s1K-sharegpt / trainer_state.json
neginr's picture
End of training
fbf6c9a verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 21,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.14285714285714285,
"grad_norm": 7.1918514360315156,
"learning_rate": 3.3333333333333333e-06,
"loss": 1.2539,
"step": 1
},
{
"epoch": 0.2857142857142857,
"grad_norm": 7.258564329223248,
"learning_rate": 6.666666666666667e-06,
"loss": 1.2536,
"step": 2
},
{
"epoch": 0.42857142857142855,
"grad_norm": 6.613027939855004,
"learning_rate": 1e-05,
"loss": 1.2389,
"step": 3
},
{
"epoch": 0.5714285714285714,
"grad_norm": 2.867616525907151,
"learning_rate": 9.924038765061042e-06,
"loss": 1.1476,
"step": 4
},
{
"epoch": 0.7142857142857143,
"grad_norm": 5.0623232957279,
"learning_rate": 9.698463103929542e-06,
"loss": 1.187,
"step": 5
},
{
"epoch": 0.8571428571428571,
"grad_norm": 5.122009865680415,
"learning_rate": 9.330127018922195e-06,
"loss": 1.19,
"step": 6
},
{
"epoch": 1.0,
"grad_norm": 5.169179697249293,
"learning_rate": 8.83022221559489e-06,
"loss": 1.1093,
"step": 7
},
{
"epoch": 1.1428571428571428,
"grad_norm": 4.245273791494235,
"learning_rate": 8.213938048432697e-06,
"loss": 1.0878,
"step": 8
},
{
"epoch": 1.2857142857142856,
"grad_norm": 2.688340622145421,
"learning_rate": 7.500000000000001e-06,
"loss": 1.102,
"step": 9
},
{
"epoch": 1.4285714285714286,
"grad_norm": 1.5102788029376382,
"learning_rate": 6.710100716628345e-06,
"loss": 1.0701,
"step": 10
},
{
"epoch": 1.5714285714285714,
"grad_norm": 1.458922754458695,
"learning_rate": 5.8682408883346535e-06,
"loss": 1.0754,
"step": 11
},
{
"epoch": 1.7142857142857144,
"grad_norm": 1.575146484429721,
"learning_rate": 5e-06,
"loss": 1.0604,
"step": 12
},
{
"epoch": 1.8571428571428572,
"grad_norm": 1.0891127424968043,
"learning_rate": 4.131759111665349e-06,
"loss": 1.0392,
"step": 13
},
{
"epoch": 2.0,
"grad_norm": 0.7807349854997148,
"learning_rate": 3.289899283371657e-06,
"loss": 1.0138,
"step": 14
},
{
"epoch": 2.142857142857143,
"grad_norm": 0.775626613409751,
"learning_rate": 2.5000000000000015e-06,
"loss": 1.0251,
"step": 15
},
{
"epoch": 2.2857142857142856,
"grad_norm": 0.8222834262472473,
"learning_rate": 1.7860619515673034e-06,
"loss": 1.018,
"step": 16
},
{
"epoch": 2.4285714285714284,
"grad_norm": 0.8048779475363516,
"learning_rate": 1.1697777844051105e-06,
"loss": 0.9789,
"step": 17
},
{
"epoch": 2.571428571428571,
"grad_norm": 0.630393040491335,
"learning_rate": 6.698729810778065e-07,
"loss": 0.962,
"step": 18
},
{
"epoch": 2.7142857142857144,
"grad_norm": 0.6033390719728393,
"learning_rate": 3.015368960704584e-07,
"loss": 1.0054,
"step": 19
},
{
"epoch": 2.857142857142857,
"grad_norm": 0.5305113909150767,
"learning_rate": 7.59612349389599e-08,
"loss": 1.0134,
"step": 20
},
{
"epoch": 3.0,
"grad_norm": 0.5221535456718317,
"learning_rate": 0.0,
"loss": 0.9561,
"step": 21
},
{
"epoch": 3.0,
"step": 21,
"total_flos": 65007551250432.0,
"train_loss": 1.0851242343584697,
"train_runtime": 629.9686,
"train_samples_per_second": 3.133,
"train_steps_per_second": 0.033
}
],
"logging_steps": 1,
"max_steps": 21,
"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": 65007551250432.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}