Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +1134 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 156,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01282051282051282,
|
| 13 |
+
"grad_norm": 5.654436927502313,
|
| 14 |
+
"learning_rate": 1.25e-07,
|
| 15 |
+
"loss": 0.1973,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02564102564102564,
|
| 20 |
+
"grad_norm": 6.626728097482495,
|
| 21 |
+
"learning_rate": 2.5e-07,
|
| 22 |
+
"loss": 0.2125,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.038461538461538464,
|
| 27 |
+
"grad_norm": 5.191773774712552,
|
| 28 |
+
"learning_rate": 3.75e-07,
|
| 29 |
+
"loss": 0.2059,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05128205128205128,
|
| 34 |
+
"grad_norm": 4.852827406613362,
|
| 35 |
+
"learning_rate": 5e-07,
|
| 36 |
+
"loss": 0.2015,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0641025641025641,
|
| 41 |
+
"grad_norm": 5.69227350455002,
|
| 42 |
+
"learning_rate": 6.249999999999999e-07,
|
| 43 |
+
"loss": 0.224,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07692307692307693,
|
| 48 |
+
"grad_norm": 4.784103798672743,
|
| 49 |
+
"learning_rate": 7.5e-07,
|
| 50 |
+
"loss": 0.2303,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08974358974358974,
|
| 55 |
+
"grad_norm": 5.22887777491138,
|
| 56 |
+
"learning_rate": 8.75e-07,
|
| 57 |
+
"loss": 0.2085,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10256410256410256,
|
| 62 |
+
"grad_norm": 8.072567924174027,
|
| 63 |
+
"learning_rate": 1e-06,
|
| 64 |
+
"loss": 0.2456,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11538461538461539,
|
| 69 |
+
"grad_norm": 4.413690659858428,
|
| 70 |
+
"learning_rate": 1.125e-06,
|
| 71 |
+
"loss": 0.2244,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.1282051282051282,
|
| 76 |
+
"grad_norm": 6.608388352589465,
|
| 77 |
+
"learning_rate": 1.2499999999999999e-06,
|
| 78 |
+
"loss": 0.2112,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14102564102564102,
|
| 83 |
+
"grad_norm": 5.926240686861022,
|
| 84 |
+
"learning_rate": 1.375e-06,
|
| 85 |
+
"loss": 0.2004,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15384615384615385,
|
| 90 |
+
"grad_norm": 5.08796650227605,
|
| 91 |
+
"learning_rate": 1.5e-06,
|
| 92 |
+
"loss": 0.1926,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16666666666666666,
|
| 97 |
+
"grad_norm": 5.2298278793003075,
|
| 98 |
+
"learning_rate": 1.625e-06,
|
| 99 |
+
"loss": 0.2305,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1794871794871795,
|
| 104 |
+
"grad_norm": 4.166605877793112,
|
| 105 |
+
"learning_rate": 1.75e-06,
|
| 106 |
+
"loss": 0.206,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.19230769230769232,
|
| 111 |
+
"grad_norm": 4.292650959846603,
|
| 112 |
+
"learning_rate": 1.8749999999999998e-06,
|
| 113 |
+
"loss": 0.2078,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.20512820512820512,
|
| 118 |
+
"grad_norm": 5.696519990963819,
|
| 119 |
+
"learning_rate": 2e-06,
|
| 120 |
+
"loss": 0.2038,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.21794871794871795,
|
| 125 |
+
"grad_norm": 6.300006556535238,
|
| 126 |
+
"learning_rate": 1.9997482349425066e-06,
|
| 127 |
+
"loss": 0.215,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23076923076923078,
|
| 132 |
+
"grad_norm": 7.520729049313368,
|
| 133 |
+
"learning_rate": 1.9989930665413145e-06,
|
| 134 |
+
"loss": 0.2057,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.24358974358974358,
|
| 139 |
+
"grad_norm": 5.558795363812876,
|
| 140 |
+
"learning_rate": 1.997734875046456e-06,
|
| 141 |
+
"loss": 0.2213,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2564102564102564,
|
| 146 |
+
"grad_norm": 6.697951711680472,
|
| 147 |
+
"learning_rate": 1.995974293995239e-06,
|
| 148 |
+
"loss": 0.2334,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2692307692307692,
|
| 153 |
+
"grad_norm": 6.202168109977011,
