End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.5,
|
| 3 |
+
"eval_loss": 2.1667492389678955,
|
| 4 |
+
"eval_runtime": 6455.0764,
|
| 5 |
+
"eval_sacrebleu": 118.26401526639184,
|
| 6 |
+
"eval_samples_per_second": 0.526,
|
| 7 |
+
"eval_steps_per_second": 0.263,
|
| 8 |
+
"eval_wer": 686.689662169017,
|
| 9 |
+
"train_loss": 0.524174674898386,
|
| 10 |
+
"train_runtime": 70623.6184,
|
| 11 |
+
"train_samples_per_second": 0.906,
|
| 12 |
+
"train_steps_per_second": 0.028
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.5,
|
| 3 |
+
"eval_loss": 2.1667492389678955,
|
| 4 |
+
"eval_runtime": 6455.0764,
|
| 5 |
+
"eval_sacrebleu": 118.26401526639184,
|
| 6 |
+
"eval_samples_per_second": 0.526,
|
| 7 |
+
"eval_steps_per_second": 0.263,
|
| 8 |
+
"eval_wer": 686.689662169017
|
| 9 |
+
}
|
runs/Apr07_09-53-36_fasso.esat.kuleuven.be/events.out.tfevents.1680931239.fasso.esat.kuleuven.be.754703.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57b7a54184c3f648a3365a832622dcfe80947071e8c6b0f930bb1103ad4bfa69
|
| 3 |
+
size 459
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.5,
|
| 3 |
+
"train_loss": 0.524174674898386,
|
| 4 |
+
"train_runtime": 70623.6184,
|
| 5 |
+
"train_samples_per_second": 0.906,
|
| 6 |
+
"train_steps_per_second": 0.028
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,905 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 118.26401526639184,
|
| 3 |
+
"best_model_checkpoint": "/esat/audioslave/qmeeus/exp/huggingface/outputs/la-whisper-small-covost2/checkpoint-50",
|
| 4 |
+
"epoch": 4.501969611705121,
|
| 5 |
+
"global_step": 2000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 6.900000000000001e-06,
|
| 13 |
+
"loss": 5.246,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.11,
|
| 18 |
+
"learning_rate": 1.44e-05,
|
| 19 |
+
"loss": 2.6943,
|
| 20 |
+
"step": 50
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.11,
|
| 24 |
+
"eval_loss": 2.1667492389678955,
|
| 25 |
+
"eval_runtime": 6734.5431,
|
| 26 |
+
"eval_sacrebleu": 118.26401526639184,
|
| 27 |
+
"eval_samples_per_second": 0.505,
|
| 28 |
+
"eval_steps_per_second": 0.252,
|
| 29 |
+
"eval_wer": 686.689662169017,
|
| 30 |
+
"step": 50
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.17,
|
| 34 |
+
"learning_rate": 2.19e-05,
|
| 35 |
+
"loss": 1.9855,
|
| 36 |
+
"step": 75
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.23,
|
| 40 |
+
"learning_rate": 2.94e-05,
|
| 41 |
+
"loss": 1.5505,
|
| 42 |
+
"step": 100
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.23,
|
| 46 |
+
"eval_loss": 1.6016461849212646,
|
| 47 |
+
"eval_runtime": 7083.1054,
|
| 48 |
+
"eval_sacrebleu": 259.9306589539216,
|
| 49 |
+
"eval_samples_per_second": 0.48,
|
| 50 |
+
"eval_steps_per_second": 0.24,
|
| 51 |
+
"eval_wer": 165.6115789117717,
|
| 52 |
+
"step": 100
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.28,
|
| 56 |
+
"learning_rate": 2.9636842105263158e-05,
|
| 57 |
+
"loss": 1.5057,
|
| 58 |
+
"step": 125
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.34,
|
| 62 |
+
"learning_rate": 2.9242105263157893e-05,
|
| 63 |
+
"loss": 1.4093,
|
| 64 |
+
"step": 150
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.34,
|
| 68 |
+
"eval_loss": 1.58578622341156,
|
| 69 |
+
"eval_runtime": 3061.6014,
|
| 70 |
+
"eval_sacrebleu": 496.73346747625675,
|
| 71 |
+
"eval_samples_per_second": 1.11,
|
| 72 |
+
"eval_steps_per_second": 0.555,
|
| 73 |
+
"eval_wer": 197.01058469446417,
|
| 74 |
+
"step": 150
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.39,
|
| 78 |
+
"learning_rate": 2.8847368421052634e-05,
|
| 79 |
+
"loss": 1.3543,
|
