Commit
•
e807b5d
1
Parent(s):
3eb4ca7
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +817 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.6920720338821411,
|
| 4 |
+
"eval_runtime": 161.6241,
|
| 5 |
+
"eval_samples": 2693,
|
| 6 |
+
"eval_samples_per_second": 16.662,
|
| 7 |
+
"eval_steps_per_second": 1.046,
|
| 8 |
+
"eval_wer": 0.3547145308659153,
|
| 9 |
+
"train_loss": 1.0753348940518237,
|
| 10 |
+
"train_runtime": 32646.4121,
|
| 11 |
+
"train_samples": 6176,
|
| 12 |
+
"train_samples_per_second": 9.459,
|
| 13 |
+
"train_steps_per_second": 0.296
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.6920720338821411,
|
| 4 |
+
"eval_runtime": 161.6241,
|
| 5 |
+
"eval_samples": 2693,
|
| 6 |
+
"eval_samples_per_second": 16.662,
|
| 7 |
+
"eval_steps_per_second": 1.046,
|
| 8 |
+
"eval_wer": 0.3547145308659153
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3850650929
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a49a6dea42ab0593abe28e44721bfebd8a4a2be769f4dc4d0e41ca62706ea67c
|
| 3 |
size 3850650929
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 1.0753348940518237,
|
| 4 |
+
"train_runtime": 32646.4121,
|
| 5 |
+
"train_samples": 6176,
|
| 6 |
+
"train_samples_per_second": 9.459,
|
| 7 |
+
"train_steps_per_second": 0.296
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,817 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"global_step": 9650,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.52,
|
| 12 |
+
"learning_rate": 3.2333333333333334e-06,
|
| 13 |
+
"loss": 11.6585,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.04,
|
| 18 |
+
"learning_rate": 6.566666666666667e-06,
|
| 19 |
+
"loss": 3.4349,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.55,
|
| 24 |
+
"learning_rate": 9.900000000000002e-06,
|
| 25 |
+
"loss": 3.0733,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.07,
|
| 30 |
+
"learning_rate": 1.3233333333333334e-05,
|
| 31 |
+
"loss": 2.0674,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.07,
|
| 36 |
+
"eval_loss": 1.341134786605835,
|
| 37 |
+
"eval_runtime": 163.433,
|
| 38 |
+
"eval_samples_per_second": 16.478,
|
| 39 |
+
"eval_steps_per_second": 1.034,
|
| 40 |
+
"eval_wer": 0.8834933444645424,
|
| 41 |
+
"step": 400
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.59,
|
| 45 |
+
"learning_rate": 1.6566666666666665e-05,
|
| 46 |
+
"loss": 1.6791,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.11,
|
| 51 |
+
"learning_rate": 1.9900000000000003e-05,
|
| 52 |
+
"loss": 1.5037,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.63,
|
| 57 |
+
"learning_rate": 2.3233333333333333e-05,
|
| 58 |
+
"loss": 1.3966,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.15,
|
| 63 |
+
"learning_rate": 2.6566666666666668e-05,
|
| 64 |
+
"loss": 1.324,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 4.15,
|
| 69 |
+
"eval_loss": 0.9311084747314453,
|
| 70 |
+
"eval_runtime": 162.7031,
|
| 71 |
+
"eval_samples_per_second": 16.552,
|
| 72 |
+
"eval_steps_per_second": 1.039,
|
| 73 |
+
"eval_wer": 0.7142062056490887,
|
| 74 |
+
"step": 800
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.66,
|
| 78 |
+
"learning_rate": 2.9900000000000002e-05,
|
| 79 |
+
"loss": 1.2854,
|
| 80 |
+
"step": 900
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.18,
|
| 84 |
+
"learning_rate": 3.323333333333333e-05,
|
| 85 |
+
"loss": 1.237,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 5.7,
|
| 90 |
+
"learning_rate": 3.656666666666666e-05,
|
| 91 |
+
