Ours-RadFM / checkpoint-2 /trainer_state.json
loopback-kr's picture
Upload folder using huggingface_hub
12c4961 verified
{
"best_metric": 0.0,
"best_model_checkpoint": "/workspace/previous_works/RadFM/output/RadFM-Llama3-8B-pretrain-0002-embed_tokens-depth32-lora/checkpoint-2",
"epoch": 0.6666666666666666,
"eval_steps": 2,
"global_step": 2,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.3333333333333333,
"grad_norm": 62.661258697509766,
"learning_rate": 0.0001,
"loss": 2.7344,
"step": 1
},
{
"epoch": 0.6666666666666666,
"grad_norm": 83.97257232666016,
"learning_rate": 9.619397662556435e-05,
"loss": 2.7904,
"step": 2
},
{
"epoch": 0.6666666666666666,
"eval_accuracy": 0.0,
"eval_loss": 2.184687852859497,
"eval_runtime": 6.6014,
"eval_samples_per_second": 1.515,
"eval_steps_per_second": 0.454,
"step": 2
}
],
"logging_steps": 1,
"max_steps": 9,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 2,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 381644150145024.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}