| 80 |
+
"step": 175
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.45,
|
| 84 |
+
"learning_rate": 2.845263157894737e-05,
|
| 85 |
+
"loss": 1.3209,
|
| 86 |
+
"step": 200
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.45,
|
| 90 |
+
"eval_loss": 1.5648311376571655,
|
| 91 |
+
"eval_runtime": 4753.6305,
|
| 92 |
+
"eval_sacrebleu": 724.2490824821294,
|
| 93 |
+
"eval_samples_per_second": 0.715,
|
| 94 |
+
"eval_steps_per_second": 0.357,
|
| 95 |
+
"eval_wer": 121.87954414798281,
|
| 96 |
+
"step": 200
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.51,
|
| 100 |
+
"learning_rate": 2.8057894736842104e-05,
|
| 101 |
+
"loss": 1.2593,
|
| 102 |
+
"step": 225
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.56,
|
| 106 |
+
"learning_rate": 2.7663157894736842e-05,
|
| 107 |
+
"loss": 1.2941,
|
| 108 |
+
"step": 250
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.56,
|
| 112 |
+
"eval_loss": 1.5596481561660767,
|
| 113 |
+
"eval_runtime": 1255.7531,
|
| 114 |
+
"eval_sacrebleu": 820.1240805356855,
|
| 115 |
+
"eval_samples_per_second": 2.706,
|
| 116 |
+
"eval_steps_per_second": 1.353,
|
| 117 |
+
"eval_wer": 161.71587027831322,
|
| 118 |
+
"step": 250
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.62,
|
| 122 |
+
"learning_rate": 2.726842105263158e-05,
|
| 123 |
+
"loss": 1.2667,
|
| 124 |
+
"step": 275
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.68,
|
| 128 |
+
"learning_rate": 2.687368421052632e-05,
|
| 129 |
+
"loss": 1.2078,
|
| 130 |
+
"step": 300
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.68,
|
| 134 |
+
"eval_loss": 1.5073782205581665,
|
| 135 |
+
"eval_runtime": 1290.5503,
|
| 136 |
+
"eval_sacrebleu": 1022.0043170726522,
|
| 137 |
+
"eval_samples_per_second": 2.633,
|
| 138 |
+
"eval_steps_per_second": 1.316,
|
| 139 |
+
"eval_wer": 140.3875418484326,
|
| 140 |
+
"step": 300
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.73,
|
| 144 |
+
"learning_rate": 2.6478947368421053e-05,
|
| 145 |
+
"loss": 1.1966,
|
| 146 |
+
"step": 325
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.79,
|
| 150 |
+
"learning_rate": 2.6084210526315788e-05,
|
| 151 |
+
"loss": 1.1532,
|
| 152 |
+
"step": 350
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.79,
|
| 156 |
+
"eval_loss": 1.497180700302124,
|
| 157 |
+
"eval_runtime": 4184.4699,
|
| 158 |
+
"eval_sacrebleu": 174.83502311336932,
|
| 159 |
+
"eval_samples_per_second": 0.812,
|
| 160 |
+
"eval_steps_per_second": 0.406,
|
| 161 |
+
"eval_wer": 610.3716478982788,
|
| 162 |
+
"step": 350
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.84,
|
| 166 |
+
"learning_rate": 2.568947368421053e-05,
|
| 167 |
+
"loss": 1.1028,
|
| 168 |
+
"step": 375
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.9,
|
| 172 |
+
"learning_rate": 2.5294736842105264e-05,
|
| 173 |
+
"loss": 1.0167,
|
| 174 |
+
"step": 400
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.9,
|
| 178 |
+
"eval_loss": 1.4551299810409546,
|
| 179 |
+
"eval_runtime": 791.8848,
|
| 180 |
+
"eval_sacrebleu": 1904.0920904088528,
|
| 181 |
+
"eval_samples_per_second": 4.291,
|
| 182 |
+
"eval_steps_per_second": 2.146,
|
| 183 |
+
"eval_wer": 82.7635183118596,
|
| 184 |
+
"step": 400
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.96,
|
| 188 |
+
"learning_rate": 2.49e-05,
|
| 189 |
+
"loss": 1.0012,
|
| 190 |
+
"step": 425
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.01,
|
| 194 |
+
"learning_rate": 2.4505263157894737e-05,
|
| 195 |
+
"loss": 0.8842,
|
| 196 |
+
"step": 450
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.01,
|
| 200 |
+
"eval_loss": 1.4295964241027832,
