Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +1120 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1120 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9887005649717513,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 154,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0129136400322841,
|
| 13 |
+
"grad_norm": 26.062711090700205,
|
| 14 |
+
"learning_rate": 1.25e-07,
|
| 15 |
+
"loss": 0.3676,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0258272800645682,
|
| 20 |
+
"grad_norm": 25.492701005475226,
|
| 21 |
+
"learning_rate": 2.5e-07,
|
| 22 |
+
"loss": 0.3636,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0387409200968523,
|
| 27 |
+
"grad_norm": 28.061770353754802,
|
| 28 |
+
"learning_rate": 3.75e-07,
|
| 29 |
+
"loss": 0.3788,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0516545601291364,
|
| 34 |
+
"grad_norm": 28.99654186845467,
|
| 35 |
+
"learning_rate": 5e-07,
|
| 36 |
+
"loss": 0.395,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0645682001614205,
|
| 41 |
+
"grad_norm": 20.935596355684492,
|
| 42 |
+
"learning_rate": 6.249999999999999e-07,
|
| 43 |
+
"loss": 0.3741,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0774818401937046,
|
| 48 |
+
"grad_norm": 26.035750645230728,
|
| 49 |
+
"learning_rate": 7.5e-07,
|
| 50 |
+
"loss": 0.3701,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0903954802259887,
|
| 55 |
+
"grad_norm": 36.159284168135784,
|
| 56 |
+
"learning_rate": 8.75e-07,
|
| 57 |
+
"loss": 0.3642,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1033091202582728,
|
| 62 |
+
"grad_norm": 32.251355945752145,
|
| 63 |
+
"learning_rate": 1e-06,
|
| 64 |
+
"loss": 0.3473,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1162227602905569,
|
| 69 |
+
"grad_norm": 35.98055219357963,
|
| 70 |
+
"learning_rate": 1.125e-06,
|
| 71 |
+
"loss": 0.3728,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.129136400322841,
|
| 76 |
+
"grad_norm": 40.267311648036056,
|
| 77 |
+
"learning_rate": 1.2499999999999999e-06,
|
| 78 |
+
"loss": 0.3472,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1420500403551251,
|
| 83 |
+
"grad_norm": 54.431376922416256,
|
| 84 |
+
"learning_rate": 1.375e-06,
|
| 85 |
+
"loss": 0.3941,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1549636803874092,
|
| 90 |
+
"grad_norm": 52.787609103487405,
|
| 91 |
+
"learning_rate": 1.5e-06,
|
| 92 |
+
"loss": 0.3694,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1678773204196933,
|
| 97 |
+
"grad_norm": 51.470919986790086,
|
| 98 |
+
"learning_rate": 1.625e-06,
|
| 99 |
+
"loss": 0.3587,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1807909604519774,
|
| 104 |
+
"grad_norm": 44.49429708878717,
|
| 105 |
+
"learning_rate": 1.75e-06,
|
| 106 |
+
"loss": 0.3369,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.1937046004842615,
|
| 111 |
+
"grad_norm": 48.61799720707049,
|
| 112 |
+
"learning_rate": 1.8749999999999998e-06,
|
| 113 |
+
"loss": 0.3811,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2066182405165456,
|
| 118 |
+
"grad_norm": 41.649896181802205,
|
| 119 |
+
"learning_rate": 2e-06,
|
| 120 |
+
"loss": 0.3911,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2195318805488297,
|
| 125 |
+
"grad_norm": 42.44898868701815,
|
| 126 |
+
"learning_rate": 1.999740884841349e-06,
|
| 127 |
+
"loss": 0.387,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2324455205811138,
|
| 132 |
+
"grad_norm": 41.183443431747065,
|
| 133 |
+
"learning_rate": 1.9989636736467275e-06,
|
| 134 |
+
"loss": 0.3946,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2453591606133979,
|
| 139 |
+
"grad_norm": 34.50686543223266,
|
| 140 |
+
"learning_rate": 1.9976687691905393e-06,
|
| 141 |
+
"loss": 0.4088,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.258272800645682,
|
| 146 |
+
"grad_norm": 27.279515728605386,
|
| 147 |
+
"learning_rate": 1.995856842531531e-06,