|
| 154 |
+
"learning_rate": 1.9937122098932426e-06,
|
| 155 |
+
"loss": 0.2044,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.28205128205128205,
|
| 160 |
+
"grad_norm": 5.709238973742468,
|
| 161 |
+
"learning_rate": 1.9909497617679347e-06,
|
| 162 |
+
"loss": 0.2113,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2948717948717949,
|
| 167 |
+
"grad_norm": 7.473600358104061,
|
| 168 |
+
"learning_rate": 1.9876883405951377e-06,
|
| 169 |
+
"loss": 0.2003,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3076923076923077,
|
| 174 |
+
"grad_norm": 6.149828829786014,
|
| 175 |
+
"learning_rate": 1.9839295885986295e-06,
|
| 176 |
+
"loss": 0.2002,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.32051282051282054,
|
| 181 |
+
"grad_norm": 6.49183662646828,
|
| 182 |
+
"learning_rate": 1.9796753984232355e-06,
|
| 183 |
+
"loss": 0.2128,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3333333333333333,
|
| 188 |
+
"grad_norm": 8.0523038385668,
|
| 189 |
+
"learning_rate": 1.9749279121818236e-06,
|
| 190 |
+
"loss": 0.2069,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.34615384615384615,
|
| 195 |
+
"grad_norm": 6.53770446262557,
|
| 196 |
+
"learning_rate": 1.9696895203766866e-06,
|
| 197 |
+
"loss": 0.197,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.358974358974359,
|
| 202 |
+
"grad_norm": 5.9687895529895805,
|
| 203 |
+
"learning_rate": 1.9639628606958534e-06,
|
| 204 |
+
"loss": 0.1874,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3717948717948718,
|
| 209 |
+
"grad_norm": 8.877460162864702,
|
| 210 |
+
"learning_rate": 1.9577508166849303e-06,
|
| 211 |
+
"loss": 0.2212,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.38461538461538464,
|
| 216 |
+
"grad_norm": 7.503998295423821,
|
| 217 |
+
"learning_rate": 1.9510565162951534e-06,
|
| 218 |
+
"loss": 0.2187,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3974358974358974,
|
| 223 |
+
"grad_norm": 6.195958215948267,
|
| 224 |
+
"learning_rate": 1.9438833303083674e-06,
|
| 225 |
+
"loss": 0.2187,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.41025641025641024,
|
| 230 |
+
"grad_norm": 7.174461354155621,
|
| 231 |
+
"learning_rate": 1.936234870639737e-06,
|
| 232 |
+
"loss": 0.1864,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4230769230769231,
|
| 237 |
+
"grad_norm": 9.029309774222588,
|
| 238 |
+
"learning_rate": 1.928114988519039e-06,
|
| 239 |
+
"loss": 0.2089,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4358974358974359,
|
| 244 |
+
"grad_norm": 5.056660519084398,
|
| 245 |
+
"learning_rate": 1.9195277725514506e-06,
|
| 246 |
+
"loss": 0.2297,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.44871794871794873,
|
| 251 |
+
"grad_norm": 7.528111764781314,
|
| 252 |
+
"learning_rate": 1.9104775466588157e-06,
|
| 253 |
+
"loss": 0.1891,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.46153846153846156,
|
| 258 |
+
"grad_norm": 5.927725291827813,
|
| 259 |
+
"learning_rate": 1.9009688679024189e-06,
|
| 260 |
+
"loss": 0.2107,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.47435897435897434,
|
| 265 |
+
"grad_norm": 5.918052578055244,
|
| 266 |
+
"learning_rate": 1.8910065241883678e-06,
|
| 267 |
+
"loss": 0.2151,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.48717948717948717,
|
| 272 |
+
"grad_norm": 8.021624884542486,
|
| 273 |
+
"learning_rate": 1.8805955318567379e-06,
|
| 274 |
+
"loss": 0.1947,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5,
|
| 279 |
+
"grad_norm": 10.034128786337845,
|
| 280 |
+
"learning_rate": 1.8697411331556953e-06,
|
| 281 |
+
"loss": 0.1958,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5128205128205128,
|
| 286 |
+
"grad_norm": 5.830942012949506,
|
| 287 |
+
"learning_rate": 1.858448793601866e-06,
|
| 288 |
+
"loss": 0.2141,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5256410256410257,
|
| 293 |
+
"grad_norm": 6.052490132533048,
|
| 294 |
+