|
| 201 |
+
"eval_runtime": 800.8358,
|
| 202 |
+
"eval_sacrebleu": 1883.6113185691272,
|
| 203 |
+
"eval_samples_per_second": 4.243,
|
| 204 |
+
"eval_steps_per_second": 2.122,
|
| 205 |
+
"eval_wer": 81.39055155388726,
|
| 206 |
+
"step": 450
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.07,
|
| 210 |
+
"learning_rate": 2.4110526315789475e-05,
|
| 211 |
+
"loss": 0.5407,
|
| 212 |
+
"step": 475
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.13,
|
| 216 |
+
"learning_rate": 2.371578947368421e-05,
|
| 217 |
+
"loss": 0.5619,
|
| 218 |
+
"step": 500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.13,
|
| 222 |
+
"eval_loss": 1.4332941770553589,
|
| 223 |
+
"eval_runtime": 899.1709,
|
| 224 |
+
"eval_sacrebleu": 1817.9439847694805,
|
| 225 |
+
"eval_samples_per_second": 3.779,
|
| 226 |
+
"eval_steps_per_second": 1.89,
|
| 227 |
+
"eval_wer": 84.93118257752528,
|
| 228 |
+
"step": 500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.18,
|
| 232 |
+
"learning_rate": 2.3321052631578948e-05,
|
| 233 |
+
"loss": 0.5429,
|
| 234 |
+
"step": 525
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.24,
|
| 238 |
+
"learning_rate": 2.2926315789473683e-05,
|
| 239 |
+
"loss": 0.5523,
|
| 240 |
+
"step": 550
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.24,
|
| 244 |
+
"eval_loss": 1.4236555099487305,
|
| 245 |
+
"eval_runtime": 1007.519,
|
| 246 |
+
"eval_sacrebleu": 1517.1744129968472,
|
| 247 |
+
"eval_samples_per_second": 3.373,
|
| 248 |
+
"eval_steps_per_second": 1.686,
|
| 249 |
+
"eval_wer": 104.09184674174021,
|
| 250 |
+
"step": 550
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.29,
|
| 254 |
+
"learning_rate": 2.2531578947368424e-05,
|
| 255 |
+
"loss": 0.5684,
|
| 256 |
+
"step": 575
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.35,
|
| 260 |
+
"learning_rate": 2.213684210526316e-05,
|
| 261 |
+
"loss": 0.4881,
|
| 262 |
+
"step": 600
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.35,
|
| 266 |
+
"eval_loss": 1.4413496255874634,
|
| 267 |
+
"eval_runtime": 983.5837,
|
| 268 |
+
"eval_sacrebleu": 1650.1807231081336,
|
| 269 |
+
"eval_samples_per_second": 3.455,
|
| 270 |
+
"eval_steps_per_second": 1.727,
|
| 271 |
+
"eval_wer": 97.20672280274594,
|
| 272 |
+
"step": 600
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.41,
|
| 276 |
+
"learning_rate": 2.1742105263157894e-05,
|
| 277 |
+
"loss": 0.5024,
|
| 278 |
+
"step": 625
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.46,
|
| 282 |
+
"learning_rate": 2.1347368421052632e-05,
|
| 283 |
+
"loss": 0.471,
|
| 284 |
+
"step": 650
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.46,
|
| 288 |
+
"eval_loss": 1.3961360454559326,
|
| 289 |
+
"eval_runtime": 818.7087,
|
| 290 |
+
"eval_sacrebleu": 1885.001364213634,
|
| 291 |
+
"eval_samples_per_second": 4.15,
|
| 292 |
+
"eval_steps_per_second": 2.075,
|
| 293 |
+
"eval_wer": 82.26640965811099,
|
| 294 |
+
"step": 650
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.52,
|
| 298 |
+
"learning_rate": 2.095263157894737e-05,
|
| 299 |
+
"loss": 0.4683,
|
| 300 |
+
"step": 675
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.58,
|
| 304 |
+
"learning_rate": 2.0557894736842105e-05,
|
| 305 |
+
"loss": 0.4412,
|
| 306 |
+
"step": 700
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.58,
|
| 310 |
+
"eval_loss": 1.3986111879348755,
|
| 311 |
+
"eval_runtime": 764.1712,
|
| 312 |
+
"eval_sacrebleu": 2145.9785743484185,
|
| 313 |
+
"eval_samples_per_second": 4.447,
|
| 314 |
+
"eval_steps_per_second": 2.223,
|
| 315 |
+
"eval_wer": 72.04693787832674,
|
| 316 |
+
"step": 700
