trainer_state.json

{
 "best_metric": null,
 "best_model_checkpoint": null,
 "epoch": 1.0,
 "eval_steps": 125,
 "global_step": 625,
 "is_hyper_param_search": false,
 "is_local_process_zero": true,
 "is_world_process_zero": true,
 "log_history": [
 {
 "epoch": 0.2,
 "grad_norm": 4.343021869659424,
 "learning_rate": 2e-05,
 "loss": 0.7338,
 "step": 125
 },
 {
 "epoch": 0.2,
 "eval_accuracy": 0.6674,
 "eval_loss": 0.6289547681808472,
 "eval_runtime": 887.8897,
 "eval_samples_per_second": 5.631,
 "eval_steps_per_second": 0.704,
 "step": 125
 },
 {
 "epoch": 0.4,
 "grad_norm": 4.281612396240234,
 "learning_rate": 1.5e-05,
 "loss": 0.5543,
 "step": 250
 },
 {
 "epoch": 0.4,
 "eval_accuracy": 0.7028,
 "eval_loss": 0.574894368648529,
 "eval_runtime": 879.5534,
 "eval_samples_per_second": 5.685,
 "eval_steps_per_second": 0.711,
 "step": 250
 },
 {
 "epoch": 0.6,
 "grad_norm": 7.856631755828857,
 "learning_rate": 1e-05,
 "loss": 0.5197,
 "step": 375
 },
 {
 "epoch": 0.6,
 "eval_accuracy": 0.7106,
 "eval_loss": 0.5658385753631592,
 "eval_runtime": 880.3303,
 "eval_samples_per_second": 5.68,
 "eval_steps_per_second": 0.71,
 "step": 375
 },
 {
 "epoch": 0.8,
 "grad_norm": 5.3395795822143555,
 "learning_rate": 5e-06,
 "loss": 0.5136,
 "step": 500
 },
 {
 "epoch": 0.8,
 "eval_accuracy": 0.7146,
 "eval_loss": 0.5568351745605469,
 "eval_runtime": 879.4101,
 "eval_samples_per_second": 5.686,
 "eval_steps_per_second": 0.711,
 "step": 500
 },
 {
 "epoch": 1.0,
 "grad_norm": 7.301570415496826,
 "learning_rate": 0.0,
 "loss": 0.506,
 "step": 625
 },
 {
 "epoch": 1.0,
 "eval_accuracy": 0.7158,
 "eval_loss": 0.5556138753890991,
 "eval_runtime": 877.516,
 "eval_samples_per_second": 5.698,
 "eval_steps_per_second": 0.712,
 "step": 625
 }
 ],
 "logging_steps": 125,
 "max_steps": 625,
 "num_input_tokens_seen": 0,
 "num_train_epochs": 1,
 "save_steps": 125,
 "stateful_callbacks": {
 "TrainerControl": {
 "args": {
 "should_epoch_stop": false,
 "should_evaluate": false,
 "should_log": false,
 "should_save": true,
 "should_training_stop": true
 },
 "attributes": {}
 }
 },
 "total_flos": 0.0,
 "train_batch_size": 12,
 "trial_name": null,
 "trial_params": null
}