375 lines
11 KiB
JSON
375 lines
11 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 342,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 0.9084768030792475,
|
||
|
|
"epoch": 0.029282576866764276,
|
||
|
|
"grad_norm": 0.6526874899864197,
|
||
|
|
"learning_rate": 9.736842105263158e-05,
|
||
|
|
"loss": 0.9780763626098633,
|
||
|
|
"mean_token_accuracy": 0.7051678966730833,
|
||
|
|
"num_tokens": 48075.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.851213363558054,
|
||
|
|
"epoch": 0.05856515373352855,
|
||
|
|
"grad_norm": 0.6124968528747559,
|
||
|
|
"learning_rate": 9.444444444444444e-05,
|
||
|
|
"loss": 0.8791687965393067,
|
||
|
|
"mean_token_accuracy": 0.7213793203234673,
|
||
|
|
"num_tokens": 94230.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8435173436999321,
|
||
|
|
"epoch": 0.08784773060029283,
|
||
|
|
"grad_norm": 0.6288742423057556,
|
||
|
|
"learning_rate": 9.152046783625731e-05,
|
||
|
|
"loss": 0.8266268730163574,
|
||
|
|
"mean_token_accuracy": 0.7377924341708422,
|
||
|
|
"num_tokens": 141629.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8042548477649689,
|
||
|
|
"epoch": 0.1171303074670571,
|
||
|
|
"grad_norm": 0.5038673281669617,
|
||
|
|
"learning_rate": 8.859649122807017e-05,
|
||
|
|
"loss": 0.8090643882751465,
|
||
|
|
"mean_token_accuracy": 0.7400693342089653,
|
||
|
|
"num_tokens": 191350.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8231701619923115,
|
||
|
|
"epoch": 0.14641288433382138,
|
||
|
|
"grad_norm": 0.5801142454147339,
|
||
|
|
"learning_rate": 8.567251461988305e-05,
|
||
|
|
"loss": 0.7851119995117187,
|
||
|
|
"mean_token_accuracy": 0.7518898636102677,
|
||
|
|
"num_tokens": 239424.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8163149386644364,
|
||
|
|
"epoch": 0.17569546120058566,
|
||
|
|
"grad_norm": 0.5873974561691284,
|
||
|
|
"learning_rate": 8.27485380116959e-05,
|
||
|
|
"loss": 0.7841955661773682,
|
||
|
|
"mean_token_accuracy": 0.7512057028710842,
|
||
|
|
"num_tokens": 286988.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.766396289691329,
|
||
|
|
"epoch": 0.20497803806734993,
|
||
|
|
"grad_norm": 0.5832107663154602,
|
||
|
|
"learning_rate": 7.982456140350878e-05,
|
||
|
|
"loss": 0.7343662738800049,
|
||
|
|
"mean_token_accuracy": 0.7698234215378761,
|
||
|
|
"num_tokens": 331211.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7477757569402457,
|
||
|
|
"epoch": 0.2342606149341142,
|
||
|
|
"grad_norm": 0.5744237303733826,
|
||
|
|
"learning_rate": 7.690058479532164e-05,
|
||
|
|
"loss": 0.7584258556365967,
|
||
|
|
"mean_token_accuracy": 0.7560943610966205,
|
||
|
|
"num_tokens": 381196.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7231789533048868,
|
||
|
|
"epoch": 0.2635431918008785,
|
||
|
|
"grad_norm": 0.580487072467804,
|
||
|
|
"learning_rate": 7.397660818713451e-05,
|
||
|
|
"loss": 0.7524959087371826,
|
||
|
|
"mean_token_accuracy": 0.771955594420433,
|
||
|
|
"num_tokens": 427843.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7700492061674595,
|
||
|
|
"epoch": 0.29282576866764276,
|
||
|
|
"grad_norm": 0.5994941592216492,
|
||
|
|
"learning_rate": 7.105263157894737e-05,
|
||
|
|
"loss": 0.7649751663208008,
|
||
|
|
"mean_token_accuracy": 0.7535097397863865,
|
||
|
|
"num_tokens": 476175.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7387165613472462,
|
||
|
|
"epoch": 0.32210834553440704,
|
||
|
|
"grad_norm": 0.566872775554657,
|
||
|
|
"learning_rate": 6.812865497076024e-05,