|
| 148 |
+
"loss": 0.3724,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2711864406779661,
|
| 153 |
+
"grad_norm": 30.83455319215996,
|
| 154 |
+
"learning_rate": 1.993528832665031e-06,
|
| 155 |
+
"loss": 0.3866,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2841000807102502,
|
| 160 |
+
"grad_norm": 26.247716082750618,
|
| 161 |
+
"learning_rate": 1.9906859460363304e-06,
|
| 162 |
+
"loss": 0.3521,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2970137207425343,
|
| 167 |
+
"grad_norm": 27.831795874032476,
|
| 168 |
+
"learning_rate": 1.9873296559154695e-06,
|
| 169 |
+
"loss": 0.4159,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3099273607748184,
|
| 174 |
+
"grad_norm": 28.5584628971039,
|
| 175 |
+
"learning_rate": 1.983461701633742e-06,
|
| 176 |
+
"loss": 0.3767,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.3228410008071025,
|
| 181 |
+
"grad_norm": 21.378018038217974,
|
| 182 |
+
"learning_rate": 1.979084087682323e-06,
|
| 183 |
+
"loss": 0.3304,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3357546408393866,
|
| 188 |
+
"grad_norm": 23.044867887515178,
|
| 189 |
+
"learning_rate": 1.9741990826734792e-06,
|
| 190 |
+
"loss": 0.3705,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3486682808716707,
|
| 195 |
+
"grad_norm": 26.947778108928976,
|
| 196 |
+
"learning_rate": 1.968809218164906e-06,
|
| 197 |
+
"loss": 0.404,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3615819209039548,
|
| 202 |
+
"grad_norm": 22.039479721325353,
|
| 203 |
+
"learning_rate": 1.9629172873477994e-06,
|
| 204 |
+
"loss": 0.3365,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3744955609362389,
|
| 209 |
+
"grad_norm": 21.274934357850633,
|
| 210 |
+
"learning_rate": 1.956526343599335e-06,
|
| 211 |
+
"loss": 0.3306,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.387409200968523,
|
| 216 |
+
"grad_norm": 23.40835330420104,
|
| 217 |
+
"learning_rate": 1.949639698900319e-06,
|
| 218 |
+
"loss": 0.3246,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4003228410008071,
|
| 223 |
+
"grad_norm": 26.03684733839572,
|
| 224 |
+
"learning_rate": 1.9422609221188204e-06,
|
| 225 |
+
"loss": 0.3121,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4132364810330912,
|
| 230 |
+
"grad_norm": 22.817733184810667,
|
| 231 |
+
"learning_rate": 1.9343938371606708e-06,
|
| 232 |
+
"loss": 0.363,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4261501210653753,
|
| 237 |
+
"grad_norm": 27.888538264779257,
|
| 238 |
+
"learning_rate": 1.926042520987805e-06,
|
| 239 |
+
"loss": 0.3599,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4390637610976594,
|
| 244 |
+
"grad_norm": 23.664544295074048,
|
| 245 |
+
"learning_rate": 1.9172113015054528e-06,
|
| 246 |
+
"loss": 0.3259,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.4519774011299435,
|
| 251 |
+
"grad_norm": 26.814661596965937,
|
| 252 |
+
"learning_rate": 1.907904755319289e-06,
|
| 253 |
+
"loss": 0.3622,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4648910411622276,
|
| 258 |
+
"grad_norm": 23.047494151516773,
|
| 259 |
+
"learning_rate": 1.898127705363696e-06,
|
| 260 |
+
"loss": 0.3595,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4778046811945117,
|
| 265 |
+
"grad_norm": 23.769391103002327,
|
| 266 |
+
"learning_rate": 1.887885218402375e-06,
|
| 267 |
+
"loss": 0.3418,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4907183212267958,
|
| 272 |
+
"grad_norm": 21.448479890321668,
|
| 273 |
+
"learning_rate": 1.8771826024025943e-06,
|
| 274 |
+
"loss": 0.3112,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5036319612590799,
|
| 279 |
+
"grad_norm": 21.618299294862208,
|
| 280 |
+
"learning_rate": 1.8660254037844386e-06,
|
| 281 |
+
"loss": 0.2935,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.516545601291364,
|
| 286 |
+
"grad_norm": 23.83881914235937,
|
| 287 |
+
"learning_rate": 1.8544194045464886e-06,