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.63,
|
| 320 |
+
"learning_rate": 2.0163157894736843e-05,
|
| 321 |
+
"loss": 0.4794,
|
| 322 |
+
"step": 725
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.69,
|
| 326 |
+
"learning_rate": 1.9768421052631577e-05,
|
| 327 |
+
"loss": 0.4625,
|
| 328 |
+
"step": 750
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.69,
|
| 332 |
+
"eval_loss": 1.3885091543197632,
|
| 333 |
+
"eval_runtime": 843.5439,
|
| 334 |
+
"eval_sacrebleu": 1837.7812162939947,
|
| 335 |
+
"eval_samples_per_second": 4.028,
|
| 336 |
+
"eval_steps_per_second": 2.014,
|
| 337 |
+
"eval_wer": 87.4471610699672,
|
| 338 |
+
"step": 750
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.74,
|
| 342 |
+
"learning_rate": 1.937368421052632e-05,
|
| 343 |
+
"loss": 0.4706,
|
| 344 |
+
"step": 775
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.8,
|
| 348 |
+
"learning_rate": 1.8978947368421054e-05,
|
| 349 |
+
"loss": 0.4195,
|
| 350 |
+
"step": 800
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.8,
|
| 354 |
+
"eval_loss": 1.4094804525375366,
|
| 355 |
+
"eval_runtime": 811.6593,
|
| 356 |
+
"eval_sacrebleu": 1909.2654907091217,
|
| 357 |
+
"eval_samples_per_second": 4.186,
|
| 358 |
+
"eval_steps_per_second": 2.093,
|
| 359 |
+
"eval_wer": 78.69196171925196,
|
| 360 |
+
"step": 800
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.86,
|
| 364 |
+
"learning_rate": 1.858421052631579e-05,
|
| 365 |
+
"loss": 0.4444,
|
| 366 |
+
"step": 825
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.91,
|
| 370 |
+
"learning_rate": 1.8189473684210527e-05,
|
| 371 |
+
"loss": 0.4532,
|
| 372 |
+
"step": 850
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.91,
|
| 376 |
+
"eval_loss": 1.3891242742538452,
|
| 377 |
+
"eval_runtime": 809.0214,
|
| 378 |
+
"eval_sacrebleu": 1925.2237669771318,
|
| 379 |
+
"eval_samples_per_second": 4.2,
|
| 380 |
+
"eval_steps_per_second": 2.1,
|
| 381 |
+
"eval_wer": 82.01616448547564,
|
| 382 |
+
"step": 850
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.97,
|
| 386 |
+
"learning_rate": 1.7794736842105265e-05,
|
| 387 |
+
"loss": 0.4294,
|
| 388 |
+
"step": 875
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 2.03,
|
| 392 |
+
"learning_rate": 1.74e-05,
|
| 393 |
+
"loss": 0.3201,
|
| 394 |
+
"step": 900
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.03,
|
| 398 |
+
"eval_loss": 1.4414749145507812,
|
| 399 |
+
"eval_runtime": 822.9952,
|
| 400 |
+
"eval_sacrebleu": 1919.20196297517,
|
| 401 |
+
"eval_samples_per_second": 4.129,
|
| 402 |
+
"eval_steps_per_second": 2.064,
|
| 403 |
+
"eval_wer": 80.44367792769943,
|
| 404 |
+
"step": 900
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.08,
|
| 408 |
+
"learning_rate": 1.7005263157894738e-05,
|
| 409 |
+
"loss": 0.1981,
|
| 410 |
+
"step": 925
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.14,
|
| 414 |
+
"learning_rate": 1.6610526315789472e-05,
|
| 415 |
+
"loss": 0.1955,
|
| 416 |
+
"step": 950
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 2.14,
|
| 420 |
+
"eval_loss": 1.4410473108291626,
|
| 421 |
+
"eval_runtime": 941.3392,
|
| 422 |
+
"eval_sacrebleu": 1540.5045548236765,
|
| 423 |
+
"eval_samples_per_second": 3.61,
|
| 424 |
+
"eval_steps_per_second": 1.805,
|
| 425 |
+
"eval_wer": 101.01450745662981,
|
| 426 |
+
"step": 950
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 2.19,
|
| 430 |
+
"learning_rate": 1.621578947368421e-05,
|
| 431 |
+
"loss": 0.191,
|
| 432 |
+
"step": 975
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.25,
|
| 436 |
+
"learning_rate": 1.582105263157895e-05,
|