|
||
|
|
"loss": 0.7395598411560058,
|
||
|
|
"mean_token_accuracy": 0.764047472178936,
|
||
|
|
"num_tokens": 525945.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6992391187697649,
|
||
|
|
"epoch": 0.3513909224011713,
|
||
|
|
"grad_norm": 0.5976600646972656,
|
||
|
|
"learning_rate": 6.52046783625731e-05,
|
||
|
|
"loss": 0.7317209720611573,
|
||
|
|
"mean_token_accuracy": 0.768539623171091,
|
||
|
|
"num_tokens": 578034.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6964165881276131,
|
||
|
|
"epoch": 0.3806734992679356,
|
||
|
|
"grad_norm": 0.5820103287696838,
|
||
|
|
"learning_rate": 6.228070175438597e-05,
|
||
|
|
"loss": 0.7210932731628418,
|
||
|
|
"mean_token_accuracy": 0.7791639149188996,
|
||
|
|
"num_tokens": 626037.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7017107924446464,
|
||
|
|
"epoch": 0.40995607613469986,
|
||
|
|
"grad_norm": 0.51661217212677,
|
||
|
|
"learning_rate": 5.935672514619883e-05,
|
||
|
|
"loss": 0.7269079685211182,
|
||
|
|
"mean_token_accuracy": 0.7747438743710517,
|
||
|
|
"num_tokens": 675700.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.664313331618905,
|
||
|
|
"epoch": 0.43923865300146414,
|
||
|
|
"grad_norm": 0.6018214821815491,
|
||
|
|
"learning_rate": 5.64327485380117e-05,
|
||
|
|
"loss": 0.7238316535949707,
|
||
|
|
"mean_token_accuracy": 0.7786699928343296,
|
||
|
|
"num_tokens": 721264.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6971867846325039,
|
||
|
|
"epoch": 0.4685212298682284,
|
||
|
|
"grad_norm": 0.5508215427398682,
|
||
|
|
"learning_rate": 5.350877192982456e-05,
|
||
|
|
"loss": 0.7297741413116455,
|
||
|
|
"mean_token_accuracy": 0.7762463159859181,
|
||
|
|
"num_tokens": 771422.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7035594137385488,
|
||
|
|
"epoch": 0.4978038067349927,
|
||
|
|
"grad_norm": 0.5616670846939087,
|
||
|
|
"learning_rate": 5.0584795321637435e-05,
|
||
|
|
"loss": 0.7257540225982666,
|
||
|
|
"mean_token_accuracy": 0.7646482229232788,
|
||
|
|
"num_tokens": 821862.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6510832279920578,
|
||
|
|
"epoch": 0.527086383601757,
|
||
|
|
"grad_norm": 0.5769926905632019,
|
||
|
|
"learning_rate": 4.7660818713450294e-05,
|
||
|
|
"loss": 0.7212032794952392,
|
||
|
|
"mean_token_accuracy": 0.7790563136339188,
|
||
|
|
"num_tokens": 865346.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6748028011992574,
|
||
|
|
"epoch": 0.5563689604685212,
|
||
|
|
"grad_norm": 0.4853920638561249,
|
||
|
|
"learning_rate": 4.473684210526316e-05,
|
||
|
|
"loss": 0.7330660820007324,
|
||
|
|
"mean_token_accuracy": 0.77990957275033,
|
||
|
|
"num_tokens": 913014.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6797729761339724,
|
||
|
|
"epoch": 0.5856515373352855,
|
||
|
|
"grad_norm": 0.6134698390960693,
|
||
|
|
"learning_rate": 4.1812865497076025e-05,
|
||
|
|
"loss": 0.7272362232208252,
|
||
|
|
"mean_token_accuracy": 0.7770261645317078,
|
||
|
|
"num_tokens": 959989.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6656798135489226,
|
||
|
|
"epoch": 0.6149341142020498,
|
||
|
|
"grad_norm": 0.578885018825531,
|
||
|
|
"learning_rate": 3.888888888888889e-05,
|
||
|
|
"loss": 0.7093584537506104,
|
||
|
|
"mean_token_accuracy": 0.7775001615285874,
|
||
|
|
"num_tokens": 1005525.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.719148834887892,
|
||
|
|
"epoch": 0.6442166910688141,
|
||
|
|
"grad_norm": 0.6008777022361755,
|
||
|
|
