End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.16924357414245605,
|
| 4 |
+
"eval_runtime": 378.8314,
|
| 5 |
+
"eval_samples": 5345,
|
| 6 |
+
"eval_samples_per_second": 14.109,
|
| 7 |
+
"eval_steps_per_second": 0.884,
|
| 8 |
+
"eval_wer": 0.1662755684767937,
|
| 9 |
+
"train_loss": 1.3962327468410898,
|
| 10 |
+
"train_runtime": 63316.5045,
|
| 11 |
+
"train_samples": 48750,
|
| 12 |
+
"train_samples_per_second": 11.549,
|
| 13 |
+
"train_steps_per_second": 0.18
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.16924357414245605,
|
| 4 |
+
"eval_runtime": 378.8314,
|
| 5 |
+
"eval_samples": 5345,
|
| 6 |
+
"eval_samples_per_second": 14.109,
|
| 7 |
+
"eval_steps_per_second": 0.884,
|
| 8 |
+
"eval_wer": 0.1662755684767937
|
| 9 |
+
}
|
runs/Feb01_12-24-21_job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba/events.out.tfevents.1643782206.job-c93f32d8-97c5-48e7-b5ec-c6c950f627ba.2341977.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:abf9373a194f91cb5ab43d9654213bf8f5bb7f21782a7713f5decd9fc2317cd5
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 1.3962327468410898,
|
| 4 |
+
"train_runtime": 63316.5045,
|
| 5 |
+
"train_samples": 48750,
|
| 6 |
+
"train_samples_per_second": 11.549,
|
| 7 |
+
"train_steps_per_second": 0.18
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,907 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.999015425008205,
|
| 5 |
+
"global_step": 11415,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.13,
|
| 12 |
+
"learning_rate": 3.7125e-06,
|
| 13 |
+
"loss": 9.2501,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.26,
|
| 18 |
+
"learning_rate": 7.4625e-06,
|
| 19 |
+
"loss": 6.0161,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.39,
|
| 24 |
+
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
+
"loss": 4.0333,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.53,
|
| 30 |
+
"learning_rate": 1.49625e-05,
|
| 31 |
+
"loss": 3.449,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.66,
|
| 36 |
+
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
+
"loss": 3.1638,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.66,
|
| 42 |
+
"eval_loss": 3.0685720443725586,
|
| 43 |
+
"eval_runtime": 399.7202,
|
| 44 |
+
"eval_samples_per_second": 13.372,
|
| 45 |
+
"eval_steps_per_second": 0.838,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.79,
|
| 51 |
+
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
+
"loss": 3.0226,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.92,
|
| 57 |
+
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
+
"loss": 2.9748,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.05,
|
| 63 |
+
"learning_rate": 2.99625e-05,
|
| 64 |
+
"loss": 2.9583,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.18,
|
| 69 |
+
"learning_rate": 3.37125e-05,
|
| 70 |
+
"loss": 2.9372,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.31,
|
| 75 |
+
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
+
"loss": 2.9311,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.31,
|
| 81 |
+
"eval_loss": 2.920839309692383,
|
| 82 |
+
"eval_runtime": 390.9897,
|
| 83 |
+
"eval_samples_per_second": 13.67,
|
| 84 |
+
"eval_steps_per_second": 0.857,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.45,
|
| 90 |
+
"learning_rate": 4.12125e-05,
|
| 91 |
+
"loss": 2.9082,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.58,
|
| 96 |
+
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
+
"loss": 2.8906,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.71,
|
| 102 |
+
"learning_rate": 4.871249999999999e-05,
|
| 103 |
+
"loss": 2.8496,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.84,
|
| 108 |
+
"learning_rate": 5.2462499999999994e-05,
|
| 109 |
+
"loss": 2.7544,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.97,
|
| 114 |
+
"learning_rate": 5.62125e-05,
|
| 115 |
+
"loss": 2.4175,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.97,
|
| 120 |
+
"eval_loss": 1.5009071826934814,
|
| 121 |
+