|
| 288 |
+
"loss": 0.3487,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5294592413236481,
|
| 293 |
+
"grad_norm": 22.34191137757626,
|
| 294 |
+
"learning_rate": 1.8423706192694115e-06,
|
| 295 |
+
"loss": 0.3337,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5423728813559322,
|
| 300 |
+
"grad_norm": 22.680526606663665,
|
| 301 |
+
"learning_rate": 1.8298852919990251e-06,
|
| 302 |
+
"loss": 0.3326,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5552865213882163,
|
| 307 |
+
"grad_norm": 21.047157564860804,
|
| 308 |
+
"learning_rate": 1.8169698930104419e-06,
|
| 309 |
+
"loss": 0.3262,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5682001614205004,
|
| 314 |
+
"grad_norm": 23.210220151582366,
|
| 315 |
+
"learning_rate": 1.8036311154549781e-06,
|
| 316 |
+
"loss": 0.3531,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5811138014527845,
|
| 321 |
+
"grad_norm": 23.059400786558072,
|
| 322 |
+
"learning_rate": 1.7898758718915585e-06,
|
| 323 |
+
"loss": 0.3042,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5940274414850686,
|
| 328 |
+
"grad_norm": 20.985554392897924,
|
| 329 |
+
"learning_rate": 1.7757112907044198e-06,
|
| 330 |
+
"loss": 0.3211,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6069410815173527,
|
| 335 |
+
"grad_norm": 24.381494037046473,
|
| 336 |
+
"learning_rate": 1.7611447124089646e-06,
|
| 337 |
+
"loss": 0.3501,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6198547215496368,
|
| 342 |
+
"grad_norm": 21.163889567259087,
|
| 343 |
+
"learning_rate": 1.7461836858476855e-06,
|
| 344 |
+
"loss": 0.3407,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.632768361581921,
|
| 349 |
+
"grad_norm": 17.673857406454214,
|
| 350 |
+
"learning_rate": 1.7308359642781241e-06,
|
| 351 |
+
"loss": 0.2912,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.645682001614205,
|
| 356 |
+
"grad_norm": 20.851302876331236,
|
| 357 |
+
"learning_rate": 1.7151095013548993e-06,
|
| 358 |
+
"loss": 0.311,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6585956416464891,
|
| 363 |
+
"grad_norm": 20.07386513124459,
|
| 364 |
+
"learning_rate": 1.699012447007882e-06,
|
| 365 |
+
"loss": 0.2891,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6715092816787732,
|
| 370 |
+
"grad_norm": 16.654008916412547,
|
| 371 |
+
"learning_rate": 1.682553143218654e-06,
|
| 372 |
+
"loss": 0.2724,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6844229217110573,
|
| 377 |
+
"grad_norm": 20.770464397707823,
|
| 378 |
+
"learning_rate": 1.6657401196974403e-06,
|
| 379 |
+
"loss": 0.332,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6973365617433414,
|
| 384 |
+
"grad_norm": 22.094352414249297,
|
| 385 |
+
"learning_rate": 1.648582089462756e-06,
|
| 386 |
+
"loss": 0.344,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7102502017756255,
|
| 391 |
+
"grad_norm": 20.494531642603356,
|
| 392 |
+
"learning_rate": 1.6310879443260529e-06,
|
| 393 |
+
"loss": 0.3041,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7231638418079096,
|
| 398 |
+
"grad_norm": 19.8177079163147,
|
| 399 |
+
"learning_rate": 1.6132667502837162e-06,
|
| 400 |
+
"loss": 0.3162,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7360774818401937,
|
| 405 |
+
"grad_norm": 19.049318629916456,
|
| 406 |
+
"learning_rate": 1.5951277428187897e-06,
|
| 407 |
+
"loss": 0.3094,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7489911218724778,
|
| 412 |
+
"grad_norm": 21.597536610054288,
|
| 413 |
+
"learning_rate": 1.5766803221148673e-06,
|
| 414 |
+
"loss": 0.3325,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7619047619047619,
|
| 419 |
+
"grad_norm": 20.70892358284607,
|
| 420 |
+
"learning_rate": 1.5579340481846335e-06,
|
| 421 |
+
"loss": 0.3397,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.774818401937046,
|
| 426 |
+
"grad_norm": 20.308915238405316,
|
| 427 |
+
