imdatta0's picture
End of training
8b1b653 verified
raw
history blame
1.41 kB
{
"best_metric": 0.17445562779903412,
"best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_metamath_ortho/checkpoint-13",
"epoch": 0.021061158363710003,
"eval_steps": 13,
"global_step": 13,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0016200891049007696,
"grad_norm": 2.756908416748047,
"learning_rate": 2.3076923076923076e-05,
"loss": 0.2557,
"step": 1
},
{
"epoch": 0.011340623734305387,
"grad_norm": 1.0947959423065186,
"learning_rate": 0.00016153846153846153,
"loss": 0.1901,
"step": 7
},
{
"epoch": 0.021061158363710003,
"eval_loss": 0.17445562779903412,
"eval_runtime": 12.3542,
"eval_samples_per_second": 40.472,
"eval_steps_per_second": 5.099,
"step": 13
}
],
"logging_steps": 7,
"max_steps": 617,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 13,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.5689170152456192e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}