| 437 |
+
"loss": 0.2111,
|
| 438 |
+
"step": 1000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.25,
|
| 442 |
+
"eval_loss": 1.434527039527893,
|
| 443 |
+
"eval_runtime": 854.9114,
|
| 444 |
+
"eval_sacrebleu": 1735.9647924357162,
|
| 445 |
+
"eval_samples_per_second": 3.975,
|
| 446 |
+
"eval_steps_per_second": 1.987,
|
| 447 |
+
"eval_wer": 90.92692164620743,
|
| 448 |
+
"step": 1000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 2.31,
|
| 452 |
+
"learning_rate": 1.5426315789473683e-05,
|
| 453 |
+
"loss": 0.2043,
|
| 454 |
+
"step": 1025
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 2.36,
|
| 458 |
+
"learning_rate": 1.5031578947368423e-05,
|
| 459 |
+
"loss": 0.1981,
|
| 460 |
+
"step": 1050
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.36,
|
| 464 |
+
"eval_loss": 1.4597235918045044,
|
| 465 |
+
"eval_runtime": 878.1026,
|
| 466 |
+
"eval_sacrebleu": 1730.324971075036,
|
| 467 |
+
"eval_samples_per_second": 3.87,
|
| 468 |
+
"eval_steps_per_second": 1.935,
|
| 469 |
+
"eval_wer": 91.53562612018533,
|
| 470 |
+
"step": 1050
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 2.42,
|
| 474 |
+
"learning_rate": 1.4636842105263158e-05,
|
| 475 |
+
"loss": 0.1862,
|
| 476 |
+
"step": 1075
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 2.48,
|
| 480 |
+
"learning_rate": 1.4242105263157896e-05,
|
| 481 |
+
"loss": 0.2052,
|
| 482 |
+
"step": 1100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.48,
|
| 486 |
+
"eval_loss": 1.443906307220459,
|
| 487 |
+
"eval_runtime": 758.1394,
|
| 488 |
+
"eval_sacrebleu": 2143.363030526311,
|
| 489 |
+
"eval_samples_per_second": 4.482,
|
| 490 |
+
"eval_steps_per_second": 2.241,
|
| 491 |
+
"eval_wer": 72.49332115924385,
|
| 492 |
+
"step": 1100
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.53,
|
| 496 |
+
"learning_rate": 1.384736842105263e-05,
|
| 497 |
+
"loss": 0.2018,
|
| 498 |
+
"step": 1125
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.59,
|
| 502 |
+
"learning_rate": 1.3452631578947369e-05,
|
| 503 |
+
"loss": 0.1886,
|
| 504 |
+
"step": 1150
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.59,
|
| 508 |
+
"eval_loss": 1.4702341556549072,
|
| 509 |
+
"eval_runtime": 802.8723,
|
| 510 |
+
"eval_sacrebleu": 1965.5005446719592,
|
| 511 |
+
"eval_samples_per_second": 4.232,
|
| 512 |
+
"eval_steps_per_second": 2.116,
|
| 513 |
+
"eval_wer": 77.75185147610834,
|
| 514 |
+
"step": 1150
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 2.64,
|
| 518 |
+
"learning_rate": 1.3057894736842105e-05,
|
| 519 |
+
"loss": 0.1874,
|
| 520 |
+
"step": 1175
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.7,
|
| 524 |
+
"learning_rate": 1.2663157894736843e-05,
|
| 525 |
+
"loss": 0.1918,
|
| 526 |
+
"step": 1200
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.7,
|
| 530 |
+
"eval_loss": 1.4518498182296753,
|
| 531 |
+
"eval_runtime": 782.9672,
|
| 532 |
+
"eval_sacrebleu": 2057.451707817952,
|
| 533 |
+
"eval_samples_per_second": 4.34,
|
| 534 |
+
"eval_steps_per_second": 2.17,
|
| 535 |
+
"eval_wer": 75.49288153934597,
|
| 536 |
+
"step": 1200
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.76,
|
| 540 |
+
"learning_rate": 1.2268421052631578e-05,
|
| 541 |
+
"loss": 0.1819,
|
| 542 |
+
"step": 1225
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.81,
|
| 546 |
+
"learning_rate": 1.1873684210526316e-05,
|
| 547 |
+
"loss": 0.1755,
|
| 548 |
+
"step": 1250
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.81,
|
| 552 |
+
"eval_loss": 1.4787976741790771,
|
| 553 |
+
"eval_runtime": 797.1285,
|
| 554 |
+
"eval_sacrebleu": 1954.2236766241735,