"learning_rate": 1.5388986359155755e-06,
|
| 428 |
+
"loss": 0.3265,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7877320419693301,
|
| 433 |
+
"grad_norm": 22.627261985685145,
|
| 434 |
+
"learning_rate": 1.5195839500354335e-06,
|
| 435 |
+
"loss": 0.329,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.8006456820016142,
|
| 440 |
+
"grad_norm": 19.565397419582293,
|
| 441 |
+
"learning_rate": 1.5e-06,
|
| 442 |
+
"loss": 0.3167,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8135593220338984,
|
| 447 |
+
"grad_norm": 18.907828874624407,
|
| 448 |
+
"learning_rate": 1.4801569348059155e-06,
|
| 449 |
+
"loss": 0.3146,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8264729620661824,
|
| 454 |
+
"grad_norm": 22.644125199961817,
|
| 455 |
+
"learning_rate": 1.460065037731152e-06,
|
| 456 |
+
"loss": 0.3072,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8393866020984665,
|
| 461 |
+
"grad_norm": 20.786645375646636,
|
| 462 |
+
"learning_rate": 1.4397347210059057e-06,
|
| 463 |
+
"loss": 0.3159,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8523002421307506,
|
| 468 |
+
"grad_norm": 18.57141588916772,
|
| 469 |
+
"learning_rate": 1.419176520416664e-06,
|
| 470 |
+
"loss": 0.3103,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8652138821630347,
|
| 475 |
+
"grad_norm": 17.225037473257153,
|
| 476 |
+
"learning_rate": 1.3984010898462415e-06,
|
| 477 |
+
"loss": 0.3027,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8781275221953188,
|
| 482 |
+
"grad_norm": 18.21584058115977,
|
| 483 |
+
"learning_rate": 1.3774191957526142e-06,
|
| 484 |
+
"loss": 0.2895,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8910411622276029,
|
| 489 |
+
"grad_norm": 20.509880420628917,
|
| 490 |
+
"learning_rate": 1.3562417115894168e-06,
|
| 491 |
+
"loss": 0.3098,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.903954802259887,
|
| 496 |
+
"grad_norm": 17.29318336230985,
|
| 497 |
+
"learning_rate": 1.334879612170986e-06,
|
| 498 |
+
"loss": 0.2954,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9168684422921711,
|
| 503 |
+
"grad_norm": 18.377551002595997,
|
| 504 |
+
"learning_rate": 1.3133439679848822e-06,
|
| 505 |
+
"loss": 0.301,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9297820823244553,
|
| 510 |
+
"grad_norm": 26.11617966128903,
|
| 511 |
+
"learning_rate": 1.2916459394548249e-06,
|
| 512 |
+
"loss": 0.3493,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9426957223567393,
|
| 517 |
+
"grad_norm": 19.11873065402881,
|
| 518 |
+
"learning_rate": 1.2697967711570242e-06,
|
| 519 |
+
"loss": 0.3055,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9556093623890234,
|
| 524 |
+
"grad_norm": 20.15187293059899,
|
| 525 |
+
"learning_rate": 1.2478077859929e-06,
|
| 526 |
+
"loss": 0.3267,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9685230024213075,
|
| 531 |
+
"grad_norm": 17.356504787996283,
|
| 532 |
+
"learning_rate": 1.2256903793212105e-06,
|
| 533 |
+
"loss": 0.3,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9814366424535916,
|
| 538 |
+
"grad_norm": 20.116822076601576,
|
| 539 |
+
"learning_rate": 1.203456013052634e-06,
|
| 540 |
+
"loss": 0.3139,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9943502824858758,
|
| 545 |
+
"grad_norm": 20.23077529953701,
|
| 546 |
+
"learning_rate": 1.1811162097098558e-06,
|
| 547 |
+
"loss": 0.2941,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.0072639225181599,
|
| 552 |
+
"grad_norm": 14.472243001169707,
|
| 553 |
+
"learning_rate": 1.1586825464562514e-06,
|
| 554 |
+
"loss": 0.217,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.020177562550444,
|
| 559 |
+
"grad_norm": 11.1760098722801,
|
| 560 |
+
"learning_rate": 1.1361666490962467e-06,
|
| 561 |
+
"loss": 0.1391,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.033091202582728,
|
| 566 |
+
"grad_norm": 12.929875937980091,
|
| 567 |
+