"loss": 1.2105,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 6.22,
|
| 96 |
+
"learning_rate": 3.99e-05,
|
| 97 |
+
"loss": 1.2023,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.22,
|
| 102 |
+
"eval_loss": 0.8060462474822998,
|
| 103 |
+
"eval_runtime": 163.1532,
|
| 104 |
+
"eval_samples_per_second": 16.506,
|
| 105 |
+
"eval_steps_per_second": 1.036,
|
| 106 |
+
"eval_wer": 0.6170400259728213,
|
| 107 |
+
"step": 1200
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 6.74,
|
| 111 |
+
"learning_rate": 4.323333333333334e-05,
|
| 112 |
+
"loss": 1.1816,
|
| 113 |
+
"step": 1300
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 7.25,
|
| 117 |
+
"learning_rate": 4.656666666666667e-05,
|
| 118 |
+
"loss": 1.1895,
|
| 119 |
+
"step": 1400
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 7.77,
|
| 123 |
+
"learning_rate": 4.99e-05,
|
| 124 |
+
"loss": 1.1743,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 8.29,
|
| 129 |
+
"learning_rate": 4.940490797546012e-05,
|
| 130 |
+
"loss": 1.1573,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 8.29,
|
| 135 |
+
"eval_loss": 0.741489589214325,
|
| 136 |
+
"eval_runtime": 162.3319,
|
| 137 |
+
"eval_samples_per_second": 16.589,
|
| 138 |
+
"eval_steps_per_second": 1.041,
|
| 139 |
+
"eval_wer": 0.49724038773711793,
|
| 140 |
+
"step": 1600
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 8.81,
|
| 144 |
+
"learning_rate": 4.879141104294479e-05,
|
| 145 |
+
"loss": 1.1409,
|
| 146 |
+
"step": 1700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 9.33,
|
| 150 |
+
"learning_rate": 4.817791411042945e-05,
|
| 151 |
+
"loss": 1.1314,
|
| 152 |
+
"step": 1800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.84,
|
| 156 |
+
"learning_rate": 4.756441717791411e-05,
|
| 157 |
+
"loss": 1.1364,
|
| 158 |
+
"step": 1900
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 10.36,
|
| 162 |
+
"learning_rate": 4.695092024539878e-05,
|
| 163 |
+
"loss": 1.1117,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 10.36,
|
| 168 |
+
"eval_loss": 0.7248229384422302,
|
| 169 |
+
"eval_runtime": 162.1078,
|
| 170 |
+
"eval_samples_per_second": 16.612,
|
| 171 |
+
"eval_steps_per_second": 1.043,
|
| 172 |
+
"eval_wer": 0.4587913362088957,
|
| 173 |
+
"step": 2000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 10.88,
|
| 177 |
+
"learning_rate": 4.633742331288344e-05,
|
| 178 |
+
"loss": 1.1017,
|
| 179 |
+
"step": 2100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 11.4,
|
| 183 |
+
"learning_rate": 4.5723926380368096e-05,
|
| 184 |
+
"loss": 1.0979,
|
| 185 |
+
"step": 2200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 11.92,
|
| 189 |
+
"learning_rate": 4.511042944785277e-05,
|
| 190 |
+
"loss": 1.0932,
|
| 191 |
+
"step": 2300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 12.44,
|
| 195 |
+
"learning_rate": 4.4496932515337425e-05,
|
| 196 |
+
"loss": 1.0672,
|
| 197 |
+
"step": 2400
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 12.44,
|
| 201 |
+
"eval_loss": 0.6728883981704712,
|
| 202 |
+
"eval_runtime": 161.2322,
|
| 203 |
+
"eval_samples_per_second": 16.703,
|
| 204 |
+
"eval_steps_per_second": 1.048,
|
| 205 |
+
"eval_wer": 0.4349983766986689,
|
| 206 |
+
"step": 2400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 12.95,
|
| 210 |
+
"learning_rate": 4.388343558282208e-05,
|
| 211 |
+
"loss": 1.0797,
|
| 212 |
+
"step": 2500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 13.47,
|
| 216 |
+
"learning_rate": 4.3269938650306755e-05,
|
| 217 |
+
"loss": 1.0521,