|
| 555 |
+
"eval_samples_per_second": 4.263,
|
| 556 |
+
"eval_steps_per_second": 2.131,
|
| 557 |
+
"eval_wer": 78.29968550268845,
|
| 558 |
+
"step": 1250
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 2.87,
|
| 562 |
+
"learning_rate": 1.1478947368421053e-05,
|
| 563 |
+
"loss": 0.172,
|
| 564 |
+
"step": 1275
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.93,
|
| 568 |
+
"learning_rate": 1.108421052631579e-05,
|
| 569 |
+
"loss": 0.1769,
|
| 570 |
+
"step": 1300
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.93,
|
| 574 |
+
"eval_loss": 1.4588199853897095,
|
| 575 |
+
"eval_runtime": 886.5481,
|
| 576 |
+
"eval_sacrebleu": 1774.1463556159297,
|
| 577 |
+
"eval_samples_per_second": 3.833,
|
| 578 |
+
"eval_steps_per_second": 1.916,
|
| 579 |
+
"eval_wer": 91.92790233674884,
|
| 580 |
+
"step": 1300
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.98,
|
| 584 |
+
"learning_rate": 1.0689473684210526e-05,
|
| 585 |
+
"loss": 0.1893,
|
| 586 |
+
"step": 1325
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.04,
|
| 590 |
+
"learning_rate": 1.0294736842105264e-05,
|
| 591 |
+
"loss": 0.1104,
|
| 592 |
+
"step": 1350
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 3.04,
|
| 596 |
+
"eval_loss": 1.5280550718307495,
|
| 597 |
+
"eval_runtime": 860.6622,
|
| 598 |
+
"eval_sacrebleu": 1838.1998881180943,
|
| 599 |
+
"eval_samples_per_second": 3.948,
|
| 600 |
+
"eval_steps_per_second": 1.974,
|
| 601 |
+
"eval_wer": 84.73166277772141,
|
| 602 |
+
"step": 1350
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 3.1,
|
| 606 |
+
"learning_rate": 9.9e-06,
|
| 607 |
+
"loss": 0.069,
|
| 608 |
+
"step": 1375
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.15,
|
| 612 |
+
"learning_rate": 9.505263157894737e-06,
|
| 613 |
+
"loss": 0.0718,
|
| 614 |
+
"step": 1400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.15,
|
| 618 |
+
"eval_loss": 1.5133227109909058,
|
| 619 |
+
"eval_runtime": 790.7595,
|
| 620 |
+
"eval_sacrebleu": 2058.095479486138,
|
| 621 |
+
"eval_samples_per_second": 4.297,
|
| 622 |
+
"eval_steps_per_second": 2.149,
|
| 623 |
+
"eval_wer": 76.03057049135978,
|
| 624 |
+
"step": 1400
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.21,
|
| 628 |
+
"learning_rate": 9.110526315789473e-06,
|
| 629 |
+
"loss": 0.0735,
|
| 630 |
+
"step": 1425
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 3.26,
|
| 634 |
+
"learning_rate": 8.715789473684211e-06,
|
| 635 |
+
"loss": 0.0855,
|
| 636 |
+
"step": 1450
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 3.26,
|
| 640 |
+
"eval_loss": 1.527108907699585,
|
| 641 |
+
"eval_runtime": 856.1919,
|
| 642 |
+
"eval_sacrebleu": 1720.107239864863,
|
| 643 |
+
"eval_samples_per_second": 3.969,
|
| 644 |
+
"eval_steps_per_second": 1.984,
|
| 645 |
+
"eval_wer": 89.13462513949477,
|
| 646 |
+
"step": 1450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 3.32,
|
| 650 |
+
"learning_rate": 8.321052631578947e-06,
|
| 651 |
+
"loss": 0.0705,
|
| 652 |
+
"step": 1475
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 3.38,
|
| 656 |
+
"learning_rate": 7.926315789473684e-06,
|
| 657 |
+
"loss": 0.0717,
|
| 658 |
+
"step": 1500
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 3.38,
|
| 662 |
+
"eval_loss": 1.5289273262023926,
|
| 663 |
+
"eval_runtime": 783.842,
|
| 664 |
+
"eval_sacrebleu": 2007.516276406263,
|
| 665 |
+
"eval_samples_per_second": 4.335,
|
| 666 |
+
"eval_steps_per_second": 2.168,
|
| 667 |
+
"eval_wer": 75.9291197456968,
|
| 668 |
+
"step": 1500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.43,
|
| 672 |
+
"learning_rate": 7.531578947368421e-06,
