whisper-clean_4 / checkpoint-200 /trainer_state.json
lyhourt's picture
Training in progress, step 200, checkpoint
5347263 verified
{
"best_metric": 12.677725118483412,
"best_model_checkpoint": "./whisper-small-clean_4/checkpoint-200",
"epoch": 1.0333333333333334,
"eval_steps": 100,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.08333333333333333,
"grad_norm": 33.14436340332031,
"learning_rate": 4.2000000000000004e-06,
"loss": 2.3942,
"step": 25
},
{
"epoch": 0.16666666666666666,
"grad_norm": 2.458979606628418,
"learning_rate": 9.200000000000002e-06,
"loss": 0.2133,
"step": 50
},
{
"epoch": 0.25,
"grad_norm": 1.5336248874664307,
"learning_rate": 9.16e-06,
"loss": 0.0856,
"step": 75
},
{
"epoch": 0.3333333333333333,
"grad_norm": 1.983283519744873,
"learning_rate": 8.16e-06,
"loss": 0.0466,
"step": 100
},
{
"epoch": 0.3333333333333333,
"eval_loss": 0.11704772710800171,
"eval_runtime": 626.1564,
"eval_samples_per_second": 2.696,
"eval_steps_per_second": 0.085,
"eval_wer": 15.995260663507107,
"step": 100
},
{
"epoch": 0.4166666666666667,
"grad_norm": 0.32364532351493835,
"learning_rate": 7.16e-06,
"loss": 0.011,
"step": 125
},
{
"epoch": 0.5,
"grad_norm": 65.41436767578125,
"learning_rate": 6.16e-06,
"loss": 0.2072,
"step": 150
},
{
"epoch": 0.5833333333333334,
"grad_norm": 20.50943946838379,
"learning_rate": 5.1600000000000006e-06,
"loss": 1.0934,
"step": 175
},
{
"epoch": 1.0333333333333334,
"grad_norm": 1.004915714263916,
"learning_rate": 4.16e-06,
"loss": 0.2315,
"step": 200
},
{
"epoch": 1.0333333333333334,
"eval_loss": 0.08518517762422562,
"eval_runtime": 600.9232,
"eval_samples_per_second": 2.809,
"eval_steps_per_second": 0.088,
"eval_wer": 12.677725118483412,
"step": 200
}
],
"logging_steps": 25,
"max_steps": 300,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.6837926350848e+18,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}