_loss": 0.8092445731163025,
|
| 582 |
+
"eval_runtime": 279.3948,
|
| 583 |
+
"eval_samples_per_second": 38.573,
|
| 584 |
+
"eval_steps_per_second": 0.605,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.025,
|
| 589 |
+
"grad_norm": 0.741980442805714,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.731,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.05,
|
| 596 |
+
"grad_norm": 0.6348046724923234,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.7254,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.075,
|
| 603 |
+
"grad_norm": 0.6998662875286445,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.722,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.1,
|
| 610 |
+
"grad_norm": 0.6662824597359785,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.7251,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.125,
|
| 617 |
+
"grad_norm": 0.5887524561855076,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.7329,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.15,
|
| 624 |
+
"grad_norm": 0.773871523528161,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.7282,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.175,
|
| 631 |
+
"grad_norm": 0.6520605524681873,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.7283,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.2,
|
| 638 |
+
"grad_norm": 0.6481774807052064,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.7333,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.225,
|
| 645 |
+
"grad_norm": 0.5980371010517554,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.7307,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.25,
|
| 652 |
+
"grad_norm": 0.6698106025945666,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.7276,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.275,
|
| 659 |
+
"grad_norm": 0.6064989242913533,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.7284,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.3,
|
| 666 |
+
"grad_norm": 0.5694949139445875,
|
| 667 |
+
"learning_rate": 5e-06,
|
| 668 |
+
"loss": 0.7287,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.325,
|
| 673 |
+
"grad_norm": 0.6518761067013964,
|
| 674 |
+
"learning_rate": 5e-06,
|
| 675 |
+
"loss": 0.7272,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.35,
|
| 680 |
+
"grad_norm": 0.6644802771746268,
|
| 681 |
+
"learning_rate": 5e-06,
|
| 682 |
+
"loss": 0.7301,
|
| 683 |
+
"step": 940
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.375,
|
| 687 |
+
"grad_norm": 0.564358869683258,
|
| 688 |
+
"learning_rate": 5e-06,
|
| 689 |
+
"loss": 0.7333,
|
| 690 |
+
"step": 950
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.4,
|
| 694 |
+
"grad_norm": 0.6052088075734858,
|
| 695 |
+
"learning_rate": 5e-06,
|
| 696 |
+
"loss": 0.7279,
|
| 697 |
+
"step": 960
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.425,
|
| 701 |
+
"grad_norm": 0.5588110088845489,
|
| 702 |
+
"learning_rate": 5e-06,
|
| 703 |
+
"loss": 0.7307,
|
| 704 |
+
"step": 970
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.45,
|
| 708 |
+
"grad_norm": 0.7921721794918313,
|
| 709 |
+
"learning_rate": 5e-06,
|
| 710 |
+
"loss": 0.727,
|
| 711 |
+
"step": 980
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.475,
|
| 715 |
+
"grad_norm": 0.633397399719071,
|
| 716 |
+
"learning_rate": 5e-06,
|
| 717 |
+
"loss": 0.7315,
|
| 718 |
+
"step": 990
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.5,
|
| 722 |
+
"grad_norm": 0.5866962766860978,
|
| 723 |
+
"learning_rate": 5e-06,
|
| 724 |
+
"loss": 0.732,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.525,
|
| 729 |
+
"grad_norm": 0.7798403321881853,
|
| 730 |
+
"learning_rate": 5e-06,
|
| 731 |
+
"loss": 0.7301,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.55,
|
| 736 |
+
"grad_norm": 0.6522313154237275,
|
| 737 |
+
"learning_rate": 5e-06,
|
| 738 |
+
"loss": 0.727,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.575,
|
| 743 |
+
"grad_norm": 0.5918752307210033,
|
| 744 |
+
"learning_rate": 5e-06,
|
| 745 |
+
"loss": 0.732,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.6,