"learning_rate": 1.8467241992282841e-06,
|
| 295 |
+
"loss": 0.2166,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5384615384615384,
|
| 300 |
+
"grad_norm": 8.489552316231327,
|
| 301 |
+
"learning_rate": 1.8345732537213026e-06,
|
| 302 |
+
"loss": 0.2352,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5512820512820513,
|
| 307 |
+
"grad_norm": 6.764408190860317,
|
| 308 |
+
"learning_rate": 1.82200207544791e-06,
|
| 309 |
+
"loss": 0.1942,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5641025641025641,
|
| 314 |
+
"grad_norm": 6.549594976256598,
|
| 315 |
+
"learning_rate": 1.8090169943749474e-06,
|
| 316 |
+
"loss": 0.1814,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5769230769230769,
|
| 321 |
+
"grad_norm": 4.88312442021333,
|
| 322 |
+
"learning_rate": 1.795624548881781e-06,
|
| 323 |
+
"loss": 0.18,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5897435897435898,
|
| 328 |
+
"grad_norm": 8.563572665281708,
|
| 329 |
+
"learning_rate": 1.7818314824680298e-06,
|
| 330 |
+
"loss": 0.2104,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6025641025641025,
|
| 335 |
+
"grad_norm": 6.252677703659704,
|
| 336 |
+
"learning_rate": 1.767644740358011e-06,
|
| 337 |
+
"loss": 0.2064,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6153846153846154,
|
| 342 |
+
"grad_norm": 6.57449376541026,
|
| 343 |
+
"learning_rate": 1.753071466003611e-06,
|
| 344 |
+
"loss": 0.2166,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6282051282051282,
|
| 349 |
+
"grad_norm": 9.94763215280549,
|
| 350 |
+
"learning_rate": 1.7381189974873407e-06,
|
| 351 |
+
"loss": 0.2115,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6410256410256411,
|
| 356 |
+
"grad_norm": 6.086169279661426,
|
| 357 |
+
"learning_rate": 1.7227948638273915e-06,
|
| 358 |
+
"loss": 0.1951,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6538461538461539,
|
| 363 |
+
"grad_norm": 6.852323468955059,
|
| 364 |
+
"learning_rate": 1.7071067811865474e-06,
|
| 365 |
+
"loss": 0.2061,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6666666666666666,
|
| 370 |
+
"grad_norm": 6.659536057802301,
|
| 371 |
+
"learning_rate": 1.6910626489868648e-06,
|
| 372 |
+
"loss": 0.2008,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6794871794871795,
|
| 377 |
+
"grad_norm": 6.278100242208402,
|
| 378 |
+
"learning_rate": 1.6746705459320744e-06,
|
| 379 |
+
"loss": 0.2204,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6923076923076923,
|
| 384 |
+
"grad_norm": 6.003617638218591,
|
| 385 |
+
"learning_rate": 1.6579387259397126e-06,
|
| 386 |
+
"loss": 0.1948,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7051282051282052,
|
| 391 |
+
"grad_norm": 5.691471477377111,
|
| 392 |
+
"learning_rate": 1.640875613985024e-06,
|
| 393 |
+
"loss": 0.2001,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.717948717948718,
|
| 398 |
+
"grad_norm": 7.26868258409944,
|
| 399 |
+
"learning_rate": 1.6234898018587336e-06,
|
| 400 |
+
"loss": 0.2325,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7307692307692307,
|
| 405 |
+
"grad_norm": 11.03004784213591,
|
| 406 |
+
"learning_rate": 1.6057900438408199e-06,
|
| 407 |
+
"loss": 0.2098,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7435897435897436,
|
| 412 |
+
"grad_norm": 11.818489855475011,
|
| 413 |
+
"learning_rate": 1.587785252292473e-06,
|
| 414 |
+
"loss": 0.2483,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7564102564102564,
|
| 419 |
+
"grad_norm": 6.651053168471789,
|
| 420 |
+
"learning_rate": 1.569484493168452e-06,
|
| 421 |
+
"loss": 0.2206,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7692307692307693,
|
| 426 |
+
"grad_norm": 6.265843765185568,
|
| 427 |
+
"learning_rate": 1.5508969814521024e-06,
|
| 428 |
+
"loss": 0.2476,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.782051282051282,
|
| 433 |
+
"grad_norm": 6.377989141523518,
|
| 434 |
+
"learning_rate": 1.5320320765153365e-06,
