whisper-tiny-en / checkpoint-1000 /trainer_state.json
lfurman's picture
Upload folder using huggingface_hub
02625b8 verified
raw
history blame
7.78 kB
{
"best_metric": 97.89625360230548,
"best_model_checkpoint": "./whisper-tiny-en/checkpoint-1000",
"epoch": 24.390243902439025,
"eval_steps": 1000,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.6097560975609756,
"grad_norm": 76.21949768066406,
"learning_rate": 4.2000000000000006e-07,
"loss": 7.5702,
"step": 25
},
{
"epoch": 1.2195121951219512,
"grad_norm": 72.27240753173828,
"learning_rate": 9.200000000000001e-07,
"loss": 7.1361,
"step": 50
},
{
"epoch": 1.8292682926829267,
"grad_norm": 27.645973205566406,
"learning_rate": 1.42e-06,
"loss": 6.4445,
"step": 75
},
{
"epoch": 2.4390243902439024,
"grad_norm": 21.712514877319336,
"learning_rate": 1.9200000000000003e-06,
"loss": 5.8148,
"step": 100
},
{
"epoch": 3.048780487804878,
"grad_norm": 19.066011428833008,
"learning_rate": 2.42e-06,
"loss": 5.1377,
"step": 125
},
{
"epoch": 3.658536585365854,
"grad_norm": 14.726506233215332,
"learning_rate": 2.92e-06,
"loss": 4.6,
"step": 150
},
{
"epoch": 4.2682926829268295,
"grad_norm": 13.597981452941895,
"learning_rate": 3.4200000000000007e-06,
"loss": 4.3595,
"step": 175
},
{
"epoch": 4.878048780487805,
"grad_norm": 16.399568557739258,
"learning_rate": 3.920000000000001e-06,
"loss": 4.171,
"step": 200
},
{
"epoch": 5.487804878048781,
"grad_norm": 15.264267921447754,
"learning_rate": 4.42e-06,
"loss": 3.9727,
"step": 225
},
{
"epoch": 6.097560975609756,
"grad_norm": 14.83893871307373,
"learning_rate": 4.92e-06,
"loss": 3.816,
"step": 250
},
{
"epoch": 6.7073170731707314,
"grad_norm": 14.460186004638672,
"learning_rate": 5.420000000000001e-06,
"loss": 3.6683,
"step": 275
},
{
"epoch": 7.317073170731708,
"grad_norm": 13.663064002990723,
"learning_rate": 5.92e-06,
"loss": 3.5392,
"step": 300
},
{
"epoch": 7.926829268292683,
"grad_norm": 14.359073638916016,
"learning_rate": 6.42e-06,
"loss": 3.4807,
"step": 325
},
{
"epoch": 8.536585365853659,
"grad_norm": 12.90373706817627,
"learning_rate": 6.92e-06,
"loss": 3.335,
"step": 350
},
{
"epoch": 9.146341463414634,
"grad_norm": 15.809270858764648,
"learning_rate": 7.420000000000001e-06,
"loss": 3.2196,
"step": 375
},
{
"epoch": 9.75609756097561,
"grad_norm": 14.483990669250488,
"learning_rate": 7.92e-06,
"loss": 3.1224,
"step": 400
},
{
"epoch": 10.365853658536585,
"grad_norm": 15.841504096984863,
"learning_rate": 8.42e-06,
"loss": 2.9858,
"step": 425
},
{
"epoch": 10.975609756097562,
"grad_norm": 16.420425415039062,
"learning_rate": 8.920000000000001e-06,
"loss": 2.9395,
"step": 450
},
{
"epoch": 11.585365853658537,
"grad_norm": 14.952699661254883,
"learning_rate": 9.42e-06,
"loss": 2.7501,
"step": 475
},
{
"epoch": 12.195121951219512,
"grad_norm": 15.765605926513672,
"learning_rate": 9.920000000000002e-06,
"loss": 2.7014,
"step": 500
},
{
"epoch": 12.804878048780488,
"grad_norm": 17.627931594848633,
"learning_rate": 9.967692307692308e-06,