"learning_rate": 1.1135801860504748e-06,
|
| 568 |
+
"loss": 0.144,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.0460048426150121,
|
| 573 |
+
"grad_norm": 14.401479453367932,
|
| 574 |
+
"learning_rate": 1.0909348623088471e-06,
|
| 575 |
+
"loss": 0.1525,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.0589184826472962,
|
| 580 |
+
"grad_norm": 11.466941187008539,
|
| 581 |
+
"learning_rate": 1.068242413364671e-06,
|
| 582 |
+
"loss": 0.1492,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.0718321226795804,
|
| 587 |
+
"grad_norm": 13.844595316123245,
|
| 588 |
+
"learning_rate": 1.0455145991329637e-06,
|
| 589 |
+
"loss": 0.1317,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0847457627118644,
|
| 594 |
+
"grad_norm": 14.390882441536627,
|
| 595 |
+
"learning_rate": 1.0227631978561055e-06,
|
| 596 |
+
"loss": 0.1574,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.0976594027441484,
|
| 601 |
+
"grad_norm": 13.517598225715695,
|
| 602 |
+
"learning_rate": 1e-06,
|
| 603 |
+
"loss": 0.1409,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.1105730427764327,
|
| 608 |
+
"grad_norm": 13.993115366941224,
|
| 609 |
+
"learning_rate": 9.772368021438942e-07,
|
| 610 |
+
"loss": 0.1464,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.1234866828087167,
|
| 615 |
+
"grad_norm": 12.2423524228492,
|
| 616 |
+
"learning_rate": 9.544854008670366e-07,
|
| 617 |
+
"loss": 0.1207,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.136400322841001,
|
| 622 |
+
"grad_norm": 14.122285998417135,
|
| 623 |
+
"learning_rate": 9.317575866353291e-07,
|
| 624 |
+
"loss": 0.1251,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.149313962873285,
|
| 629 |
+
"grad_norm": 15.560999029473845,
|
| 630 |
+
"learning_rate": 9.09065137691153e-07,
|
| 631 |
+
"loss": 0.1209,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.162227602905569,
|
| 636 |
+
"grad_norm": 13.507106399175507,
|
| 637 |
+
"learning_rate": 8.86419813949525e-07,
|
| 638 |
+
"loss": 0.1091,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1751412429378532,
|
| 643 |
+
"grad_norm": 13.891730510521498,
|
| 644 |
+
"learning_rate": 8.638333509037535e-07,
|
| 645 |
+
"loss": 0.1256,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1880548829701372,
|
| 650 |
+
"grad_norm": 14.992783015617869,
|
| 651 |
+
"learning_rate": 8.413174535437487e-07,
|
| 652 |
+
"loss": 0.137,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.2009685230024214,
|
| 657 |
+
"grad_norm": 15.910575644376518,
|
| 658 |
+
"learning_rate": 8.188837902901441e-07,
|
| 659 |
+
"loss": 0.1347,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.2138821630347054,
|
| 664 |
+
"grad_norm": 19.326658299865034,
|
| 665 |
+
"learning_rate": 7.965439869473663e-07,
|
| 666 |
+
"loss": 0.1473,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.2267958030669894,
|
| 671 |
+
"grad_norm": 21.139159288452323,
|
| 672 |
+
"learning_rate": 7.743096206787893e-07,
|
| 673 |
+
"loss": 0.1449,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2397094430992737,
|
| 678 |
+
"grad_norm": 19.57794361709339,
|
| 679 |
+
"learning_rate": 7.521922140071003e-07,
|
| 680 |
+
"loss": 0.1429,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.2526230831315577,
|
| 685 |
+
"grad_norm": 18.770894293602584,
|
| 686 |
+
"learning_rate": 7.302032288429756e-07,
|
| 687 |
+
"loss": 0.1533,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.2655367231638417,
|
| 692 |
+
"grad_norm": 19.148357746011182,
|
| 693 |
+
"learning_rate": 7.083540605451749e-07,
|
| 694 |
+
"loss": 0.164,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.278450363196126,
|
| 699 |
+
"grad_norm": 15.077187564406879,
|
| 700 |
+
"learning_rate": 6.866560320151178e-07,
|
| 701 |
+
"loss": 0.1404,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.29136400322841,
|
| 706 |
+
"grad_norm": 16.058568106787366,
|
| 707 |
+
"learning_rate": 6.651203878290138e-07,