|
| 435 |
+
"loss": 0.2256,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.7948717948717948,
|
| 440 |
+
"grad_norm": 6.77172097560254,
|
| 441 |
+
"learning_rate": 1.5128992774059062e-06,
|
| 442 |
+
"loss": 0.2192,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8076923076923077,
|
| 447 |
+
"grad_norm": 8.034536751348226,
|
| 448 |
+
"learning_rate": 1.4935082180643467e-06,
|
| 449 |
+
"loss": 0.2045,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8205128205128205,
|
| 454 |
+
"grad_norm": 7.238602784260153,
|
| 455 |
+
"learning_rate": 1.4738686624729987e-06,
|
| 456 |
+
"loss": 0.2097,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8333333333333334,
|
| 461 |
+
"grad_norm": 6.255397368652637,
|
| 462 |
+
"learning_rate": 1.4539904997395467e-06,
|
| 463 |
+
"loss": 0.1961,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8461538461538461,
|
| 468 |
+
"grad_norm": 5.344086626520604,
|
| 469 |
+
"learning_rate": 1.433883739117558e-06,
|
| 470 |
+
"loss": 0.1929,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8589743589743589,
|
| 475 |
+
"grad_norm": 8.83134621904589,
|
| 476 |
+
"learning_rate": 1.4135585049665206e-06,
|
| 477 |
+
"loss": 0.2045,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8717948717948718,
|
| 482 |
+
"grad_norm": 8.50453690901984,
|
| 483 |
+
"learning_rate": 1.3930250316539235e-06,
|
| 484 |
+
"loss": 0.2069,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8846153846153846,
|
| 489 |
+
"grad_norm": 7.6397867354678874,
|
| 490 |
+
"learning_rate": 1.3722936584019451e-06,
|
| 491 |
+
"loss": 0.1913,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.8974358974358975,
|
| 496 |
+
"grad_norm": 7.745065793286678,
|
| 497 |
+
"learning_rate": 1.3513748240813427e-06,
|
| 498 |
+
"loss": 0.1833,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9102564102564102,
|
| 503 |
+
"grad_norm": 6.807770861970589,
|
| 504 |
+
"learning_rate": 1.3302790619551672e-06,
|
| 505 |
+
"loss": 0.202,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9230769230769231,
|
| 510 |
+
"grad_norm": 5.865099001869984,
|
| 511 |
+
"learning_rate": 1.3090169943749473e-06,
|
| 512 |
+
"loss": 0.2106,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9358974358974359,
|
| 517 |
+
"grad_norm": 5.443624271225363,
|
| 518 |
+
"learning_rate": 1.2875993274320173e-06,
|
| 519 |
+
"loss": 0.1905,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9487179487179487,
|
| 524 |
+
"grad_norm": 8.268666970710147,
|
| 525 |
+
"learning_rate": 1.266036845566675e-06,
|
| 526 |
+
"loss": 0.1967,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9615384615384616,
|
| 531 |
+
"grad_norm": 9.115366281353827,
|
| 532 |
+
"learning_rate": 1.244340406137894e-06,
|
| 533 |
+
"loss": 0.1991,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9743589743589743,
|
| 538 |
+
"grad_norm": 7.710114282346724,
|
| 539 |
+
"learning_rate": 1.2225209339563143e-06,
|
| 540 |
+
"loss": 0.1993,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9871794871794872,
|
| 545 |
+
"grad_norm": 5.780906649992342,
|
| 546 |
+
"learning_rate": 1.2005894157832728e-06,
|
| 547 |
+
"loss": 0.2047,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.0,
|
| 552 |
+
"grad_norm": 7.36349391438361,
|
| 553 |
+
"learning_rate": 1.1785568947986366e-06,
|
| 554 |
+
"loss": 0.1974,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.0128205128205128,
|
| 559 |
+
"grad_norm": 4.653198666191018,
|
| 560 |
+
"learning_rate": 1.156434465040231e-06,
|
| 561 |
+
"loss": 0.2002,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.0256410256410255,
|
| 566 |
+
"grad_norm": 3.7980826556751635,
|
| 567 |
+
"learning_rate": 1.1342332658176555e-06,
|
| 568 |
+
"loss": 0.2004,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.0384615384615385,
|
| 573 |
+
"grad_norm": 3.5134848623818344,
|
| 574 |
+
"learning_rate": 1.1119644761033077e-06,
|
| 575 |
+