|
| 673 |
+
"loss": 0.0728,
|
| 674 |
+
"step": 1525
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 3.49,
|
| 678 |
+
"learning_rate": 7.1368421052631585e-06,
|
| 679 |
+
"loss": 0.0707,
|
| 680 |
+
"step": 1550
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 3.49,
|
| 684 |
+
"eval_loss": 1.5366449356079102,
|
| 685 |
+
"eval_runtime": 778.1169,
|
| 686 |
+
"eval_sacrebleu": 2149.64780863619,
|
| 687 |
+
"eval_samples_per_second": 4.367,
|
| 688 |
+
"eval_steps_per_second": 2.183,
|
| 689 |
+
"eval_wer": 71.95225051570796,
|
| 690 |
+
"step": 1550
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 3.55,
|
| 694 |
+
"learning_rate": 6.742105263157895e-06,
|
| 695 |
+
"loss": 0.0715,
|
| 696 |
+
"step": 1575
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 3.6,
|
| 700 |
+
"learning_rate": 6.347368421052632e-06,
|
| 701 |
+
"loss": 0.0704,
|
| 702 |
+
"step": 1600
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 3.6,
|
| 706 |
+
"eval_loss": 1.5355422496795654,
|
| 707 |
+
"eval_runtime": 742.6818,
|
| 708 |
+
"eval_sacrebleu": 2179.5147468855334,
|
| 709 |
+
"eval_samples_per_second": 4.575,
|
| 710 |
+
"eval_steps_per_second": 2.288,
|
| 711 |
+
"eval_wer": 69.87589192113896,
|
| 712 |
+
"step": 1600
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 3.66,
|
| 716 |
+
"learning_rate": 5.952631578947369e-06,
|
| 717 |
+
"loss": 0.0667,
|
| 718 |
+
"step": 1625
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 3.71,
|
| 722 |
+
"learning_rate": 5.557894736842106e-06,
|
| 723 |
+
"loss": 0.0676,
|
| 724 |
+
"step": 1650
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 3.71,
|
| 728 |
+
"eval_loss": 1.5392969846725464,
|
| 729 |
+
"eval_runtime": 767.4869,
|
| 730 |
+
"eval_sacrebleu": 2086.2196870181765,
|
| 731 |
+
"eval_samples_per_second": 4.427,
|
| 732 |
+
"eval_steps_per_second": 2.214,
|
| 733 |
+
"eval_wer": 73.24743836867201,
|
| 734 |
+
"step": 1650
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.77,
|
| 738 |
+
"learning_rate": 5.163157894736842e-06,
|
| 739 |
+
"loss": 0.0725,
|
| 740 |
+
"step": 1675
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 3.83,
|
| 744 |
+
"learning_rate": 4.76842105263158e-06,
|
| 745 |
+
"loss": 0.0748,
|
| 746 |
+
"step": 1700
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 3.83,
|
| 750 |
+
"eval_loss": 1.5397995710372925,
|
| 751 |
+
"eval_runtime": 800.0285,
|
| 752 |
+
"eval_sacrebleu": 1879.1609534484007,
|
| 753 |
+
"eval_samples_per_second": 4.247,
|
| 754 |
+
"eval_steps_per_second": 2.124,
|
| 755 |
+
"eval_wer": 80.72774001555578,
|
| 756 |
+
"step": 1700
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 3.88,
|
| 760 |
+
"learning_rate": 4.373684210526316e-06,
|
| 761 |
+
"loss": 0.0761,
|
| 762 |
+
"step": 1725
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 3.94,
|
| 766 |
+
"learning_rate": 3.978947368421053e-06,
|
| 767 |
+
"loss": 0.0695,
|
| 768 |
+
"step": 1750
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.94,
|
| 772 |
+
"eval_loss": 1.5350741147994995,
|
| 773 |
+
"eval_runtime": 802.9873,
|
| 774 |
+
"eval_sacrebleu": 2001.847643597036,
|
| 775 |
+
"eval_samples_per_second": 4.232,
|
| 776 |
+
"eval_steps_per_second": 2.116,
|
| 777 |
+
"eval_wer": 78.83061107165804,
|
| 778 |
+
"step": 1750
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 4.0,
|
| 782 |
+
"learning_rate": 3.5842105263157898e-06,
|
| 783 |
+
"loss": 0.0718,
|
| 784 |
+
"step": 1775
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 4.05,
|
| 788 |
+
"learning_rate": 3.1894736842105266e-06,
|
| 789 |
+
"loss": 0.033,
|
| 790 |
+
