192 lines
4.6 KiB
JSON
192 lines
4.6 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_steps": 200,
|
||
|
|
"global_step": 204,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.14814814814814814,
|
||
|
|
"grad_norm": 2.5787038803100586,
|
||
|
|
"learning_rate": 1.6363636363636366e-05,
|
||
|
|
"loss": 0.8932,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2962962962962963,
|
||
|
|
"grad_norm": 1.3832252025604248,
|
||
|
|
"learning_rate": 1.991533160993366e-05,
|
||
|
|
"loss": 0.7355,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4444444444444444,
|
||
|
|
"grad_norm": 1.1969517469406128,
|
||
|
|
"learning_rate": 1.9573822052208013e-05,
|
||
|
|
"loss": 0.6846,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5925925925925926,
|
||
|
|
"grad_norm": 1.1520695686340332,
|
||
|
|
"learning_rate": 1.8979201311434434e-05,
|
||
|
|
"loss": 0.6672,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7407407407407407,
|
||
|
|
"grad_norm": 1.3113605976104736,
|
||
|
|
"learning_rate": 1.8147189876080463e-05,
|
||
|
|
"loss": 0.6607,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8888888888888888,
|
||
|
|
"grad_norm": 1.1071128845214844,
|
||
|
|
"learning_rate": 1.709978433199901e-05,
|
||
|
|
"loss": 0.6538,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0296296296296297,
|
||
|
|
"grad_norm": 1.3430020809173584,
|
||
|
|
"learning_rate": 1.586467582019392e-05,
|
||
|
|
"loss": 0.6235,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.1777777777777778,
|
||
|
|
"grad_norm": 1.1768577098846436,
|
||
|
|
"learning_rate": 1.4474517942843173e-05,
|
||
|
|
"loss": 0.5681,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.325925925925926,
|
||
|
|
"grad_norm": 1.1502654552459717,
|
||
|
|
"learning_rate": 1.2966063472563686e-05,
|
||
|
|
"loss": 0.5632,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.474074074074074,
|
||
|
|
"grad_norm": 1.2426036596298218,
|
||
|
|
"learning_rate": 1.137919268846074e-05,
|
||
|
|
"loss": 0.5595,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.6222222222222222,
|
||
|
|
"grad_norm": 1.1577049493789673,
|
||
|
|
"learning_rate": 9.755859027658848e-06,
|
||
|
|
"loss": 0.5516,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.7703703703703704,
|
||
|
|
"grad_norm": 1.0877997875213623,
|
||
|
|
"learning_rate": 8.138979927010964e-06,
|
||
|
|
"loss": 0.552,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.9185185185185185,
|
||
|
|
"grad_norm": 1.0529100894927979,
|
||
|
|
"learning_rate": 6.571302178736404e-06,
|
||
|
|
"loss": 0.5492,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0592592592592593,
|
||
|
|
"grad_norm": 1.0376404523849487,
|
||
|
|
"learning_rate": 5.094271797534404e-06,
|
||
|
|
"loss": 0.5272,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.2074074074074073,
|
||
|
|
"grad_norm": 1.0315033197402954,
|
||
|
|
"learning_rate": 3.74693827744665e-06,
|
||
|
|
"loss": 0.492,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.3555555555555556,
|
||
|
|
"grad_norm": 0.9723921418190002,
|
||
|
|
"learning_rate": 2.564922207551337e-06,
|
||
|
|
"loss": 0.4981,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.5037037037037035,
|
||
|
|
"grad_norm": 0.9567559957504272,
|
||
|
|
"learning_rate": 1.579473540500751e-06,
|
||
|
|
"loss": 0.4926,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.651851851851852,
|
||
|
|
"grad_norm": 0.994361937046051,
|
||
|
|
"learning_rate": 8.166454112487287e-07,
|
||
|
|
"loss": 0.4956,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.8,
|
||
|
|
"grad_norm": 0.9385385513305664,
|
||
|
|
"learning_rate": 2.9660534841699175e-07,
|
||
|
|
"loss": 0.4903,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.948148148148148,
|
||
|
|
"grad_norm": 0.9814600944519043,
|
||
|
|
"learning_rate": 3.310208838458562e-08,
|
||
|
|
"loss": 0.4883,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.948148148148148,
|
||
|
|
"eval_loss": 0.6684299111366272,
|
||
|
|
"eval_runtime": 0.3376,
|
||
|
|
"eval_samples_per_second": 130.315,
|
||
|
|
"eval_steps_per_second": 17.77,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.0,
|
||
|
|
"step": 204,
|
||
|
|
"total_flos": 5.359421722984448e+16,
|
||
|
|
"train_loss": 0.5856129340097016,
|
||
|
|
"train_runtime": 310.1981,
|
||
|
|
"train_samples_per_second": 41.751,
|
||
|
|
"train_steps_per_second": 0.658
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 204,
|
||
|
|
"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": 5.359421722984448e+16,
|
||
|
|
"train_batch_size": 4,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|