"loss": 0.1806,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.0512820512820513,
|
| 580 |
+
"grad_norm": 3.83273521354433,
|
| 581 |
+
"learning_rate": 1.0896393089034335e-06,
|
| 582 |
+
"loss": 0.1843,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.064102564102564,
|
| 587 |
+
"grad_norm": 3.611360912304485,
|
| 588 |
+
"learning_rate": 1.0672690056120398e-06,
|
| 589 |
+
"loss": 0.1877,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0769230769230769,
|
| 594 |
+
"grad_norm": 3.448510808199711,
|
| 595 |
+
"learning_rate": 1.044864830350515e-06,
|
| 596 |
+
"loss": 0.1883,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.0897435897435896,
|
| 601 |
+
"grad_norm": 4.569903767106023,
|
| 602 |
+
"learning_rate": 1.022438064295805e-06,
|
| 603 |
+
"loss": 0.1679,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.1025641025641026,
|
| 608 |
+
"grad_norm": 3.8192214552271277,
|
| 609 |
+
"learning_rate": 1e-06,
|
| 610 |
+
"loss": 0.1874,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.1153846153846154,
|
| 615 |
+
"grad_norm": 3.8192368467830313,
|
| 616 |
+
"learning_rate": 9.77561935704195e-07,
|
| 617 |
+
"loss": 0.1754,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.1282051282051282,
|
| 622 |
+
"grad_norm": 3.610547145371744,
|
| 623 |
+
"learning_rate": 9.551351696494853e-07,
|
| 624 |
+
"loss": 0.1822,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.141025641025641,
|
| 629 |
+
"grad_norm": 3.589131977726136,
|
| 630 |
+
"learning_rate": 9.327309943879603e-07,
|
| 631 |
+
"loss": 0.1651,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.1538461538461537,
|
| 636 |
+
"grad_norm": 4.194399981796768,
|
| 637 |
+
"learning_rate": 9.103606910965665e-07,
|
| 638 |
+
"loss": 0.1756,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1666666666666667,
|
| 643 |
+
"grad_norm": 3.408278500756967,
|
| 644 |
+
"learning_rate": 8.880355238966921e-07,
|
| 645 |
+
"loss": 0.1744,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1794871794871795,
|
| 650 |
+
"grad_norm": 3.6044589661783224,
|
| 651 |
+
"learning_rate": 8.657667341823448e-07,
|
| 652 |
+
"loss": 0.1822,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.1923076923076923,
|
| 657 |
+
"grad_norm": 4.046000102159344,
|
| 658 |
+
"learning_rate": 8.435655349597689e-07,
|
| 659 |
+
"loss": 0.1717,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.205128205128205,
|
| 664 |
+
"grad_norm": 3.9296326372487544,
|
| 665 |
+
"learning_rate": 8.214431052013634e-07,
|
| 666 |
+
"loss": 0.1883,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.217948717948718,
|
| 671 |
+
"grad_norm": 3.700661482476124,
|
| 672 |
+
"learning_rate": 7.994105842167272e-07,
|
| 673 |
+
"loss": 0.1764,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2307692307692308,
|
| 678 |
+
"grad_norm": 5.372030890569249,
|
| 679 |
+
"learning_rate": 7.774790660436857e-07,
|
| 680 |
+
"loss": 0.1965,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.2435897435897436,
|
| 685 |
+
"grad_norm": 4.242334560823488,
|
| 686 |
+
"learning_rate": 7.556595938621058e-07,
|
| 687 |
+
"loss": 0.1827,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.2564102564102564,
|
| 692 |
+
"grad_norm": 3.799513498528545,
|
| 693 |
+
"learning_rate": 7.33963154433325e-07,
|
| 694 |
+
"loss": 0.1743,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.2692307692307692,
|
| 699 |
+
"grad_norm": 3.934607344620737,
|
| 700 |
+
"learning_rate": 7.124006725679828e-07,
|
| 701 |
+
"loss": 0.1823,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.282051282051282,
|
| 706 |
+
"grad_norm": 4.925229210556202,
|
| 707 |
+
"learning_rate": 6.909830056250526e-07,
|
| 708 |
+
"loss": 0.1777,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.294871794871795,
|
| 713 |
+
"grad_norm": 3.5402723653658845,
|
| 714 |
+
"learning_rate": 6.697209380448332e-07,
|
| 715 |
+
"loss": 0.1611,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.3076923076923077,