|
| 218 |
+
"step": 2600
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 13.99,
|
| 222 |
+
"learning_rate": 4.265644171779141e-05,
|
| 223 |
+
"loss": 1.051,
|
| 224 |
+
"step": 2700
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.51,
|
| 228 |
+
"learning_rate": 4.204294478527608e-05,
|
| 229 |
+
"loss": 1.0336,
|
| 230 |
+
"step": 2800
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 14.51,
|
| 234 |
+
"eval_loss": 0.7116607427597046,
|
| 235 |
+
"eval_runtime": 160.91,
|
| 236 |
+
"eval_samples_per_second": 16.736,
|
| 237 |
+
"eval_steps_per_second": 1.05,
|
| 238 |
+
"eval_wer": 0.4345809563563842,
|
| 239 |
+
"step": 2800
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 15.03,
|
| 243 |
+
"learning_rate": 4.142944785276074e-05,
|
| 244 |
+
"loss": 1.0411,
|
| 245 |
+
"step": 2900
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 15.54,
|
| 249 |
+
"learning_rate": 4.08159509202454e-05,
|
| 250 |
+
"loss": 1.0135,
|
| 251 |
+
"step": 3000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 16.06,
|
| 255 |
+
"learning_rate": 4.0202453987730065e-05,
|
| 256 |
+
"loss": 1.0102,
|
| 257 |
+
"step": 3100
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 16.58,
|
| 261 |
+
"learning_rate": 3.958895705521473e-05,
|
| 262 |
+
"loss": 1.0025,
|
| 263 |
+
"step": 3200
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 16.58,
|
| 267 |
+
"eval_loss": 0.7018883228302002,
|
| 268 |
+
"eval_runtime": 160.9879,
|
| 269 |
+
"eval_samples_per_second": 16.728,
|
| 270 |
+
"eval_steps_per_second": 1.05,
|
| 271 |
+
"eval_wer": 0.4271601502713232,
|
| 272 |
+
"step": 3200
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 17.1,
|
| 276 |
+
"learning_rate": 3.897546012269939e-05,
|
| 277 |
+
"loss": 0.9754,
|
| 278 |
+
"step": 3300
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 17.62,
|
| 282 |
+
"learning_rate": 3.836196319018405e-05,
|
| 283 |
+
"loss": 0.9695,
|
| 284 |
+
"step": 3400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 18.13,
|
| 288 |
+
"learning_rate": 3.774846625766871e-05,
|
| 289 |
+
"loss": 0.9663,
|
| 290 |
+
"step": 3500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 18.65,
|
| 294 |
+
"learning_rate": 3.7134969325153375e-05,
|
| 295 |
+
"loss": 0.9578,
|
| 296 |
+
"step": 3600
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 18.65,
|
| 300 |
+
"eval_loss": 0.6791622042655945,
|
| 301 |
+
"eval_runtime": 162.2664,
|
| 302 |
+
"eval_samples_per_second": 16.596,
|
| 303 |
+
"eval_steps_per_second": 1.041,
|
| 304 |
+
"eval_wer": 0.41176197764482164,
|
| 305 |
+
"step": 3600
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 19.17,
|
| 309 |
+
"learning_rate": 3.652147239263804e-05,
|
| 310 |
+
"loss": 0.9392,
|
| 311 |
+
"step": 3700
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 19.69,
|
| 315 |
+
"learning_rate": 3.59079754601227e-05,
|
| 316 |
+
"loss": 0.9504,
|
| 317 |
+
"step": 3800
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 20.21,
|
| 321 |
+
"learning_rate": 3.529447852760737e-05,
|
| 322 |
+
"loss": 0.9389,
|
| 323 |
+
"step": 3900
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 20.73,
|
| 327 |
+
"learning_rate": 3.468098159509203e-05,
|
| 328 |
+
"loss": 0.9272,
|
| 329 |
+
"step": 4000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.73,
|
| 333 |
+
"eval_loss": 0.6863027811050415,
|
| 334 |
+
"eval_runtime": 161.4481,
|
| 335 |
+
"eval_samples_per_second": 16.68,
|
| 336 |
+
"eval_steps_per_second": 1.047,
|
| 337 |
+
"eval_wer": 0.41561152080144703,
