ocr-test-q3b-7k / trainer_state.json
ahmedheakl's picture
End of training
be8ad60 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.8465608465608465,
"eval_steps": 500,
"global_step": 15,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.1693121693121693,
"grad_norm": 3.812271470654861,
"learning_rate": 0.0,
"loss": 0.4358,
"step": 1
},
{
"epoch": 0.3386243386243386,
"grad_norm": 3.6837345877956977,
"learning_rate": 1e-05,
"loss": 0.437,
"step": 2
},
{
"epoch": 0.5079365079365079,
"grad_norm": 5.4942591993038805,
"learning_rate": 2e-05,
"loss": 0.4015,
"step": 3
},
{
"epoch": 0.6772486772486772,
"grad_norm": 3.1981448047695795,
"learning_rate": 1.9709418174260523e-05,
"loss": 0.362,
"step": 4
},
{
"epoch": 0.8465608465608465,
"grad_norm": 2.08023395165645,
"learning_rate": 1.8854560256532098e-05,
"loss": 0.3566,
"step": 5
},
{
"epoch": 1.1693121693121693,
"grad_norm": 13.21818033067951,
"learning_rate": 1.7485107481711014e-05,
"loss": 0.4189,
"step": 6
},
{
"epoch": 1.3386243386243386,
"grad_norm": 2.5599008897358635,
"learning_rate": 1.568064746731156e-05,
"loss": 0.3168,
"step": 7
},
{
"epoch": 1.507936507936508,
"grad_norm": 2.634569116956323,
"learning_rate": 1.3546048870425356e-05,
"loss": 0.3148,
"step": 8
},
{
"epoch": 1.6772486772486772,
"grad_norm": 1.7050997879118408,
"learning_rate": 1.1205366802553231e-05,
"loss": 0.2987,
"step": 9
},
{
"epoch": 1.8465608465608465,
"grad_norm": 2.7839305132633823,
"learning_rate": 8.79463319744677e-06,
"loss": 0.2887,
"step": 10
},
{
"epoch": 2.1693121693121693,
"grad_norm": 3.6052622772527414,
"learning_rate": 6.453951129574644e-06,
"loss": 0.3454,
"step": 11
},
{
"epoch": 2.3386243386243386,
"grad_norm": 2.8886720844496065,
"learning_rate": 4.319352532688444e-06,
"loss": 0.2702,
"step": 12
},
{
"epoch": 2.507936507936508,
"grad_norm": 1.8242696525345412,
"learning_rate": 2.514892518288988e-06,
"loss": 0.2734,
"step": 13
},
{
"epoch": 2.677248677248677,
"grad_norm": 2.437269171769602,
"learning_rate": 1.1454397434679022e-06,
"loss": 0.2841,
"step": 14
},
{
"epoch": 2.8465608465608465,
"grad_norm": 2.7728903491320915,
"learning_rate": 2.905818257394799e-07,
"loss": 0.2681,
"step": 15
},
{
"epoch": 2.8465608465608465,
"step": 15,
"total_flos": 434379331469312.0,
"train_loss": 0.33813624978065493,
"train_runtime": 7450.7673,
"train_samples_per_second": 2.43,
"train_steps_per_second": 0.002
}
],
"logging_steps": 1,
"max_steps": 15,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 434379331469312.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}