|
| 720 |
+
"grad_norm": 3.67954011660819,
|
| 721 |
+
"learning_rate": 6.486251759186572e-07,
|
| 722 |
+
"loss": 0.166,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.3205128205128205,
|
| 727 |
+
"grad_norm": 3.539381923301239,
|
| 728 |
+
"learning_rate": 6.277063415980548e-07,
|
| 729 |
+
"loss": 0.1928,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.3333333333333333,
|
| 734 |
+
"grad_norm": 3.5828829512416642,
|
| 735 |
+
"learning_rate": 6.069749683460764e-07,
|
| 736 |
+
"loss": 0.1712,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.3461538461538463,
|
| 741 |
+
"grad_norm": 3.495907721286862,
|
| 742 |
+
"learning_rate": 5.864414950334795e-07,
|
| 743 |
+
"loss": 0.1597,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.358974358974359,
|
| 748 |
+
"grad_norm": 3.448718183120504,
|
| 749 |
+
"learning_rate": 5.661162608824419e-07,
|
| 750 |
+
"loss": 0.1885,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.3717948717948718,
|
| 755 |
+
"grad_norm": 3.6865447674417484,
|
| 756 |
+
"learning_rate": 5.460095002604532e-07,
|
| 757 |
+
"loss": 0.1771,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.3846153846153846,
|
| 762 |
+
"grad_norm": 4.132287434133233,
|
| 763 |
+
"learning_rate": 5.261313375270013e-07,
|
| 764 |
+
"loss": 0.1833,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.3974358974358974,
|
| 769 |
+
"grad_norm": 3.6813685770538824,
|
| 770 |
+
"learning_rate": 5.064917819356531e-07,
|
| 771 |
+
"loss": 0.1596,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.4102564102564101,
|
| 776 |
+
"grad_norm": 4.011642360472788,
|
| 777 |
+
"learning_rate": 4.871007225940939e-07,
|
| 778 |
+
"loss": 0.1734,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.4230769230769231,
|
| 783 |
+
"grad_norm": 4.9153871108994345,
|
| 784 |
+
"learning_rate": 4.6796792348466353e-07,
|
| 785 |
+
"loss": 0.1938,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.435897435897436,
|
| 790 |
+
"grad_norm": 3.6473161562273577,
|
| 791 |
+
"learning_rate": 4.4910301854789755e-07,
|
| 792 |
+
"loss": 0.1811,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.4487179487179487,
|
| 797 |
+
"grad_norm": 5.4491593387650115,
|
| 798 |
+
"learning_rate": 4.3051550683154804e-07,
|
| 799 |
+
"loss": 0.1783,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.4615384615384617,
|
| 804 |
+
"grad_norm": 3.6123921470095146,
|
| 805 |
+
"learning_rate": 4.1221474770752696e-07,
|
| 806 |
+
"loss": 0.1682,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.4743589743589745,
|
| 811 |
+
"grad_norm": 3.479526179911105,
|
| 812 |
+
"learning_rate": 3.942099561591802e-07,
|
| 813 |
+
"loss": 0.1625,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.4871794871794872,
|
| 818 |
+
"grad_norm": 3.527085972143578,
|
| 819 |
+
"learning_rate": 3.765101981412665e-07,
|
| 820 |
+
"loss": 0.1595,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.5,
|
| 825 |
+
"grad_norm": 3.751067158642893,
|
| 826 |
+
"learning_rate": 3.5912438601497584e-07,
|
| 827 |
+
"loss": 0.1749,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.5128205128205128,
|
| 832 |
+
"grad_norm": 3.2729349470702305,
|
| 833 |
+
"learning_rate": 3.420612740602874e-07,
|
| 834 |
+
"loss": 0.1743,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.5256410256410255,
|
| 839 |
+
"grad_norm": 3.8112995810635075,
|
| 840 |
+
"learning_rate": 3.253294540679257e-07,
|
| 841 |
+
"loss": 0.1687,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.5384615384615383,
|
| 846 |
+
"grad_norm": 3.8584077848619542,
|
| 847 |
+
"learning_rate": 3.0893735101313535e-07,
|
| 848 |
+
"loss": 0.1797,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.5512820512820513,
|
| 853 |
+
"grad_norm": 2.9480553215055396,
|
| 854 |
+
"learning_rate": 2.9289321881345254e-07,
|
| 855 |
+
"loss": 0.1652,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.564102564102564,