|
| 708 |
+
"loss": 0.1327,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.3042776432606942,
|
| 713 |
+
"grad_norm": 14.029168964991912,
|
| 714 |
+
"learning_rate": 6.437582884105834e-07,
|
| 715 |
+
"loss": 0.136,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.3171912832929782,
|
| 720 |
+
"grad_norm": 15.10100507251639,
|
| 721 |
+
"learning_rate": 6.225808042473857e-07,
|
| 722 |
+
"loss": 0.1149,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.3301049233252624,
|
| 727 |
+
"grad_norm": 11.718976894577146,
|
| 728 |
+
"learning_rate": 6.015989101537586e-07,
|
| 729 |
+
"loss": 0.1207,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.3430185633575464,
|
| 734 |
+
"grad_norm": 12.627764504381856,
|
| 735 |
+
"learning_rate": 5.808234795833362e-07,
|
| 736 |
+
"loss": 0.1081,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.3559322033898304,
|
| 741 |
+
"grad_norm": 15.041188760891716,
|
| 742 |
+
"learning_rate": 5.602652789940941e-07,
|
| 743 |
+
"loss": 0.1262,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.3688458434221147,
|
| 748 |
+
"grad_norm": 21.09158607094679,
|
| 749 |
+
"learning_rate": 5.399349622688478e-07,
|
| 750 |
+
"loss": 0.1314,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.3817594834543987,
|
| 755 |
+
"grad_norm": 20.9326453419062,
|
| 756 |
+
"learning_rate": 5.198430651940845e-07,
|
| 757 |
+
"loss": 0.1346,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.3946731234866827,
|
| 762 |
+
"grad_norm": 12.206782639953575,
|
| 763 |
+
"learning_rate": 5.000000000000002e-07,
|
| 764 |
+
"loss": 0.1157,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.407586763518967,
|
| 769 |
+
"grad_norm": 13.75893208112109,
|
| 770 |
+
"learning_rate": 4.804160499645667e-07,
|
| 771 |
+
"loss": 0.1384,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.420500403551251,
|
| 776 |
+
"grad_norm": 13.597298102282235,
|
| 777 |
+
"learning_rate": 4.6110136408442445e-07,
|
| 778 |
+
"loss": 0.1137,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.4334140435835352,
|
| 783 |
+
"grad_norm": 13.538763536743675,
|
| 784 |
+
"learning_rate": 4.4206595181536664e-07,
|
| 785 |
+
"loss": 0.1203,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.4463276836158192,
|
| 790 |
+
"grad_norm": 12.774053600703025,
|
| 791 |
+
"learning_rate": 4.233196778851329e-07,
|
| 792 |
+
"loss": 0.1122,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.4592413236481034,
|
| 797 |
+
"grad_norm": 12.870680344192039,
|
| 798 |
+
"learning_rate": 4.0487225718121045e-07,
|
| 799 |
+
"loss": 0.1228,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.4721549636803875,
|
| 804 |
+
"grad_norm": 15.02538486601625,
|
| 805 |
+
"learning_rate": 3.867332497162835e-07,
|
| 806 |
+
"loss": 0.1615,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.4850686037126715,
|
| 811 |
+
"grad_norm": 12.30281015942022,
|
| 812 |
+
"learning_rate": 3.6891205567394746e-07,
|
| 813 |
+
"loss": 0.099,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.4979822437449557,
|
| 818 |
+
"grad_norm": 10.419235242267765,
|
| 819 |
+
"learning_rate": 3.5141791053724404e-07,
|
| 820 |
+
"loss": 0.1096,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.5108958837772397,
|
| 825 |
+
"grad_norm": 11.16767240028713,
|
| 826 |
+
"learning_rate": 3.3425988030255945e-07,
|
| 827 |
+
"loss": 0.1199,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.5238095238095237,
|
| 832 |
+
"grad_norm": 13.158987202921047,
|
| 833 |
+
"learning_rate": 3.174468567813461e-07,
|
| 834 |
+
"loss": 0.1324,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.536723163841808,
|
| 839 |
+
"grad_norm": 11.844158983050455,
|
| 840 |
+
"learning_rate": 3.009875529921181e-07,
|
| 841 |
+
"loss": 0.1106,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.549636803874092,
|
| 846 |
+
"grad_norm": 14.439077228213756,
|
| 847 |
+
