mix_full_ret / checkpoint-5000 /trainer_state.json
praveensonu's picture
Upload LoRA adapter from gd_full_retain_model
8cfd458 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.9230769230769231,
"eval_steps": 500,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.19230769230769232,
"grad_norm": 76.94072723388672,
"learning_rate": 9.040384615384617e-06,
"loss": -102.1462,
"step": 500
},
{
"epoch": 0.38461538461538464,
"grad_norm": 531.77978515625,
"learning_rate": 8.078846153846154e-06,
"loss": -151.8301,
"step": 1000
},
{
"epoch": 0.5769230769230769,
"grad_norm": 139.83151245117188,
"learning_rate": 7.117307692307693e-06,
"loss": -155.388,
"step": 1500
},
{
"epoch": 0.7692307692307693,
"grad_norm": 114.4839096069336,
"learning_rate": 6.15576923076923e-06,
"loss": -158.7632,
"step": 2000
},
{
"epoch": 0.9615384615384616,
"grad_norm": 116.69457244873047,
"learning_rate": 5.194230769230769e-06,
"loss": -160.9429,
"step": 2500
},
{
"epoch": 1.1538461538461537,
"grad_norm": 109.9537353515625,
"learning_rate": 4.232692307692308e-06,
"loss": -163.2363,
"step": 3000
},
{
"epoch": 1.3461538461538463,
"grad_norm": 123.96151733398438,
"learning_rate": 3.271153846153847e-06,
"loss": -164.1284,
"step": 3500
},
{
"epoch": 1.5384615384615383,
"grad_norm": 140.77725219726562,
"learning_rate": 2.309615384615385e-06,
"loss": -166.1304,
"step": 4000
},
{
"epoch": 1.7307692307692308,
"grad_norm": 89.94447326660156,
"learning_rate": 1.348076923076923e-06,
"loss": -166.7199,
"step": 4500
},
{
"epoch": 1.9230769230769231,
"grad_norm": 115.57528686523438,
"learning_rate": 3.865384615384616e-07,
"loss": -167.5511,
"step": 5000
}
],
"logging_steps": 500,
"max_steps": 5200,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}