|
| 860 |
+
"grad_norm": 3.665600196030296,
|
| 861 |
+
"learning_rate": 2.7720513617260855e-07,
|
| 862 |
+
"loss": 0.1611,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.5769230769230769,
|
| 867 |
+
"grad_norm": 4.1018359378878,
|
| 868 |
+
"learning_rate": 2.6188100251265943e-07,
|
| 869 |
+
"loss": 0.1661,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.5897435897435899,
|
| 874 |
+
"grad_norm": 4.065847262008942,
|
| 875 |
+
"learning_rate": 2.4692853399638913e-07,
|
| 876 |
+
"loss": 0.1618,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.6025641025641026,
|
| 881 |
+
"grad_norm": 4.687554993645887,
|
| 882 |
+
"learning_rate": 2.3235525964198888e-07,
|
| 883 |
+
"loss": 0.1713,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.6153846153846154,
|
| 888 |
+
"grad_norm": 4.844279539372532,
|
| 889 |
+
"learning_rate": 2.181685175319702e-07,
|
| 890 |
+
"loss": 0.1729,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.6282051282051282,
|
| 895 |
+
"grad_norm": 4.804893114610019,
|
| 896 |
+
"learning_rate": 2.043754511182191e-07,
|
| 897 |
+
"loss": 0.1735,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.641025641025641,
|
| 902 |
+
"grad_norm": 3.172109679268063,
|
| 903 |
+
"learning_rate": 1.9098300562505264e-07,
|
| 904 |
+
"loss": 0.1631,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.6538461538461537,
|
| 909 |
+
"grad_norm": 3.0808645279202587,
|
| 910 |
+
"learning_rate": 1.7799792455209016e-07,
|
| 911 |
+
"loss": 0.1736,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.6666666666666665,
|
| 916 |
+
"grad_norm": 6.873539949503569,
|
| 917 |
+
"learning_rate": 1.6542674627869734e-07,
|
| 918 |
+
"loss": 0.173,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.6794871794871795,
|
| 923 |
+
"grad_norm": 6.772028203722931,
|
| 924 |
+
"learning_rate": 1.5327580077171588e-07,
|
| 925 |
+
"loss": 0.175,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.6923076923076923,
|
| 930 |
+
"grad_norm": 3.5318656132751522,
|
| 931 |
+
"learning_rate": 1.415512063981339e-07,
|
| 932 |
+
"loss": 0.1608,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.7051282051282053,
|
| 937 |
+
"grad_norm": 3.542583252285712,
|
| 938 |
+
"learning_rate": 1.3025886684430465e-07,
|
| 939 |
+
"loss": 0.1676,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.717948717948718,
|
| 944 |
+
"grad_norm": 3.261417812424828,
|
| 945 |
+
"learning_rate": 1.19404468143262e-07,
|
| 946 |
+
"loss": 0.1651,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.7307692307692308,
|
| 951 |
+
"grad_norm": 3.4294106221389455,
|
| 952 |
+
"learning_rate": 1.089934758116322e-07,
|
| 953 |
+
"loss": 0.1646,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.7435897435897436,
|
| 958 |
+
"grad_norm": 3.332436518189792,
|
| 959 |
+
"learning_rate": 9.903113209758096e-08,
|
| 960 |
+
"loss": 0.1848,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.7564102564102564,
|
| 965 |
+
"grad_norm": 3.259323557596099,
|
| 966 |
+
"learning_rate": 8.952245334118413e-08,
|
| 967 |
+
"loss": 0.1622,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.7692307692307692,
|
| 972 |
+
"grad_norm": 2.683942222088084,
|
| 973 |
+
"learning_rate": 8.047222744854942e-08,
|
| 974 |
+
"loss": 0.1755,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.782051282051282,
|
| 979 |
+
"grad_norm": 3.529116813078909,
|
| 980 |
+
"learning_rate": 7.188501148096116e-08,
|
| 981 |
+
"loss": 0.1769,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.7948717948717947,
|
| 986 |
+
"grad_norm": 3.457639855339149,
|
| 987 |
+
"learning_rate": 6.376512936026279e-08,
|
| 988 |
+
"loss": 0.1791,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.8076923076923077,
|
| 993 |
+
"grad_norm": 6.966811792098972,
|
| 994 |
+
"learning_rate": 5.611666969163242e-08,
|
| 995 |
+
"loss": 0.1966,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.8205128205128205,