"eval_runtime": 386.8095,
|
| 122 |
+
"eval_samples_per_second": 13.818,
|
| 123 |
+
"eval_steps_per_second": 0.866,
|
| 124 |
+
"eval_wer": 0.904880074758592,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.1,
|
| 129 |
+
"learning_rate": 5.9962499999999994e-05,
|
| 130 |
+
"loss": 2.07,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.23,
|
| 135 |
+
"learning_rate": 6.37125e-05,
|
| 136 |
+
"loss": 1.7743,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.36,
|
| 141 |
+
"learning_rate": 6.746249999999999e-05,
|
| 142 |
+
"loss": 1.6002,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.5,
|
| 147 |
+
"learning_rate": 7.121249999999999e-05,
|
| 148 |
+
"loss": 1.5316,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.63,
|
| 153 |
+
"learning_rate": 7.49625e-05,
|
| 154 |
+
"loss": 1.4442,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 2.63,
|
| 159 |
+
"eval_loss": 0.4426315724849701,
|
| 160 |
+
"eval_runtime": 388.1138,
|
| 161 |
+
"eval_samples_per_second": 13.772,
|
| 162 |
+
"eval_steps_per_second": 0.863,
|
| 163 |
+
"eval_wer": 0.37832000830651025,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.76,
|
| 168 |
+
"learning_rate": 7.42113648433351e-05,
|
| 169 |
+
"loss": 1.4054,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.89,
|
| 174 |
+
"learning_rate": 7.341476367498672e-05,
|
| 175 |
+
"loss": 1.3519,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.02,
|
| 180 |
+
"learning_rate": 7.261816250663834e-05,
|
| 181 |
+
"loss": 1.3285,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.15,
|
| 186 |
+
"learning_rate": 7.182156133828995e-05,
|
| 187 |
+
"loss": 1.2726,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.28,
|
| 192 |
+
"learning_rate": 7.102496016994158e-05,
|
| 193 |
+
"loss": 1.2624,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.28,
|
| 198 |
+
"eval_loss": 0.3192683756351471,
|
| 199 |
+
"eval_runtime": 387.478,
|
| 200 |
+
"eval_samples_per_second": 13.794,
|
| 201 |
+
"eval_steps_per_second": 0.865,
|
| 202 |
+
"eval_wer": 0.29977157096874674,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.42,
|
| 207 |
+
"learning_rate": 7.022835900159319e-05,
|
| 208 |
+
"loss": 1.2737,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.55,
|
| 213 |
+
"learning_rate": 6.943175783324481e-05,
|
| 214 |
+
"loss": 1.2333,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.68,
|
| 219 |
+
"learning_rate": 6.863515666489644e-05,
|
| 220 |
+
"loss": 1.2028,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.81,
|
| 225 |
+
"learning_rate": 6.783855549654805e-05,
|
| 226 |
+
"loss": 1.1867,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.94,
|
| 231 |
+
"learning_rate": 6.704992033988316e-05,
|
| 232 |
+
"loss": 1.1889,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.94,
|
| 237 |
+
"eval_loss": 0.2867003381252289,
|
| 238 |
+
"eval_runtime": 384.7872,
|
| 239 |
+
"eval_samples_per_second": 13.891,
|
| 240 |
+
"eval_steps_per_second": 0.871,
|
| 241 |
+
"eval_wer": 0.2630048800747586,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.07,
|
| 246 |
+
"learning_rate": 6.625331917153477e-05,
|
| 247 |
+
"loss": 1.1945,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.2,
|
| 252 |
+
"learning_rate": 6.54567180031864e-05,
|
| 253 |
+
"loss": 1.1473,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.34,
|
| 258 |
+
"learning_rate": 6.466011683483802e-05,
|
| 259 |
+
"loss": 1.1408,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.47,
|
| 264 |
+
"learning_rate": 6.386351566648965e-05,
|
| 265 |
+
"loss": 1.1376,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.6,
|
| 270 |
+
"learning_rate": 6.306691449814126e-05,
|
| 271 |
+
"loss": 1.1315,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.6,
|
| 276 |
+
"eval_loss": 0.25655683875083923,
|
| 277 |
+
"eval_runtime": 384.1782,
|
| 278 |
+
"eval_samples_per_second": 13.913,
|
| 279 |
+
"eval_steps_per_second": 0.872,
|
| 280 |
+
"eval_wer": 0.2443775308898349,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 4.73,