"step": 1800
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 4.05,
|
| 794 |
+
"eval_loss": 1.5806833505630493,
|
| 795 |
+
"eval_runtime": 818.7159,
|
| 796 |
+
"eval_sacrebleu": 1892.0435299189799,
|
| 797 |
+
"eval_samples_per_second": 4.15,
|
| 798 |
+
"eval_steps_per_second": 2.075,
|
| 799 |
+
"eval_wer": 82.26302796658888,
|
| 800 |
+
"step": 1800
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 4.11,
|
| 804 |
+
"learning_rate": 2.7947368421052635e-06,
|
| 805 |
+
"loss": 0.0297,
|
| 806 |
+
"step": 1825
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 4.16,
|
| 810 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 811 |
+
"loss": 0.0317,
|
| 812 |
+
"step": 1850
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 4.16,
|
| 816 |
+
"eval_loss": 1.5843309164047241,
|
| 817 |
+
"eval_runtime": 799.3661,
|
| 818 |
+
"eval_sacrebleu": 1967.1172131517576,
|
| 819 |
+
"eval_samples_per_second": 4.251,
|
| 820 |
+
"eval_steps_per_second": 2.125,
|
| 821 |
+
"eval_wer": 78.77650400730445,
|
| 822 |
+
"step": 1850
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 4.22,
|
| 826 |
+
"learning_rate": 2.005263157894737e-06,
|
| 827 |
+
"loss": 0.0306,
|
| 828 |
+
"step": 1875
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 4.28,
|
| 832 |
+
"learning_rate": 1.6105263157894738e-06,
|
| 833 |
+
"loss": 0.0302,
|
| 834 |
+
"step": 1900
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 4.28,
|
| 838 |
+
"eval_loss": 1.5848406553268433,
|
| 839 |
+
"eval_runtime": 792.2648,
|
| 840 |
+
"eval_sacrebleu": 1969.6752766606157,
|
| 841 |
+
"eval_samples_per_second": 4.289,
|
| 842 |
+
"eval_steps_per_second": 2.144,
|
| 843 |
+
"eval_wer": 79.12481823408069,
|
| 844 |
+
"step": 1900
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 4.33,
|
| 848 |
+
"learning_rate": 1.2157894736842105e-06,
|
| 849 |
+
"loss": 0.027,
|
| 850 |
+
"step": 1925
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 4.39,
|
| 854 |
+
"learning_rate": 8.210526315789474e-07,
|
| 855 |
+
"loss": 0.0337,
|
| 856 |
+
"step": 1950
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 4.39,
|
| 860 |
+
"eval_loss": 1.5807627439498901,
|
| 861 |
+
"eval_runtime": 771.2125,
|
| 862 |
+
"eval_sacrebleu": 2062.9545573784067,
|
| 863 |
+
"eval_samples_per_second": 4.406,
|
| 864 |
+
"eval_steps_per_second": 2.203,
|
| 865 |
+
"eval_wer": 74.15373169659463,
|
| 866 |
+
"step": 1950
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 4.45,
|
| 870 |
+
"learning_rate": 4.263157894736842e-07,
|
| 871 |
+
"loss": 0.0308,
|
| 872 |
+
"step": 1975
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 4.5,
|
| 876 |
+
"learning_rate": 3.157894736842105e-08,
|
| 877 |
+
"loss": 0.0306,
|
| 878 |
+
"step": 2000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 4.5,
|
| 882 |
+
"eval_loss": 1.5844563245773315,
|
| 883 |
+
"eval_runtime": 771.453,
|
| 884 |
+
"eval_sacrebleu": 2090.6715503549444,
|
| 885 |
+
"eval_samples_per_second": 4.405,
|
| 886 |
+
"eval_steps_per_second": 2.202,
|
| 887 |
+
"eval_wer": 73.00057488755876,
|
| 888 |
+
"step": 2000
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 4.5,
|
| 892 |
+
"step": 2000,
|
| 893 |
+
"total_flos": 1.846946562048e+19,
|
| 894 |
+
"train_loss": 0.524174674898386,
|
| 895 |
+
"train_runtime": 70623.6184,
|
| 896 |
+
"train_samples_per_second": 0.906,
|
| 897 |
+
"train_steps_per_second": 0.028
|
| 898 |
+
}
|
| 899 |
+
],
|
| 900 |
+
"max_steps": 2000,
|
| 901 |
+
"num_train_epochs": 5,
|
| 902 |
+
"total_flos": 1.846946562048e+19,
|
| 903 |
+
"trial_name": null,
|
| 904 |
+
"trial_params": null
|
| 905 |
+
}
|