|
| 338 |
+
"step": 4000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 21.24,
|
| 342 |
+
"learning_rate": 3.4067484662576685e-05,
|
| 343 |
+
"loss": 0.926,
|
| 344 |
+
"step": 4100
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 21.76,
|
| 348 |
+
"learning_rate": 3.3453987730061356e-05,
|
| 349 |
+
"loss": 0.9145,
|
| 350 |
+
"step": 4200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 22.28,
|
| 354 |
+
"learning_rate": 3.284662576687117e-05,
|
| 355 |
+
"loss": 0.9118,
|
| 356 |
+
"step": 4300
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 22.8,
|
| 360 |
+
"learning_rate": 3.223312883435583e-05,
|
| 361 |
+
"loss": 0.9321,
|
| 362 |
+
"step": 4400
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 22.8,
|
| 366 |
+
"eval_loss": 0.6534942984580994,
|
| 367 |
+
"eval_runtime": 160.5933,
|
| 368 |
+
"eval_samples_per_second": 16.769,
|
| 369 |
+
"eval_steps_per_second": 1.052,
|
| 370 |
+
"eval_wer": 0.3971986457028895,
|
| 371 |
+
"step": 4400
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 23.32,
|
| 375 |
+
"learning_rate": 3.161963190184049e-05,
|
| 376 |
+
"loss": 0.9064,
|
| 377 |
+
"step": 4500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 23.83,
|
| 381 |
+
"learning_rate": 3.100613496932516e-05,
|
| 382 |
+
"loss": 0.8915,
|
| 383 |
+
"step": 4600
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 24.35,
|
| 387 |
+
"learning_rate": 3.0392638036809817e-05,
|
| 388 |
+
"loss": 0.8907,
|
| 389 |
+
"step": 4700
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 24.87,
|
| 393 |
+
"learning_rate": 2.977914110429448e-05,
|
| 394 |
+
"loss": 0.8802,
|
| 395 |
+
"step": 4800
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 24.87,
|
| 399 |
+
"eval_loss": 0.6766195297241211,
|
| 400 |
+
"eval_runtime": 161.8159,
|
| 401 |
+
"eval_samples_per_second": 16.642,
|
| 402 |
+
"eval_steps_per_second": 1.044,
|
| 403 |
+
"eval_wer": 0.39056630026436623,
|
| 404 |
+
"step": 4800
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 25.39,
|
| 408 |
+
"learning_rate": 2.9171779141104295e-05,
|
| 409 |
+
"loss": 0.8834,
|
| 410 |
+
"step": 4900
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 25.91,
|
| 414 |
+
"learning_rate": 2.855828220858896e-05,
|
| 415 |
+
"loss": 0.866,
|
| 416 |
+
"step": 5000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 26.42,
|
| 420 |
+
"learning_rate": 2.794478527607362e-05,
|
| 421 |
+
"loss": 0.8497,
|
| 422 |
+
"step": 5100
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 26.94,
|
| 426 |
+
"learning_rate": 2.7331288343558282e-05,
|
| 427 |
+
"loss": 0.844,
|
| 428 |
+
"step": 5200
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 26.94,
|
| 432 |
+
"eval_loss": 0.678193986415863,
|
| 433 |
+
"eval_runtime": 161.4084,
|
| 434 |
+
"eval_samples_per_second": 16.684,
|
| 435 |
+
"eval_steps_per_second": 1.047,
|
| 436 |
+
"eval_wer": 0.39492602383933956,
|
| 437 |
+
"step": 5200
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 27.46,
|
| 441 |
+
"learning_rate": 2.6717791411042947e-05,
|
| 442 |
+
"loss": 0.8466,
|
| 443 |
+
"step": 5300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 27.98,
|
| 447 |
+
"learning_rate": 2.6104294478527608e-05,
|
| 448 |
+
"loss": 0.831,
|
| 449 |
+
"step": 5400
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 28.5,
|
| 453 |
+
"learning_rate": 2.5490797546012273e-05,
|
| 454 |
+
"loss": 0.82,
|
| 455 |
+
"step": 5500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 29.02,
|
| 459 |
+
"learning_rate": 2.4877300613496934e-05,