"learning_rate": 2.848904986451005e-07,
|
| 848 |
+
"loss": 0.117,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.562550443906376,
|
| 853 |
+
"grad_norm": 12.407600386041583,
|
| 854 |
+
"learning_rate": 2.6916403572187587e-07,
|
| 855 |
+
"loss": 0.1159,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.5754640839386602,
|
| 860 |
+
"grad_norm": 13.151590889043437,
|
| 861 |
+
"learning_rate": 2.538163141523145e-07,
|
| 862 |
+
"loss": 0.1236,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.5883777239709445,
|
| 867 |
+
"grad_norm": 12.342415661391085,
|
| 868 |
+
"learning_rate": 2.3885528759103536e-07,
|
| 869 |
+
"loss": 0.1191,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.6012913640032282,
|
| 874 |
+
"grad_norm": 25.46717666484534,
|
| 875 |
+
"learning_rate": 2.2428870929558007e-07,
|
| 876 |
+
"loss": 0.1464,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.6142050040355125,
|
| 881 |
+
"grad_norm": 13.835165114851305,
|
| 882 |
+
"learning_rate": 2.101241281084416e-07,
|
| 883 |
+
"loss": 0.114,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.6271186440677967,
|
| 888 |
+
"grad_norm": 12.278805533417628,
|
| 889 |
+
"learning_rate": 1.9636888454502177e-07,
|
| 890 |
+
"loss": 0.1095,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.6400322841000807,
|
| 895 |
+
"grad_norm": 9.656229159282553,
|
| 896 |
+
"learning_rate": 1.83030106989558e-07,
|
| 897 |
+
"loss": 0.1048,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.6529459241323647,
|
| 902 |
+
"grad_norm": 10.723649324385244,
|
| 903 |
+
"learning_rate": 1.7011470800097495e-07,
|
| 904 |
+
"loss": 0.1144,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.665859564164649,
|
| 909 |
+
"grad_norm": 11.18317780807273,
|
| 910 |
+
"learning_rate": 1.576293807305885e-07,
|
| 911 |
+
"loss": 0.1124,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.678773204196933,
|
| 916 |
+
"grad_norm": 19.22525610881903,
|
| 917 |
+
"learning_rate": 1.4558059545351142e-07,
|
| 918 |
+
"loss": 0.1185,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.691686844229217,
|
| 923 |
+
"grad_norm": 13.840544386244854,
|
| 924 |
+
"learning_rate": 1.3397459621556128e-07,
|
| 925 |
+
"loss": 0.1172,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.7046004842615012,
|
| 930 |
+
"grad_norm": 9.863106547796383,
|
| 931 |
+
"learning_rate": 1.2281739759740574e-07,
|
| 932 |
+
"loss": 0.1123,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.7175141242937855,
|
| 937 |
+
"grad_norm": 12.249146236722565,
|
| 938 |
+
"learning_rate": 1.1211478159762478e-07,
|
| 939 |
+
"loss": 0.1026,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.7304277643260693,
|
| 944 |
+
"grad_norm": 19.53823199805843,
|
| 945 |
+
"learning_rate": 1.0187229463630398e-07,
|
| 946 |
+
"loss": 0.1245,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.7433414043583535,
|
| 951 |
+
"grad_norm": 13.813268314685585,
|
| 952 |
+
"learning_rate": 9.209524468071095e-08,
|
| 953 |
+
"loss": 0.1259,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.7562550443906377,
|
| 958 |
+
"grad_norm": 13.002603579106614,
|
| 959 |
+
"learning_rate": 8.278869849454717e-08,
|
| 960 |
+
"loss": 0.126,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.7691686844229217,
|
| 965 |
+
"grad_norm": 11.790731140563368,
|
| 966 |
+
"learning_rate": 7.395747901219473e-08,
|
| 967 |
+
"loss": 0.1163,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.7820823244552058,
|
| 972 |
+
"grad_norm": 15.611451697022517,
|
| 973 |
+
"learning_rate": 6.560616283932896e-08,
|
| 974 |
+
"loss": 0.1226,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.79499596448749,
|
| 979 |
+
"grad_norm": 24.410099015402505,
|
| 980 |
+
"learning_rate": 5.77390778811796e-08,
|
| 981 |
+
"loss": 0.1155,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.807909604519774,
|
| 986 |
+
"grad_norm": 9.829584476780473,
