tsunayoshi21's picture
Subiendo mi modelo entrenado finetuneado desde Trisert/tinyllama-alpaca
fe516e4 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9747899159663865,
"eval_steps": 8,
"global_step": 29,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03361344537815126,
"grad_norm": 1.0234375,
"learning_rate": 2e-05,
"loss": 2.6905,
"step": 1
},
{
"epoch": 0.03361344537815126,
"eval_loss": 2.729154109954834,
"eval_runtime": 20.6566,
"eval_samples_per_second": 2.421,
"eval_steps_per_second": 0.339,
"step": 1
},
{
"epoch": 0.06722689075630252,
"grad_norm": 1.0390625,
"learning_rate": 4e-05,
"loss": 2.687,
"step": 2
},
{
"epoch": 0.10084033613445378,
"grad_norm": 1.1171875,
"learning_rate": 6e-05,
"loss": 2.8062,
"step": 3
},
{
"epoch": 0.13445378151260504,
"grad_norm": 1.03125,
"learning_rate": 8e-05,
"loss": 2.6619,
"step": 4
},
{
"epoch": 0.16806722689075632,
"grad_norm": 1.0,
"learning_rate": 0.0001,
"loss": 2.5939,
"step": 5
},
{
"epoch": 0.20168067226890757,
"grad_norm": 0.95703125,
"learning_rate": 0.00012,
"loss": 2.6307,
"step": 6
},
{
"epoch": 0.23529411764705882,
"grad_norm": 0.86328125,
"learning_rate": 0.00014,
"loss": 2.5751,
"step": 7
},
{
"epoch": 0.2689075630252101,
"grad_norm": 0.82421875,
"learning_rate": 0.00016,
"loss": 2.4725,
"step": 8
},
{
"epoch": 0.2689075630252101,
"eval_loss": 2.3971643447875977,
"eval_runtime": 20.4746,
"eval_samples_per_second": 2.442,
"eval_steps_per_second": 0.342,
"step": 8
},
{
"epoch": 0.3025210084033613,
"grad_norm": 0.6953125,
"learning_rate": 0.00018,
"loss": 2.2905,
"step": 9
},
{
"epoch": 0.33613445378151263,
"grad_norm": 0.6015625,
"learning_rate": 0.0002,
"loss": 2.3883,
"step": 10
},
{
"epoch": 0.3697478991596639,
"grad_norm": 0.53125,
"learning_rate": 0.00019995608365087946,
"loss": 2.3071,
"step": 11
},
{
"epoch": 0.40336134453781514,
"grad_norm": 0.46484375,
"learning_rate": 0.00019982437317643217,
"loss": 2.0817,
"step": 12
},
{
"epoch": 0.4369747899159664,
"grad_norm": 0.56640625,
"learning_rate": 0.0001996049842615217,
"loss": 2.0883,
"step": 13
},
{
"epoch": 0.47058823529411764,
"grad_norm": 0.5,
"learning_rate": 0.00019929810960135172,
"loss": 2.1018,
"step": 14
},
{
"epoch": 0.5042016806722689,
"grad_norm": 0.5390625,
"learning_rate": 0.0001989040187322164,
"loss": 2.1015,
"step": 15
},
{
"epoch": 0.5378151260504201,
"grad_norm": 0.51953125,
"learning_rate": 0.00019842305779475968,
"loss": 1.9891,
"step": 16
},
{
"epoch": 0.5378151260504201,
"eval_loss": 2.071833610534668,
"eval_runtime": 20.5312,
"eval_samples_per_second": 2.435,
"eval_steps_per_second": 0.341,
"step": 16
},
{
"epoch": 0.5714285714285714,
"grad_norm": 0.4921875,
"learning_rate": 0.0001978556492299504,
"loss": 2.1245,
"step": 17
},
{
"epoch": 0.6050420168067226,
"grad_norm": 0.46875,
"learning_rate": 0.0001972022914080411,
"loss": 1.9485,
"step": 18
},
{
"epoch": 0.6386554621848739,
"grad_norm": 0.4140625,
"learning_rate": 0.00019646355819083589,
"loss": 2.0621,
"step": 19
},
{
"epoch": 0.6722689075630253,
"grad_norm": 0.41015625,
"learning_rate": 0.00019564009842765225,
"loss": 1.9076,
"step": 20
},
{
"epoch": 0.7058823529411765,
"grad_norm": 0.4375,
"learning_rate": 0.00019473263538541914,
"loss": 1.8968,
"step": 21
},
{
"epoch": 0.7394957983193278,
"grad_norm": 0.326171875,
"learning_rate": 0.0001937419661134121,
"loss": 1.8977,
"step": 22
},
{
"epoch": 0.773109243697479,
"grad_norm": 0.314453125,
"learning_rate": 0.00019266896074318334,
"loss": 1.9926,
"step": 23
},
{
"epoch": 0.8067226890756303,
"grad_norm": 0.27734375,
"learning_rate": 0.00019151456172430183,
"loss": 1.8345,
"step": 24
},
{
"epoch": 0.8067226890756303,
"eval_loss": 1.932895302772522,
"eval_runtime": 20.5148,
"eval_samples_per_second": 2.437,
"eval_steps_per_second": 0.341,
"step": 24
},
{
"epoch": 0.8403361344537815,
"grad_norm": 0.291015625,
"learning_rate": 0.00019027978299657436,
"loss": 1.8168,
"step": 25
},
{
"epoch": 0.8739495798319328,
"grad_norm": 0.29296875,
"learning_rate": 0.00018896570909947475,
"loss": 1.8794,
"step": 26
},
{
"epoch": 0.907563025210084,
"grad_norm": 0.306640625,
"learning_rate": 0.0001875734942195637,
"loss": 1.9604,
"step": 27
},
{
"epoch": 0.9411764705882353,
"grad_norm": 0.33984375,
"learning_rate": 0.00018610436117673555,
"loss": 1.824,
"step": 28
},
{
"epoch": 0.9747899159663865,
"grad_norm": 0.365234375,
"learning_rate": 0.0001845596003501826,
"loss": 1.9186,
"step": 29
}
],
"logging_steps": 1,
"max_steps": 116,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 29,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3366220896141312.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}