|
| 285 |
+
"learning_rate": 6.227031332979288e-05,
|
| 286 |
+
"loss": 1.1225,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 4.86,
|
| 291 |
+
"learning_rate": 6.14737121614445e-05,
|
| 292 |
+
"loss": 1.1386,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.99,
|
| 297 |
+
"learning_rate": 6.067711099309612e-05,
|
| 298 |
+
"loss": 1.1098,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.12,
|
| 303 |
+
"learning_rate": 5.9880509824747736e-05,
|
| 304 |
+
"loss": 1.1149,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 5.26,
|
| 309 |
+
"learning_rate": 5.908390865639936e-05,
|
| 310 |
+
"loss": 1.0864,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 5.26,
|
| 315 |
+
"eval_loss": 0.23677974939346313,
|
| 316 |
+
"eval_runtime": 394.8394,
|
| 317 |
+
"eval_samples_per_second": 13.537,
|
| 318 |
+
"eval_steps_per_second": 0.848,
|
| 319 |
+
"eval_wer": 0.2293946630671789,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.39,
|
| 324 |
+
"learning_rate": 5.828730748805098e-05,
|
| 325 |
+
"loss": 1.0996,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 5.52,
|
| 330 |
+
"learning_rate": 5.74907063197026e-05,
|
| 331 |
+
"loss": 1.0714,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 5.65,
|
| 336 |
+
"learning_rate": 5.6694105151354214e-05,
|
| 337 |
+
"loss": 1.0849,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 5.78,
|
| 342 |
+
"learning_rate": 5.589750398300583e-05,
|
| 343 |
+
"loss": 1.069,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 5.91,
|
| 348 |
+
"learning_rate": 5.5100902814657456e-05,
|
| 349 |
+
"loss": 1.093,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.91,
|
| 354 |
+
"eval_loss": 0.2239702194929123,
|
| 355 |
+
"eval_runtime": 385.949,
|
| 356 |
+
"eval_samples_per_second": 13.849,
|
| 357 |
+
"eval_steps_per_second": 0.868,
|
| 358 |
+
"eval_wer": 0.21509708233828262,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 6.04,
|
| 363 |
+
"learning_rate": 5.4304301646309074e-05,
|
| 364 |
+
"loss": 1.0677,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 6.18,
|
| 369 |
+
"learning_rate": 5.35077004779607e-05,
|
| 370 |
+
"loss": 1.089,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 6.31,
|
| 375 |
+
"learning_rate": 5.2711099309612316e-05,
|
| 376 |
+
"loss": 1.0475,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 6.44,
|
| 381 |
+
"learning_rate": 5.191449814126394e-05,
|
| 382 |
+
"loss": 1.0455,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 6.57,
|
| 387 |
+
"learning_rate": 5.111789697291556e-05,
|
| 388 |
+
"loss": 1.0368,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 6.57,
|
| 393 |
+
"eval_loss": 0.21173720061779022,
|
| 394 |
+
"eval_runtime": 384.8447,
|
| 395 |
+
"eval_samples_per_second": 13.889,
|
| 396 |
+
"eval_steps_per_second": 0.87,
|
| 397 |
+
"eval_wer": 0.205575744990136,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 6.7,
|
| 402 |
+
"learning_rate": 5.032129580456717e-05,
|
| 403 |
+
"loss": 1.0459,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 6.83,
|
| 408 |
+
"learning_rate": 4.9524694636218794e-05,
|
| 409 |
+
"loss": 1.0375,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 6.96,
|
| 414 |
+
"learning_rate": 4.872809346787041e-05,
|
| 415 |
+
"loss": 1.033,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 7.1,
|
| 420 |
+
"learning_rate": 4.7931492299522036e-05,
|
| 421 |
+
"loss": 1.0291,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.23,
|
| 426 |
+
"learning_rate": 4.7134891131173654e-05,
|
| 427 |
+
"loss": 1.0178,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 7.23,
|
| 432 |
+
"eval_loss": 0.20199091732501984,
|
| 433 |
+
"eval_runtime": 385.2062,
|
| 434 |
+
"eval_samples_per_second": 13.876,
|
| 435 |
+
"eval_steps_per_second": 0.87,
|
| 436 |
+
"eval_wer": 0.19541065309936662,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 7.36,
|
| 441 |
+
"learning_rate": 4.634625597450876e-05,
|
| 442 |
+
