pipeline_test / checkpoint-100 /trainer_state.json
beamaia's picture
Training in progress, step 100, checkpoint
f98097a verified
{
"best_metric": 0.5742923021316528,
"best_model_checkpoint": "./llama3/30-08-24-Weni-Pipeline_test_Experiment with SFT and Llama3 70b-2_max_steps-1362_batch_8_2024-08-30/checkpoint-100",
"epoch": 0.44004400440044006,
"eval_steps": 100,
"global_step": 100,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04400440044004401,
"grad_norm": 0.5568628907203674,
"learning_rate": 7.5e-05,
"loss": 2.0875,
"step": 10
},
{
"epoch": 0.08800880088008801,
"grad_norm": 0.2537558972835541,
"learning_rate": 0.00015,
"loss": 0.9378,
"step": 20
},
{
"epoch": 0.132013201320132,
"grad_norm": 0.24558919668197632,
"learning_rate": 0.000225,
"loss": 0.7,
"step": 30
},
{
"epoch": 0.17601760176017603,
"grad_norm": 0.13937097787857056,
"learning_rate": 0.0003,
"loss": 0.6298,
"step": 40
},
{
"epoch": 0.22002200220022003,
"grad_norm": 0.1871194988489151,
"learning_rate": 0.00029995764763563235,
"loss": 0.6321,
"step": 50
},
{
"epoch": 0.264026402640264,
"grad_norm": 0.14626263082027435,
"learning_rate": 0.00029983061445883305,
"loss": 0.6403,
"step": 60
},
{
"epoch": 0.30803080308030806,
"grad_norm": 0.12049665302038193,
"learning_rate": 0.0002996189722050073,
"loss": 0.5998,
"step": 70
},
{
"epoch": 0.35203520352035206,
"grad_norm": 0.13617923855781555,
"learning_rate": 0.0002993228403881531,
"loss": 0.5942,
"step": 80
},
{
"epoch": 0.39603960396039606,
"grad_norm": 0.1271793246269226,
"learning_rate": 0.00029894238623337174,
"loss": 0.5647,
"step": 90
},
{
"epoch": 0.44004400440044006,
"grad_norm": 0.18757876753807068,
"learning_rate": 0.00029847782458243663,
"loss": 0.5619,
"step": 100
},
{
"epoch": 0.44004400440044006,
"eval_accuracy": 1.0,
"eval_f1": 1.0,
"eval_f1_macro": 1.0,
"eval_f1_micro": 1.0,
"eval_loss": 0.5742923021316528,
"eval_precision": 1.0,
"eval_precision_macro": 1.0,
"eval_precision_micro": 1.0,
"eval_recall": 1.0,
"eval_recall_macro": 1.0,
"eval_recall_micro": 1.0,
"eval_runtime": 90.5857,
"eval_samples_per_second": 4.46,
"eval_steps_per_second": 1.115,
"step": 100
}
],
"logging_steps": 10,
"max_steps": 1362,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.3925298981778227e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}