ZIN_N_V1 / checkpoint-1164 /trainer_state.json
EYEDOL's picture
Upload folder using huggingface_hub
481012b verified
{
"best_global_step": 1000,
"best_metric": 48.43466157076462,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"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": 8.694071769714355,
"learning_rate": 9.800000000000001e-06,
"loss": 1.7867,
"step": 200
},
{
"epoch": 0.6878761822871883,
"grad_norm": 6.00631046295166,
"learning_rate": 7.966804979253112e-06,
"loss": 0.6235,
"step": 400
},
{
"epoch": 0.8598452278589854,
"eval_loss": 0.49836987257003784,
"eval_runtime": 2179.0562,
"eval_samples_per_second": 2.134,
"eval_steps_per_second": 0.267,
"eval_wer": 50.81982683369901,
"step": 500
},
{
"epoch": 1.0309544282029235,
"grad_norm": 4.783409118652344,
"learning_rate": 5.892116182572615e-06,
"loss": 0.4852,
"step": 600
},
{
"epoch": 1.3748925193465176,
"grad_norm": 4.614686012268066,
"learning_rate": 3.817427385892116e-06,
"loss": 0.353,
"step": 800
},
{
"epoch": 1.7188306104901119,
"grad_norm": 4.383596897125244,
"learning_rate": 1.7427385892116182e-06,
"loss": 0.3364,
"step": 1000
},
{
"epoch": 1.7188306104901119,
"eval_loss": 0.43098220229148865,
"eval_runtime": 2160.7143,
"eval_samples_per_second": 2.152,
"eval_steps_per_second": 0.269,
"eval_wer": 48.43466157076462,
"step": 1000
}
],
"logging_steps": 200,
"max_steps": 1164,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 1500,
"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
}