|
| 460 |
+
"loss": 0.8387,
|
| 461 |
+
"step": 5600
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 29.02,
|
| 465 |
+
"eval_loss": 0.6915661692619324,
|
| 466 |
+
"eval_runtime": 161.098,
|
| 467 |
+
"eval_samples_per_second": 16.717,
|
| 468 |
+
"eval_steps_per_second": 1.049,
|
| 469 |
+
"eval_wer": 0.3921432215574417,
|
| 470 |
+
"step": 5600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 29.53,
|
| 474 |
+
"learning_rate": 2.4263803680981595e-05,
|
| 475 |
+
"loss": 0.8205,
|
| 476 |
+
"step": 5700
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 30.05,
|
| 480 |
+
"learning_rate": 2.3650306748466257e-05,
|
| 481 |
+
"loss": 0.8139,
|
| 482 |
+
"step": 5800
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 30.57,
|
| 486 |
+
"learning_rate": 2.303680981595092e-05,
|
| 487 |
+
"loss": 0.8197,
|
| 488 |
+
"step": 5900
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 31.09,
|
| 492 |
+
"learning_rate": 2.2423312883435586e-05,
|
| 493 |
+
"loss": 0.8042,
|
| 494 |
+
"step": 6000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 31.09,
|
| 498 |
+
"eval_loss": 0.6805949807167053,
|
| 499 |
+
"eval_runtime": 161.7661,
|
| 500 |
+
"eval_samples_per_second": 16.647,
|
| 501 |
+
"eval_steps_per_second": 1.045,
|
| 502 |
+
"eval_wer": 0.3796669913269329,
|
| 503 |
+
"step": 6000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 31.61,
|
| 507 |
+
"learning_rate": 2.1809815950920244e-05,
|
| 508 |
+
"loss": 0.7911,
|
| 509 |
+
"step": 6100
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 32.12,
|
| 513 |
+
"learning_rate": 2.119631901840491e-05,
|
| 514 |
+
"loss": 0.7926,
|
| 515 |
+
"step": 6200
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 32.64,
|
| 519 |
+
"learning_rate": 2.0582822085889574e-05,
|
| 520 |
+
"loss": 0.7922,
|
| 521 |
+
"step": 6300
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 33.16,
|
| 525 |
+
"learning_rate": 1.9969325153374235e-05,
|
| 526 |
+
"loss": 0.793,
|
| 527 |
+
"step": 6400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 33.16,
|
| 531 |
+
"eval_loss": 0.7119700908660889,
|
| 532 |
+
"eval_runtime": 161.5775,
|
| 533 |
+
"eval_samples_per_second": 16.667,
|
| 534 |
+
"eval_steps_per_second": 1.046,
|
| 535 |
+
"eval_wer": 0.383052734103242,
|
| 536 |
+
"step": 6400
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 33.68,
|
| 540 |
+
"learning_rate": 1.9355828220858896e-05,
|
| 541 |
+
"loss": 0.7779,
|
| 542 |
+
"step": 6500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 34.2,
|
| 546 |
+
"learning_rate": 1.874233128834356e-05,
|
| 547 |
+
"loss": 0.7739,
|
| 548 |
+
"step": 6600
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 34.72,
|
| 552 |
+
"learning_rate": 1.8128834355828222e-05,
|
| 553 |
+
"loss": 0.7579,
|
| 554 |
+
"step": 6700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 35.23,
|
| 558 |
+
"learning_rate": 1.7515337423312884e-05,
|
| 559 |
+
"loss": 0.7567,
|
| 560 |
+
"step": 6800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 35.23,
|
| 564 |
+
"eval_loss": 0.686227560043335,
|
| 565 |
+
"eval_runtime": 161.2767,
|
| 566 |
+
"eval_samples_per_second": 16.698,
|
| 567 |
+
"eval_steps_per_second": 1.048,
|
| 568 |
+
"eval_wer": 0.3808264922777237,
|
| 569 |
+
"step": 6800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 35.75,
|
| 573 |
+
"learning_rate": 1.6901840490797545e-05,
|
| 574 |
+
"loss": 0.7623,
|
| 575 |
+
"step": 6900
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 36.27,
|
| 579 |
+