"loss": 1.0258,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.49,
|
| 447 |
+
"learning_rate": 4.554965480616038e-05,
|
| 448 |
+
"loss": 1.0161,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 7.62,
|
| 453 |
+
"learning_rate": 4.4753053637812e-05,
|
| 454 |
+
"loss": 1.0443,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 7.75,
|
| 459 |
+
"learning_rate": 4.395645246946362e-05,
|
| 460 |
+
"loss": 1.0143,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 7.88,
|
| 465 |
+
"learning_rate": 4.3159851301115245e-05,
|
| 466 |
+
"loss": 1.0035,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 7.88,
|
| 471 |
+
"eval_loss": 0.20046772062778473,
|
| 472 |
+
"eval_runtime": 386.8691,
|
| 473 |
+
"eval_samples_per_second": 13.816,
|
| 474 |
+
"eval_steps_per_second": 0.866,
|
| 475 |
+
"eval_wer": 0.192378776866369,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 8.02,
|
| 480 |
+
"learning_rate": 4.2363250132766856e-05,
|
| 481 |
+
"loss": 1.0233,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 8.15,
|
| 486 |
+
"learning_rate": 4.1566648964418474e-05,
|
| 487 |
+
"loss": 0.9893,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 8.28,
|
| 492 |
+
"learning_rate": 4.07700477960701e-05,
|
| 493 |
+
"loss": 0.9896,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.41,
|
| 498 |
+
"learning_rate": 3.9973446627721716e-05,
|
| 499 |
+
"loss": 0.9879,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.54,
|
| 504 |
+
"learning_rate": 3.917684545937334e-05,
|
| 505 |
+
"loss": 0.9759,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 8.54,
|
| 510 |
+
"eval_loss": 0.1971476823091507,
|
| 511 |
+
"eval_runtime": 383.9373,
|
| 512 |
+
"eval_samples_per_second": 13.922,
|
| 513 |
+
"eval_steps_per_second": 0.873,
|
| 514 |
+
"eval_wer": 0.18630464126258955,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 8.67,
|
| 519 |
+
"learning_rate": 3.838024429102496e-05,
|
| 520 |
+
"loss": 0.9969,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 8.8,
|
| 525 |
+
"learning_rate": 3.758364312267658e-05,
|
| 526 |
+
"loss": 1.0003,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 8.93,
|
| 531 |
+
"learning_rate": 3.6787041954328194e-05,
|
| 532 |
+
"loss": 1.0125,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 9.07,
|
| 537 |
+
"learning_rate": 3.599044078597982e-05,
|
| 538 |
+
"loss": 0.9907,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 9.2,
|
| 543 |
+
"learning_rate": 3.5193839617631436e-05,
|
| 544 |
+
"loss": 0.9795,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 9.2,
|
| 549 |
+
"eval_loss": 0.18917934596538544,
|
| 550 |
+
"eval_runtime": 385.8026,
|
| 551 |
+
"eval_samples_per_second": 13.854,
|
| 552 |
+
"eval_steps_per_second": 0.868,
|
| 553 |
+
"eval_wer": 0.18121690374831273,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 9.33,
|
| 558 |
+
"learning_rate": 3.4397238449283054e-05,
|
| 559 |
+
"loss": 0.9768,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 9.46,
|
| 564 |
+
"learning_rate": 3.360063728093468e-05,
|
| 565 |
+
"loss": 1.0074,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 9.59,
|
| 570 |
+
"learning_rate": 3.2804036112586296e-05,
|
| 571 |
+
"loss": 0.9729,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 9.72,
|
| 576 |
+
"learning_rate": 3.2007434944237914e-05,
|
| 577 |
+
"loss": 0.9584,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 9.85,
|
| 582 |
+
"learning_rate": 3.121083377588953e-05,
|
| 583 |
+
"loss": 0.9601,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 9.85,
|
| 588 |
+
"eval_loss": 0.18627513945102692,
|
| 589 |
+
"eval_runtime": 384.7444,
|
| 590 |
+
"eval_samples_per_second": 13.892,
|
| 591 |
+
"eval_steps_per_second": 0.871,
|
| 592 |
+
"eval_wer": 0.17949330287612916,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 9.99,
|
| 597 |
+
"learning_rate": 3.0414232607541153e-05,
|
| 598 |
+
"loss": 0.9729,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 10.12,
|
| 603 |
+
"learning_rate": 2.9617631439192774e-05,
|
