SmolLM-135M-10M-word / model /trainer_state.json
amin's picture
Upload 11 files
aefafe1 verified
raw
history blame
2.31 kB
{
"best_metric": 3.5722010135650635,
"best_model_checkpoint": "/home/dousti_group/ghanizadeh/babyLM/model-save/5-epoch-tv-tokenized/checkpoint-113408",
"epoch": 4.0,
"eval_steps": 500,
"global_step": 113408,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"grad_norm": 5.181252479553223,
"learning_rate": 4.014476851196376e-05,
"loss": 4.3405,
"step": 28352
},
{
"epoch": 1.0,
"eval_loss": 3.8782432079315186,
"eval_runtime": 163.7675,
"eval_samples_per_second": 291.572,
"eval_steps_per_second": 36.448,
"step": 28352
},
{
"epoch": 2.0,
"grad_norm": 4.64185094833374,
"learning_rate": 3.0112558402944925e-05,
"loss": 3.7749,
"step": 56704
},
{
"epoch": 2.0,
"eval_loss": 3.6707634925842285,
"eval_runtime": 158.0831,
"eval_samples_per_second": 302.056,
"eval_steps_per_second": 37.759,
"step": 56704
},
{
"epoch": 3.0,
"grad_norm": 6.706910610198975,
"learning_rate": 2.0081056208410027e-05,
"loss": 3.493,
"step": 85056
},
{
"epoch": 3.0,
"eval_loss": 3.5801351070404053,
"eval_runtime": 158.2132,
"eval_samples_per_second": 301.808,
"eval_steps_per_second": 37.728,
"step": 85056
},
{
"epoch": 4.0,
"grad_norm": 10.424848556518555,
"learning_rate": 1.005061588560102e-05,
"loss": 3.2244,
"step": 113408
},
{
"epoch": 4.0,
"eval_loss": 3.5722010135650635,
"eval_runtime": 157.7259,
"eval_samples_per_second": 302.74,
"eval_steps_per_second": 37.844,
"step": 113408
}
],
"logging_steps": 500,
"max_steps": 141760,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"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": 2.959934159506637e+17,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}