ZIN_V3 / checkpoint-1164 /trainer_state.json
EYEDOL's picture
Upload folder using huggingface_hub
75d57dc verified
{
"best_global_step": 800,
"best_metric": 7.272170821856534,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 400,
"global_step": 1164,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.34393809114359414,
"grad_norm": 4.169175624847412,
"learning_rate": 9.950000000000001e-06,
"loss": 0.1094,
"step": 200
},
{
"epoch": 0.6878761822871883,
"grad_norm": 3.851189613342285,
"learning_rate": 7.946058091286308e-06,
"loss": 0.1075,
"step": 400
},
{
"epoch": 0.6878761822871883,
"eval_loss": 0.10680446028709412,
"eval_runtime": 3328.993,
"eval_samples_per_second": 1.397,
"eval_steps_per_second": 0.175,
"eval_wer": 8.390423052475457,
"step": 400
},
{
"epoch": 1.0309544282029235,
"grad_norm": 2.2491328716278076,
"learning_rate": 5.87136929460581e-06,
"loss": 0.0919,
"step": 600
},
{
"epoch": 1.3748925193465176,
"grad_norm": 1.6838488578796387,
"learning_rate": 3.7966804979253114e-06,
"loss": 0.0335,
"step": 800
},
{
"epoch": 1.3748925193465176,
"eval_loss": 0.10098689794540405,
"eval_runtime": 2901.1453,
"eval_samples_per_second": 1.603,
"eval_steps_per_second": 0.201,
"eval_wer": 7.272170821856534,
"step": 800
},
{
"epoch": 1.7188306104901119,
"grad_norm": 3.0112545490264893,
"learning_rate": 1.7219917012448133e-06,
"loss": 0.0339,
"step": 1000
}
],
"logging_steps": 200,
"max_steps": 1164,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 1200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.073364537950208e+19,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}