| 604 |
+
"loss": 0.9712,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 10.25,
|
| 609 |
+
"learning_rate": 2.882899628252788e-05,
|
| 610 |
+
"loss": 0.9746,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 10.38,
|
| 615 |
+
"learning_rate": 2.80323951141795e-05,
|
| 616 |
+
"loss": 0.9595,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 10.51,
|
| 621 |
+
"learning_rate": 2.7235793945831116e-05,
|
| 622 |
+
"loss": 0.9673,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 10.51,
|
| 627 |
+
"eval_loss": 0.1808742731809616,
|
| 628 |
+
"eval_runtime": 383.472,
|
| 629 |
+
"eval_samples_per_second": 13.938,
|
| 630 |
+
"eval_steps_per_second": 0.874,
|
| 631 |
+
"eval_wer": 0.17611878309625167,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 10.64,
|
| 636 |
+
"learning_rate": 2.6439192777482737e-05,
|
| 637 |
+
"loss": 0.9465,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 10.77,
|
| 642 |
+
"learning_rate": 2.564259160913436e-05,
|
| 643 |
+
"loss": 0.9628,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 10.91,
|
| 648 |
+
"learning_rate": 2.484599044078598e-05,
|
| 649 |
+
"loss": 0.946,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 11.04,
|
| 654 |
+
"learning_rate": 2.4049389272437594e-05,
|
| 655 |
+
"loss": 0.9316,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 11.17,
|
| 660 |
+
"learning_rate": 2.3252788104089215e-05,
|
| 661 |
+
"loss": 0.9233,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 11.17,
|
| 666 |
+
"eval_loss": 0.18183015286922455,
|
| 667 |
+
"eval_runtime": 379.914,
|
| 668 |
+
"eval_samples_per_second": 14.069,
|
| 669 |
+
"eval_steps_per_second": 0.882,
|
| 670 |
+
"eval_wer": 0.17553732738033434,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 11.3,
|
| 675 |
+
"learning_rate": 2.2456186935740836e-05,
|
| 676 |
+
"loss": 0.9521,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 11.43,
|
| 681 |
+
"learning_rate": 2.1659585767392457e-05,
|
| 682 |
+
"loss": 0.9411,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 11.56,
|
| 687 |
+
"learning_rate": 2.0862984599044075e-05,
|
| 688 |
+
"loss": 0.9333,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 11.69,
|
| 693 |
+
"learning_rate": 2.0066383430695696e-05,
|
| 694 |
+
"loss": 0.937,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 11.83,
|
| 699 |
+
"learning_rate": 1.9269782262347317e-05,
|
| 700 |
+
"loss": 0.9382,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 11.83,
|
| 705 |
+
"eval_loss": 0.17674756050109863,
|
| 706 |
+
"eval_runtime": 384.6492,
|
| 707 |
+
"eval_samples_per_second": 13.896,
|
| 708 |
+
"eval_steps_per_second": 0.871,
|
| 709 |
+
"eval_wer": 0.1741459869172464,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 11.96,
|
| 714 |
+
"learning_rate": 1.847318109399894e-05,
|
| 715 |
+
"loss": 0.9453,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 12.09,
|
| 720 |
+
"learning_rate": 1.7676579925650556e-05,
|
| 721 |
+
"loss": 0.9626,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 12.22,
|
| 726 |
+
"learning_rate": 1.6879978757302177e-05,
|
| 727 |
+
"loss": 0.9289,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 12.35,
|
| 732 |
+
"learning_rate": 1.6083377588953795e-05,
|
| 733 |
+
"loss": 0.9282,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 12.48,
|
| 738 |
+
"learning_rate": 1.5286776420605416e-05,
|
| 739 |
+
"loss": 0.9242,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 12.48,
|
| 744 |
+
"eval_loss": 0.17428240180015564,
|
| 745 |
+
"eval_runtime": 381.3036,
|
| 746 |
+
"eval_samples_per_second": 14.018,
|
| 747 |
+
"eval_steps_per_second": 0.879,
|
| 748 |
+
"eval_wer": 0.17025231024815698,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 12.61,
|
| 753 |
+
"learning_rate": 1.4490175252257036e-05,
|
| 754 |
+
"loss": 0.9273,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 12.75,
|
| 759 |
+
"learning_rate": 1.3693574083908655e-05,
|
| 760 |
+
"loss": 0.9254,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 12.88,