"learning_rate": 1.628834355828221e-05,
|
| 580 |
+
"loss": 0.7603,
|
| 581 |
+
"step": 7000
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 36.79,
|
| 585 |
+
"learning_rate": 1.5674846625766874e-05,
|
| 586 |
+
"loss": 0.7416,
|
| 587 |
+
"step": 7100
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 37.31,
|
| 591 |
+
"learning_rate": 1.5061349693251534e-05,
|
| 592 |
+
"loss": 0.7463,
|
| 593 |
+
"step": 7200
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 37.31,
|
| 597 |
+
"eval_loss": 0.6893351078033447,
|
| 598 |
+
"eval_runtime": 161.5435,
|
| 599 |
+
"eval_samples_per_second": 16.67,
|
| 600 |
+
"eval_steps_per_second": 1.046,
|
| 601 |
+
"eval_wer": 0.3709011641389546,
|
| 602 |
+
"step": 7200
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 37.82,
|
| 606 |
+
"learning_rate": 1.445398773006135e-05,
|
| 607 |
+
"loss": 0.7333,
|
| 608 |
+
"step": 7300
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 38.34,
|
| 612 |
+
"learning_rate": 1.3840490797546013e-05,
|
| 613 |
+
"loss": 0.7142,
|
| 614 |
+
"step": 7400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 38.86,
|
| 618 |
+
"learning_rate": 1.3226993865030676e-05,
|
| 619 |
+
"loss": 0.7291,
|
| 620 |
+
"step": 7500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 39.38,
|
| 624 |
+
"learning_rate": 1.2613496932515337e-05,
|
| 625 |
+
"loss": 0.7053,
|
| 626 |
+
"step": 7600
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 39.38,
|
| 630 |
+
"eval_loss": 0.7095528841018677,
|
| 631 |
+
"eval_runtime": 160.1545,
|
| 632 |
+
"eval_samples_per_second": 16.815,
|
| 633 |
+
"eval_steps_per_second": 1.055,
|
| 634 |
+
"eval_wer": 0.3700663234543852,
|
| 635 |
+
"step": 7600
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 39.9,
|
| 639 |
+
"learning_rate": 1.2e-05,
|
| 640 |
+
"loss": 0.727,
|
| 641 |
+
"step": 7700
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 40.41,
|
| 645 |
+
"learning_rate": 1.1386503067484663e-05,
|
| 646 |
+
"loss": 0.7157,
|
| 647 |
+
"step": 7800
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 40.93,
|
| 651 |
+
"learning_rate": 1.0773006134969325e-05,
|
| 652 |
+
"loss": 0.6954,
|
| 653 |
+
"step": 7900
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 41.45,
|
| 657 |
+
"learning_rate": 1.015950920245399e-05,
|
| 658 |
+
"loss": 0.6906,
|
| 659 |
+
"step": 8000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 41.45,
|
| 663 |
+
"eval_loss": 0.6921066045761108,
|
| 664 |
+
"eval_runtime": 161.634,
|
| 665 |
+
"eval_samples_per_second": 16.661,
|
| 666 |
+
"eval_steps_per_second": 1.046,
|
| 667 |
+
"eval_wer": 0.3675618014006771,
|
| 668 |
+
"step": 8000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 41.97,
|
| 672 |
+
"learning_rate": 9.54601226993865e-06,
|
| 673 |
+
"loss": 0.6817,
|
| 674 |
+
"step": 8100
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 42.49,
|
| 678 |
+
"learning_rate": 8.932515337423314e-06,
|
| 679 |
+
"loss": 0.6861,
|
| 680 |
+
"step": 8200
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 43.01,
|
| 684 |
+
"learning_rate": 8.319018404907975e-06,
|
| 685 |
+
"loss": 0.6942,
|
| 686 |
+
"step": 8300
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 43.52,
|
| 690 |
+
"learning_rate": 7.705521472392638e-06,
|
| 691 |
+
"loss": 0.6891,
|
| 692 |
+
"step": 8400
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 43.52,
|
| 696 |
+
"eval_loss": 0.7166885733604431,
|
| 697 |
+
"eval_runtime": 160.1699,
|
| 698 |
+
"eval_samples_per_second": 16.813,
|
| 699 |
+