|
| 1000 |
+
"grad_norm": 3.408397971457261,
|
| 1001 |
+
"learning_rate": 4.8943483704846465e-08,
|
| 1002 |
+
"loss": 0.17,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.8333333333333335,
|
| 1007 |
+
"grad_norm": 3.741568289882954,
|
| 1008 |
+
"learning_rate": 4.224918331506955e-08,
|
| 1009 |
+
"loss": 0.1762,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.8461538461538463,
|
| 1014 |
+
"grad_norm": 3.6826280424952826,
|
| 1015 |
+
"learning_rate": 3.6037139304146756e-08,
|
| 1016 |
+
"loss": 0.1889,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.858974358974359,
|
| 1021 |
+
"grad_norm": 3.937760918906877,
|
| 1022 |
+
"learning_rate": 3.0310479623313125e-08,
|
| 1023 |
+
"loss": 0.1759,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.8717948717948718,
|
| 1028 |
+
"grad_norm": 5.314409301382863,
|
| 1029 |
+
"learning_rate": 2.507208781817638e-08,
|
| 1030 |
+
"loss": 0.1712,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.8846153846153846,
|
| 1035 |
+
"grad_norm": 3.137976564129638,
|
| 1036 |
+
"learning_rate": 2.032460157676452e-08,
|
| 1037 |
+
"loss": 0.1762,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.8974358974358974,
|
| 1042 |
+
"grad_norm": 3.8473953332791253,
|
| 1043 |
+
"learning_rate": 1.607041140137033e-08,
|
| 1044 |
+
"loss": 0.1808,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.9102564102564101,
|
| 1049 |
+
"grad_norm": 3.6582469574336174,
|
| 1050 |
+
"learning_rate": 1.231165940486234e-08,
|
| 1051 |
+
"loss": 0.169,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.9230769230769231,
|
| 1056 |
+
"grad_norm": 3.7902053426671833,
|
| 1057 |
+
"learning_rate": 9.050238232065299e-09,
|
| 1058 |
+
"loss": 0.1503,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.935897435897436,
|
| 1063 |
+
"grad_norm": 3.168334679520932,
|
| 1064 |
+
"learning_rate": 6.2877901067573955e-09,
|
| 1065 |
+
"loss": 0.1711,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.9487179487179487,
|
| 1070 |
+
"grad_norm": 4.039399422254096,
|
| 1071 |
+
"learning_rate": 4.025706004760931e-09,
|
| 1072 |
+
"loss": 0.1925,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.9615384615384617,
|
| 1077 |
+
"grad_norm": 3.784288177455713,
|
| 1078 |
+
"learning_rate": 2.2651249535439177e-09,
|
| 1079 |
+
"loss": 0.1727,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.9743589743589745,
|
| 1084 |
+
"grad_norm": 2.959451188655635,
|
| 1085 |
+
"learning_rate": 1.0069334586854105e-09,
|
| 1086 |
+
"loss": 0.1559,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.9871794871794872,
|
| 1091 |
+
"grad_norm": 3.690014088756706,
|
| 1092 |
+
"learning_rate": 2.517650574934693e-10,
|
| 1093 |
+
"loss": 0.1742,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.0,
|
| 1098 |
+
"grad_norm": 3.80353675897014,
|
| 1099 |
+
"learning_rate": 0.0,
|
| 1100 |
+
"loss": 0.173,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.0,
|
| 1105 |
+
"step": 156,
|
| 1106 |
+
"total_flos": 63560640233472.0,
|
| 1107 |
+
"train_loss": 0.19183813665921873,
|
| 1108 |
+
"train_runtime": 2071.3478,
|
| 1109 |
+
"train_samples_per_second": 19.268,
|
| 1110 |
+
"train_steps_per_second": 0.075
|
| 1111 |
+
}
|
| 1112 |
+
],
|
| 1113 |
+
"logging_steps": 1,
|
| 1114 |
+
"max_steps": 156,
|
| 1115 |
+
"num_input_tokens_seen": 0,
|
| 1116 |
+
"num_train_epochs": 2,
|
| 1117 |
+
"save_steps": 100,
|
| 1118 |
+
"stateful_callbacks": {
|
| 1119 |
+
"TrainerControl": {
|
| 1120 |
+
"args": {
|
| 1121 |
+
"should_epoch_stop": false,
|
| 1122 |
+
"should_evaluate": false,
|
| 1123 |
+
"should_log": false,
|
| 1124 |
+
"should_save": true,
|
| 1125 |
+
"should_training_stop": true
|
| 1126 |
+
},
|
| 1127 |
+
"attributes": {}
|
| 1128 |
+
}
|
| 1129 |
+
},
|
| 1130 |
+
"total_flos": 63560640233472.0,
|
| 1131 |
+
"train_batch_size": 2,
|
| 1132 |
+
"trial_name": null,
|
| 1133 |
+
"trial_params": null
|
| 1134 |
+
}
|