"loss": 2.601,
"step": 525
},
{
"epoch": 13.414634146341463,
"grad_norm": 16.36449432373047,
"learning_rate": 9.92923076923077e-06,
"loss": 2.4699,
"step": 550
},
{
"epoch": 14.024390243902438,
"grad_norm": 15.8140869140625,
"learning_rate": 9.890769230769231e-06,
"loss": 2.3859,
"step": 575
},
{
"epoch": 14.634146341463415,
"grad_norm": 15.983495712280273,
"learning_rate": 9.852307692307693e-06,
"loss": 2.2235,
"step": 600
},
{
"epoch": 15.24390243902439,
"grad_norm": 15.25550651550293,
"learning_rate": 9.813846153846155e-06,
"loss": 2.1583,
"step": 625
},
{
"epoch": 15.853658536585366,
"grad_norm": 18.167879104614258,
"learning_rate": 9.775384615384616e-06,
"loss": 2.0671,
"step": 650
},
{
"epoch": 16.463414634146343,
"grad_norm": 16.86029624938965,
"learning_rate": 9.736923076923078e-06,
"loss": 1.9547,
"step": 675
},
{
"epoch": 17.073170731707318,
"grad_norm": 16.195270538330078,
"learning_rate": 9.698461538461539e-06,
"loss": 1.8877,
"step": 700
},
{
"epoch": 17.682926829268293,
"grad_norm": 17.331113815307617,
"learning_rate": 9.66e-06,
"loss": 1.7336,
"step": 725
},
{
"epoch": 18.29268292682927,
"grad_norm": 18.00438690185547,
"learning_rate": 9.621538461538463e-06,
"loss": 1.6845,
"step": 750
},
{
"epoch": 18.902439024390244,
"grad_norm": 17.1490478515625,
"learning_rate": 9.583076923076923e-06,
"loss": 1.6027,
"step": 775
},
{
"epoch": 19.51219512195122,
"grad_norm": 17.924156188964844,
"learning_rate": 9.544615384615385e-06,
"loss": 1.4647,
"step": 800
},
{
"epoch": 20.121951219512194,
"grad_norm": 17.920028686523438,
"learning_rate": 9.506153846153848e-06,
"loss": 1.4291,
"step": 825
},
{
"epoch": 20.73170731707317,
"grad_norm": 17.502954483032227,
"learning_rate": 9.467692307692308e-06,
"loss": 1.3113,
"step": 850
},
{
"epoch": 21.341463414634145,
"grad_norm": 18.125986099243164,
"learning_rate": 9.42923076923077e-06,
"loss": 1.232,
"step": 875
},
{
"epoch": 21.951219512195124,
"grad_norm": 16.026338577270508,
"learning_rate": 9.39076923076923e-06,
"loss": 1.1903,
"step": 900
},
{
"epoch": 22.5609756097561,
"grad_norm": 16.795467376708984,
"learning_rate": 9.352307692307693e-06,
"loss": 1.0742,
"step": 925
},
{
"epoch": 23.170731707317074,
"grad_norm": 19.86144256591797,
"learning_rate": 9.313846153846155e-06,
"loss": 1.0346,
"step": 950
},
{
"epoch": 23.78048780487805,
"grad_norm": 17.94236183166504,
"learning_rate": 9.275384615384616e-06,
"loss": 0.9589,
"step": 975
},
{
"epoch": 24.390243902439025,
"grad_norm": 17.72712516784668,
"learning_rate": 9.236923076923078e-06,
"loss": 0.8757,
"step": 1000
},
{
"epoch": 24.390243902439025,
"eval_loss": 4.123460292816162,
"eval_runtime": 64.6053,
"eval_samples_per_second": 2.492,
"eval_steps_per_second": 0.325,
"eval_wer": 97.89625360230548,
"step": 1000
}
],
"logging_steps": 25,
"max_steps": 7000,
"num_input_tokens_seen": 0,
"num_train_epochs": 171,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.8622106976256e+17,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}