233 lines
5.6 KiB
JSON
233 lines
5.6 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 274,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.0364963503649635,
|
||
|
|
"grad_norm": 2.037085771560669,
|
||
|
|
"learning_rate": 6.4285714285714295e-06,
|
||
|
|
"loss": 0.7733,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.072992700729927,
|
||
|
|
"grad_norm": 1.040379524230957,
|
||
|
|
"learning_rate": 1.3571428571428574e-05,
|
||
|
|
"loss": 0.55,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10948905109489052,
|
||
|
|
"grad_norm": 1.7783838510513306,
|
||
|
|
"learning_rate": 1.9999184556954777e-05,
|
||
|
|
"loss": 0.2368,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.145985401459854,
|
||
|
|
"grad_norm": 0.8324124217033386,
|
||
|
|
"learning_rate": 1.9901492206471325e-05,
|
||
|
|
"loss": 0.1277,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18248175182481752,
|
||
|
|
"grad_norm": 0.272566020488739,
|
||
|
|
"learning_rate": 1.964253495453141e-05,
|
||
|
|
"loss": 0.1016,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21897810218978103,
|
||
|
|
"grad_norm": 0.3388763666152954,
|
||
|
|
"learning_rate": 1.9226530419534834e-05,
|
||
|
|
"loss": 0.0831,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.25547445255474455,
|
||
|
|
"grad_norm": 0.16205106675624847,
|
||
|
|
"learning_rate": 1.866025403784439e-05,
|
||
|
|
"loss": 0.0815,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.291970802919708,
|
||
|
|
"grad_norm": 0.17601074278354645,
|
||
|
|
"learning_rate": 1.7952928712734266e-05,
|
||
|
|
"loss": 0.0722,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3284671532846715,
|
||
|
|
"grad_norm": 0.33516547083854675,
|
||
|
|
"learning_rate": 1.7116074601618418e-05,
|
||
|
|
"loss": 0.0686,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.36496350364963503,
|
||
|
|
"grad_norm": 0.2756238877773285,
|
||
|
|
"learning_rate": 1.6163321488063636e-05,
|
||
|
|
"loss": 0.0648,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.40145985401459855,
|
||
|
|
"grad_norm": 0.357218474149704,
|
||
|
|
"learning_rate": 1.5110186794471105e-05,
|
||
|
|
"loss": 0.062,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.43795620437956206,
|
||
|
|
"grad_norm": 0.20831607282161713,
|
||
|
|
"learning_rate": 1.3973822850918055e-05,
|
||
|
|
"loss": 0.0571,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4744525547445255,
|
||
|
|
"grad_norm": 0.2679807245731354,
|
||
|
|
"learning_rate": 1.277273753637408e-05,
|
||
|
|
"loss": 0.0573,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5109489051094891,
|
||
|
|
"grad_norm": 0.2546704411506653,
|
||
|
|
"learning_rate": 1.1526492842188746e-05,
|
||
|
|
"loss": 0.053,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5474452554744526,
|
||
|
|
"grad_norm": 0.1792953908443451,
|
||
|
|
"learning_rate": 1.0255386267325602e-05,
|
||
|
|
"loss": 0.0508,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.583941605839416,
|
||
|
|
"grad_norm": 0.17367389798164368,
|
||
|
|
"learning_rate": 8.98012023443585e-06,
|
||
|
|
"loss": 0.0532,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6204379562043796,
|
||
|
|
"grad_norm": 0.14929895102977753,
|
||
|
|
"learning_rate": 7.721464910968628e-06,
|
||
|
|
"loss": 0.0521,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.656934306569343,
|
||
|
|
"grad_norm": 0.17261824011802673,
|
||
|
|
"learning_rate": 6.499919926926566e-06,
|
||
|
|
"loss": 0.0466,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6934306569343066,
|
||
|
|
"grad_norm": 0.1643795520067215,
|
||
|
|
"learning_rate": 5.335380498845559e-06,
|
||
|
|
"loss": 0.0517,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7299270072992701,
|
||
|
|
"grad_norm": 0.14236749708652496,
|
||
|
|
"learning_rate": 4.2468133978137945e-06,
|
||
|
|
"loss": 0.0565,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7664233576642335,
|
||
|
|
"grad_norm": 0.15162824094295502,
|
||
|
|
"learning_rate": 3.2519480390159806e-06,
|
||
|
|
"loss": 0.0485,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8029197080291971,
|
||
|
|
"grad_norm": 0.13050702214241028,
|
||
|
|
"learning_rate": 2.366987724005404e-06,
|
||
|
|
"loss": 0.0491,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8394160583941606,
|
||
|
|
"grad_norm": 0.11725970357656479,
|
||
|
|
"learning_rate": 1.6063457386805004e-06,
|
||
|
|
"loss": 0.0431,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8759124087591241,
|
||
|
|
"grad_norm": 0.1599208265542984,
|
||
|
|
"learning_rate": 9.82410605119486e-07,
|
||
|
|
"loss": 0.0542,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9124087591240876,
|
||
|
|
"grad_norm": 0.1203792542219162,
|
||
|
|
"learning_rate": 5.053443105997069e-07,
|
||
|
|
"loss": 0.0473,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.948905109489051,
|
||
|
|
"grad_norm": 0.11177993565797806,
|
||
|
|
"learning_rate": 1.8291680003145074e-07,
|
||
|
|
"loss": 0.0446,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9854014598540146,
|
||
|
|
"grad_norm": 0.0998225063085556,
|
||
|
|
"learning_rate": 2.037942741615617e-08,
|
||
|
|
"loss": 0.0467,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"step": 274,
|
||
|
|
"total_flos": 1.3702381181159342e+18,
|
||
|
|
"train_loss": 0.11128336125916809,
|
||
|
|
"train_runtime": 25797.0773,
|
||
|
|
"train_samples_per_second": 1.359,
|
||
|
|
"train_steps_per_second": 0.011
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 274,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 50,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 1.3702381181159342e+18,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|