"eval_steps_per_second": 1.055,
|
| 700 |
+
"eval_wer": 0.3662631603357915,
|
| 701 |
+
"step": 8400
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 44.04,
|
| 705 |
+
"learning_rate": 7.092024539877301e-06,
|
| 706 |
+
"loss": 0.6844,
|
| 707 |
+
"step": 8500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 44.56,
|
| 711 |
+
"learning_rate": 6.478527607361963e-06,
|
| 712 |
+
"loss": 0.6693,
|
| 713 |
+
"step": 8600
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 45.08,
|
| 717 |
+
"learning_rate": 5.865030674846626e-06,
|
| 718 |
+
"loss": 0.6703,
|
| 719 |
+
"step": 8700
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 45.6,
|
| 723 |
+
"learning_rate": 5.251533742331288e-06,
|
| 724 |
+
"loss": 0.658,
|
| 725 |
+
"step": 8800
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 45.6,
|
| 729 |
+
"eval_loss": 0.6833298802375793,
|
| 730 |
+
"eval_runtime": 159.9929,
|
| 731 |
+
"eval_samples_per_second": 16.832,
|
| 732 |
+
"eval_steps_per_second": 1.056,
|
| 733 |
+
"eval_wer": 0.3580075135661611,
|
| 734 |
+
"step": 8800
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 46.11,
|
| 738 |
+
"learning_rate": 4.6380368098159506e-06,
|
| 739 |
+
"loss": 0.6549,
|
| 740 |
+
"step": 8900
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 46.63,
|
| 744 |
+
"learning_rate": 4.0245398773006136e-06,
|
| 745 |
+
"loss": 0.6665,
|
| 746 |
+
"step": 9000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 47.15,
|
| 750 |
+
"learning_rate": 3.411042944785276e-06,
|
| 751 |
+
"loss": 0.6537,
|
| 752 |
+
"step": 9100
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 47.67,
|
| 756 |
+
"learning_rate": 2.7975460122699388e-06,
|
| 757 |
+
"loss": 0.6576,
|
| 758 |
+
"step": 9200
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 47.67,
|
| 762 |
+
"eval_loss": 0.6914297342300415,
|
| 763 |
+
"eval_runtime": 160.7459,
|
| 764 |
+
"eval_samples_per_second": 16.753,
|
| 765 |
+
"eval_steps_per_second": 1.051,
|
| 766 |
+
"eval_wer": 0.3569407726914336,
|
| 767 |
+
"step": 9200
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 48.19,
|
| 771 |
+
"learning_rate": 2.1840490797546013e-06,
|
| 772 |
+
"loss": 0.6601,
|
| 773 |
+
"step": 9300
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 48.7,
|
| 777 |
+
"learning_rate": 1.570552147239264e-06,
|
| 778 |
+
"loss": 0.6327,
|
| 779 |
+
"step": 9400
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 49.22,
|
| 783 |
+
"learning_rate": 9.570552147239263e-07,
|
| 784 |
+
"loss": 0.6448,
|
| 785 |
+
"step": 9500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 49.74,
|
| 789 |
+
"learning_rate": 3.43558282208589e-07,
|
| 790 |
+
"loss": 0.6358,
|
| 791 |
+
"step": 9600
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 49.74,
|
| 795 |
+
"eval_loss": 0.6921652555465698,
|
| 796 |
+
"eval_runtime": 161.3961,
|
| 797 |
+
"eval_samples_per_second": 16.686,
|
| 798 |
+
"eval_steps_per_second": 1.047,
|
| 799 |
+
"eval_wer": 0.3551319512082,
|
| 800 |
+
"step": 9600
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 50.0,
|
| 804 |
+
"step": 9650,
|
| 805 |
+
"total_flos": 1.2240549585337914e+20,
|
| 806 |
+
"train_loss": 1.0753348940518237,
|
| 807 |
+
"train_runtime": 32646.4121,
|
| 808 |
+
"train_samples_per_second": 9.459,
|
| 809 |
+
"train_steps_per_second": 0.296
|
| 810 |
+
}
|
| 811 |
+
],
|
| 812 |
+
"max_steps": 9650,
|
| 813 |
+
"num_train_epochs": 50,
|
| 814 |
+
"total_flos": 1.2240549585337914e+20,
|
| 815 |
+
"trial_name": null,
|
| 816 |
+
"trial_params": null
|
| 817 |
+
}
|