smollm-ft-onlyqa-adapter / trainer_state.json
aipib's picture
End of training
9330548 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.0,
"eval_steps": 500,
"global_step": 96,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.20833333333333334,
"grad_norm": 0.19238783419132233,
"learning_rate": 0.001,
"loss": 2.1609,
"step": 5
},
{
"epoch": 0.4166666666666667,
"grad_norm": 0.25501957535743713,
"learning_rate": 0.0009450549450549451,
"loss": 112.7474,
"step": 10
},
{
"epoch": 0.625,
"grad_norm": 0.26403358578681946,
"learning_rate": 0.00089010989010989,
"loss": 1.7251,
"step": 15
},
{
"epoch": 0.8333333333333334,
"grad_norm": 0.22951294481754303,
"learning_rate": 0.0008351648351648352,
"loss": 1.6202,
"step": 20
},
{
"epoch": 1.0416666666666667,
"grad_norm": 0.24705559015274048,
"learning_rate": 0.0007802197802197802,
"loss": 1.6083,
"step": 25
},
{
"epoch": 1.25,
"grad_norm": 0.27762821316719055,
"learning_rate": 0.0007252747252747253,
"loss": 94.5299,
"step": 30
},
{
"epoch": 1.4583333333333333,
"grad_norm": 0.31244632601737976,
"learning_rate": 0.0006703296703296702,
"loss": 1.4189,
"step": 35
},
{
"epoch": 1.6666666666666665,
"grad_norm": 0.26604393124580383,
"learning_rate": 0.0006153846153846154,
"loss": 1.3986,
"step": 40
},
{
"epoch": 1.875,
"grad_norm": 0.2991505563259125,
"learning_rate": 0.0005604395604395605,
"loss": 1.2591,
"step": 45
},
{
"epoch": 2.0833333333333335,
"grad_norm": 0.24768340587615967,
"learning_rate": 0.0005054945054945055,
"loss": 1.2605,
"step": 50
},
{
"epoch": 2.2916666666666665,
"grad_norm": 0.2997336983680725,
"learning_rate": 0.00045054945054945057,
"loss": 121.1995,
"step": 55
},
{
"epoch": 2.5,
"grad_norm": 0.3109113574028015,
"learning_rate": 0.00039560439560439557,
"loss": 1.1185,
"step": 60
},
{
"epoch": 2.7083333333333335,
"grad_norm": 0.3374277353286743,
"learning_rate": 0.0003406593406593407,
"loss": 1.0705,
"step": 65
},
{
"epoch": 2.9166666666666665,
"grad_norm": 0.3387056887149811,
"learning_rate": 0.0002857142857142857,
"loss": 1.137,
"step": 70
},
{
"epoch": 3.125,
"grad_norm": 1.2697887420654297,
"learning_rate": 0.0002307692307692308,
"loss": 132.8697,
"step": 75
},
{
"epoch": 3.3333333333333335,
"grad_norm": 0.311249315738678,
"learning_rate": 0.00017582417582417585,
"loss": 0.9697,
"step": 80
},
{
"epoch": 3.5416666666666665,
"grad_norm": 0.2974355220794678,
"learning_rate": 0.00012087912087912089,
"loss": 0.916,
"step": 85
},
{
"epoch": 3.75,
"grad_norm": 0.37431657314300537,
"learning_rate": 6.593406593406594e-05,
"loss": 1.0749,
"step": 90
},
{
"epoch": 3.9583333333333335,
"grad_norm": 0.29740434885025024,
"learning_rate": 1.098901098901099e-05,
"loss": 1.0644,
"step": 95
},
{
"epoch": 4.0,
"step": 96,
"total_flos": 85256425787904.0,
"train_loss": 25.069714224586885,
"train_runtime": 1158.3272,
"train_samples_per_second": 0.332,
"train_steps_per_second": 0.083
}
],
"logging_steps": 5,
"max_steps": 96,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 85256425787904.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}