|
| 765 |
+
"learning_rate": 1.2896972915560275e-05,
|
| 766 |
+
"loss": 0.9209,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 13.01,
|
| 771 |
+
"learning_rate": 1.2100371747211896e-05,
|
| 772 |
+
"loss": 0.9292,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 13.14,
|
| 777 |
+
"learning_rate": 1.1303770578863514e-05,
|
| 778 |
+
"loss": 0.9703,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 13.14,
|
| 783 |
+
"eval_loss": 0.17106124758720398,
|
| 784 |
+
"eval_runtime": 382.5906,
|
| 785 |
+
"eval_samples_per_second": 13.971,
|
| 786 |
+
"eval_steps_per_second": 0.876,
|
| 787 |
+
"eval_wer": 0.17105181185754334,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 13.27,
|
| 792 |
+
"learning_rate": 1.0507169410515135e-05,
|
| 793 |
+
"loss": 0.9186,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 13.4,
|
| 798 |
+
"learning_rate": 9.710568242166754e-06,
|
| 799 |
+
"loss": 0.9227,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 13.53,
|
| 804 |
+
"learning_rate": 8.913967073818375e-06,
|
| 805 |
+
"loss": 0.909,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 13.67,
|
| 810 |
+
"learning_rate": 8.117365905469995e-06,
|
| 811 |
+
"loss": 0.9028,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 13.8,
|
| 816 |
+
"learning_rate": 7.328730748805097e-06,
|
| 817 |
+
"loss": 0.9139,
|
| 818 |
+
"step": 10500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 13.8,
|
| 822 |
+
"eval_loss": 0.1718469262123108,
|
| 823 |
+
"eval_runtime": 380.6631,
|
| 824 |
+
"eval_samples_per_second": 14.041,
|
| 825 |
+
"eval_steps_per_second": 0.88,
|
| 826 |
+
"eval_wer": 0.1671996677395909,
|
| 827 |
+
"step": 10500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.93,
|
| 831 |
+
"learning_rate": 6.532129580456718e-06,
|
| 832 |
+
"loss": 0.9146,
|
| 833 |
+
"step": 10600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 14.06,
|
| 837 |
+
"learning_rate": 5.735528412108338e-06,
|
| 838 |
+
"loss": 0.9239,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 14.19,
|
| 843 |
+
"learning_rate": 4.938927243759957e-06,
|
| 844 |
+
"loss": 0.9351,
|
| 845 |
+
"step": 10800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 14.32,
|
| 849 |
+
"learning_rate": 4.142326075411577e-06,
|
| 850 |
+
"loss": 0.9127,
|
| 851 |
+
"step": 10900
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 14.45,
|
| 855 |
+
"learning_rate": 3.3457249070631965e-06,
|
| 856 |
+
"loss": 0.9073,
|
| 857 |
+
"step": 11000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 14.45,
|
| 861 |
+
"eval_loss": 0.16996563971042633,
|
| 862 |
+
"eval_runtime": 380.1012,
|
| 863 |
+
"eval_samples_per_second": 14.062,
|
| 864 |
+
"eval_steps_per_second": 0.881,
|
| 865 |
+
"eval_wer": 0.1664728480946942,
|
| 866 |
+
"step": 11000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 14.59,
|
| 870 |
+
"learning_rate": 2.5491237387148164e-06,
|
| 871 |
+
"loss": 0.9023,
|
| 872 |
+
"step": 11100
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 14.72,
|
| 876 |
+
"learning_rate": 1.76048858204992e-06,
|
| 877 |
+
"loss": 0.908,
|
| 878 |
+
"step": 11200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 14.85,
|
| 882 |
+
"learning_rate": 9.6388741370154e-07,
|
| 883 |
+
"loss": 0.9096,
|
| 884 |
+
"step": 11300
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 14.98,
|
| 888 |
+
"learning_rate": 1.6728624535315985e-07,
|
| 889 |
+
"loss": 0.9436,
|
| 890 |
+
"step": 11400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 15.0,
|
| 894 |
+
"step": 11415,
|
| 895 |
+
"total_flos": 1.4956070851005566e+20,
|
| 896 |
+
"train_loss": 1.3962327468410898,
|
| 897 |
+
"train_runtime": 63316.5045,
|
| 898 |
+
"train_samples_per_second": 11.549,
|
| 899 |
+
"train_steps_per_second": 0.18
|
| 900 |
+
}
|
| 901 |
+
],
|
| 902 |
+
"max_steps": 11415,
|
| 903 |
+
"num_train_epochs": 15,
|
| 904 |
+
"total_flos": 1.4956070851005566e+20,
|
| 905 |
+
"trial_name": null,
|
| 906 |
+
"trial_params": null
|
| 907 |
+
}
|