|
| 987 |
+
"learning_rate": 5.036030109968081e-08,
|
| 988 |
+
"loss": 0.0972,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.820823244552058,
|
| 993 |
+
"grad_norm": 14.542859998120877,
|
| 994 |
+
"learning_rate": 4.347365640066525e-08,
|
| 995 |
+
"loss": 0.1126,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.8337368845843423,
|
| 1000 |
+
"grad_norm": 20.231191724041285,
|
| 1001 |
+
"learning_rate": 3.708271265220087e-08,
|
| 1002 |
+
"loss": 0.1175,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.8466505246166263,
|
| 1007 |
+
"grad_norm": 12.407632368224395,
|
| 1008 |
+
"learning_rate": 3.119078183509372e-08,
|
| 1009 |
+
"loss": 0.1149,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.8595641646489103,
|
| 1014 |
+
"grad_norm": 14.44661290494769,
|
| 1015 |
+
"learning_rate": 2.580091732652101e-08,
|
| 1016 |
+
"loss": 0.1181,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.8724778046811945,
|
| 1021 |
+
"grad_norm": 18.79024412476877,
|
| 1022 |
+
"learning_rate": 2.0915912317677088e-08,
|
| 1023 |
+
"loss": 0.1475,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.8853914447134787,
|
| 1028 |
+
"grad_norm": 15.376557680187412,
|
| 1029 |
+
"learning_rate": 1.6538298366257974e-08,
|
| 1030 |
+
"loss": 0.1255,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.8983050847457628,
|
| 1035 |
+
"grad_norm": 15.33944835023639,
|
| 1036 |
+
"learning_rate": 1.2670344084530382e-08,
|
| 1037 |
+
"loss": 0.1034,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.9112187247780468,
|
| 1042 |
+
"grad_norm": 13.112291504434255,
|
| 1043 |
+
"learning_rate": 9.314053963669244e-09,
|
| 1044 |
+
"loss": 0.1086,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.924132364810331,
|
| 1049 |
+
"grad_norm": 12.045688670027003,
|
| 1050 |
+
"learning_rate": 6.471167334968886e-09,
|
| 1051 |
+
"loss": 0.1096,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.937046004842615,
|
| 1056 |
+
"grad_norm": 15.127302877916637,
|
| 1057 |
+
"learning_rate": 4.143157468468716e-09,
|
| 1058 |
+
"loss": 0.1126,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.949959644874899,
|
| 1063 |
+
"grad_norm": 13.774563816455347,
|
| 1064 |
+
"learning_rate": 2.331230809460738e-09,
|
| 1065 |
+
"loss": 0.1245,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.9628732849071833,
|
| 1070 |
+
"grad_norm": 32.82358453731887,
|
| 1071 |
+
"learning_rate": 1.036326353272443e-09,
|
| 1072 |
+
"loss": 0.1181,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.9757869249394673,
|
| 1077 |
+
"grad_norm": 14.92150992963698,
|
| 1078 |
+
"learning_rate": 2.5911515865084663e-10,
|
| 1079 |
+
"loss": 0.1143,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.9887005649717513,
|
| 1084 |
+
"grad_norm": 13.25981887405618,
|
| 1085 |
+
"learning_rate": 0.0,
|
| 1086 |
+
"loss": 0.1276,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.9887005649717513,
|
| 1091 |
+
"step": 154,
|
| 1092 |
+
"total_flos": 72055202119680.0,
|
| 1093 |
+
"train_loss": 0.2331618887747263,
|
| 1094 |
+
"train_runtime": 3106.2852,
|
| 1095 |
+
"train_samples_per_second": 12.759,
|
| 1096 |
+
"train_steps_per_second": 0.05
|
| 1097 |
+
}
|
| 1098 |
+
],
|
| 1099 |
+
"logging_steps": 1,
|
| 1100 |
+
"max_steps": 154,
|
| 1101 |
+
"num_input_tokens_seen": 0,
|
| 1102 |
+
"num_train_epochs": 2,
|
| 1103 |
+
"save_steps": 100,
|
| 1104 |
+
"stateful_callbacks": {
|
| 1105 |
+
"TrainerControl": {
|
| 1106 |
+
"args": {
|
| 1107 |
+
"should_epoch_stop": false,
|
| 1108 |
+
"should_evaluate": false,
|
| 1109 |
+
"should_log": false,
|
| 1110 |
+
"should_save": true,
|
| 1111 |
+
"should_training_stop": true
|
| 1112 |
+
},
|
| 1113 |
+
"attributes": {}
|
| 1114 |
+
}
|
| 1115 |
+
},
|
| 1116 |
+
"total_flos": 72055202119680.0,
|
| 1117 |
+
"train_batch_size": 2,
|
| 1118 |
+
"trial_name": null,
|
| 1119 |
+
"trial_params": null
|
| 1120 |
+
}
|