|
| 750 |
+
"grad_norm": 0.7013642379103808,
|
| 751 |
+
"learning_rate": 5e-06,
|
| 752 |
+
"loss": 0.7304,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.625,
|
| 757 |
+
"grad_norm": 0.6019984161152461,
|
| 758 |
+
"learning_rate": 5e-06,
|
| 759 |
+
"loss": 0.7301,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.65,
|
| 764 |
+
"grad_norm": 0.6025521914638767,
|
| 765 |
+
"learning_rate": 5e-06,
|
| 766 |
+
"loss": 0.7303,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.675,
|
| 771 |
+
"grad_norm": 0.8383133859288988,
|
| 772 |
+
"learning_rate": 5e-06,
|
| 773 |
+
"loss": 0.7352,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.7,
|
| 778 |
+
"grad_norm": 0.7176072744526589,
|
| 779 |
+
"learning_rate": 5e-06,
|
| 780 |
+
"loss": 0.7325,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.725,
|
| 785 |
+
"grad_norm": 0.6330508627766337,
|
| 786 |
+
"learning_rate": 5e-06,
|
| 787 |
+
"loss": 0.7297,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.75,
|
| 792 |
+
"grad_norm": 0.708854127664677,
|
| 793 |
+
"learning_rate": 5e-06,
|
| 794 |
+
"loss": 0.7302,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.775,
|
| 799 |
+
"grad_norm": 0.7080026819402018,
|
| 800 |
+
"learning_rate": 5e-06,
|
| 801 |
+
"loss": 0.7338,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.8,
|
| 806 |
+
"grad_norm": 0.6486046388056755,
|
| 807 |
+
"learning_rate": 5e-06,
|
| 808 |
+
"loss": 0.7333,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.825,
|
| 813 |
+
"grad_norm": 0.6098247605462774,
|
| 814 |
+
"learning_rate": 5e-06,
|
| 815 |
+
"loss": 0.7368,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.85,
|
| 820 |
+
"grad_norm": 0.707595465934189,
|
| 821 |
+
"learning_rate": 5e-06,
|
| 822 |
+
"loss": 0.7315,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.875,
|
| 827 |
+
"grad_norm": 0.6598068743344158,
|
| 828 |
+
"learning_rate": 5e-06,
|
| 829 |
+
"loss": 0.7328,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.9,
|
| 834 |
+
"grad_norm": 0.6256793408527994,
|
| 835 |
+
"learning_rate": 5e-06,
|
| 836 |
+
"loss": 0.7335,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.925,
|
| 841 |
+
"grad_norm": 0.6477061017951039,
|
| 842 |
+
"learning_rate": 5e-06,
|
| 843 |
+
"loss": 0.7296,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.95,
|
| 848 |
+
"grad_norm": 0.6150717312512685,
|
| 849 |
+
"learning_rate": 5e-06,
|
| 850 |
+
"loss": 0.7336,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.975,
|
| 855 |
+
"grad_norm": 0.5375035486490041,
|
| 856 |
+
"learning_rate": 5e-06,
|
| 857 |
+
"loss": 0.7316,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 3.0,
|
| 862 |
+
"grad_norm": 0.6000906442308059,
|
| 863 |
+
"learning_rate": 5e-06,
|
| 864 |
+
"loss": 0.7331,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.0,
|
| 869 |
+
"eval_loss": 0.8070117235183716,
|
| 870 |
+
"eval_runtime": 271.2171,
|
| 871 |
+
"eval_samples_per_second": 39.736,
|
| 872 |
+
"eval_steps_per_second": 0.623,
|
| 873 |
+
"step": 1200
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 3.0,
|
| 877 |
+
"step": 1200,
|
| 878 |
+
"total_flos": 2009835314872320.0,
|
| 879 |
+
"train_loss": 0.7931587568918864,
|
| 880 |
+
"train_runtime": 40308.7538,
|
| 881 |
+
"train_samples_per_second": 15.238,
|
| 882 |
+
"train_steps_per_second": 0.03
|
| 883 |
+
}
|
| 884 |
+
],
|
| 885 |
+
"logging_steps": 10,
|
| 886 |
+
"max_steps": 1200,
|
| 887 |
+
"num_input_tokens_seen": 0,
|
| 888 |
+
"num_train_epochs": 3,
|
| 889 |
+
"save_steps": 500,
|
| 890 |
+
"stateful_callbacks": {
|
| 891 |
+
"TrainerControl": {
|
| 892 |
+
"args": {
|
| 893 |
+
"should_epoch_stop": false,
|
| 894 |
+
"should_evaluate": false,
|
| 895 |
+
"should_log": false,
|
| 896 |
+
"should_save": true,
|
| 897 |
+
"should_training_stop": true
|
| 898 |
+
},
|
| 899 |
+
"attributes": {}
|
| 900 |
+
}
|
| 901 |
+
},
|
| 902 |
+
"total_flos": 2009835314872320.0,
|
| 903 |
+
"train_batch_size": 8,
|
| 904 |
+
"trial_name": null,
|
| 905 |
+
"trial_params": null
|
| 906 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|