"learning_rate": 3.5964912280701756e-05,
|
||
|
|
"loss": 0.7323647022247315,
|
||
|
|
"mean_token_accuracy": 0.7657537378370762,
|
||
|
|
"num_tokens": 1057639.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6825897304341197,
|
||
|
|
"epoch": 0.6734992679355783,
|
||
|
|
"grad_norm": 0.5272810459136963,
|
||
|
|
"learning_rate": 3.304093567251462e-05,
|
||
|
|
"loss": 0.7061307907104493,
|
||
|
|
"mean_token_accuracy": 0.7762037940323353,
|
||
|
|
"num_tokens": 1105174.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.661959953326732,
|
||
|
|
"epoch": 0.7027818448023426,
|
||
|
|
"grad_norm": 0.5122461915016174,
|
||
|
|
"learning_rate": 3.0116959064327488e-05,
|
||
|
|
"loss": 0.7107499599456787,
|
||
|
|
"mean_token_accuracy": 0.7818717740476131,
|
||
|
|
"num_tokens": 1151577.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6610155651345849,
|
||
|
|
"epoch": 0.7320644216691069,
|
||
|
|
"grad_norm": 0.526084303855896,
|
||
|
|
"learning_rate": 2.7192982456140354e-05,
|
||
|
|
"loss": 0.7069365501403808,
|
||
|
|
"mean_token_accuracy": 0.778365894407034,
|
||
|
|
"num_tokens": 1197176.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6569115126505494,
|
||
|
|
"epoch": 0.7613469985358712,
|
||
|
|
"grad_norm": 0.5765594840049744,
|
||
|
|
"learning_rate": 2.4269005847953216e-05,
|
||
|
|
"loss": 0.7107649326324463,
|
||
|
|
"mean_token_accuracy": 0.7810819767415523,
|
||
|
|
"num_tokens": 1242452.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.680330157559365,
|
||
|
|
"epoch": 0.7906295754026355,
|
||
|
|
"grad_norm": 0.48215794563293457,
|
||
|
|
"learning_rate": 2.134502923976608e-05,
|
||
|
|
"loss": 0.7114960193634033,
|
||
|
|
"mean_token_accuracy": 0.780608157068491,
|
||
|
|
"num_tokens": 1292083.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6911160614341497,
|
||
|
|
"epoch": 0.8199121522693997,
|
||
|
|
"grad_norm": 0.5212950110435486,
|
||
|
|
"learning_rate": 1.8421052631578947e-05,
|
||
|
|
"loss": 0.7070685863494873,
|
||
|
|
"mean_token_accuracy": 0.7759679570794106,
|
||
|
|
"num_tokens": 1341196.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6736268265172839,
|
||
|
|
"epoch": 0.849194729136164,
|
||
|
|
"grad_norm": 0.5101751685142517,
|
||
|
|
"learning_rate": 1.5497076023391813e-05,
|
||
|
|
"loss": 0.7038165092468261,
|
||
|
|
"mean_token_accuracy": 0.7737922541797161,
|
||
|
|
"num_tokens": 1392259.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6745592596940696,
|
||
|
|
"epoch": 0.8784773060029283,
|
||
|
|
"grad_norm": 0.6226794719696045,
|
||
|
|
"learning_rate": 1.2573099415204679e-05,
|
||
|
|
"loss": 0.7034701824188232,
|
||
|
|
"mean_token_accuracy": 0.7773476600646972,
|
||
|
|
"num_tokens": 1438469.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6509192291647196,
|
||
|
|
"epoch": 0.9077598828696926,
|
||
|
|
"grad_norm": 0.6452497243881226,
|
||
|
|
"learning_rate": 9.649122807017545e-06,
|
||
|
|
"loss": 0.7247503280639649,
|
||
|
|
"mean_token_accuracy": 0.7795319095253944,
|
||
|
|
"num_tokens": 1482993.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6876670928671956,
|
||
|
|
"epoch": 0.9370424597364568,
|
||
|
|
"grad_norm": 0.5731638073921204,
|
||
|
|
"learning_rate": 6.725146198830409e-06,
|
||
|
|
"loss": 0.7193249702453614,
|
||
|
|
"mean_token_accuracy": 0.772569590061903,
|
||
|
|
"num_tokens": 1532345.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6438878546468914,
|
||
|
|
"epoch": 0.9663250366032211,
|
||
|
|
"grad_norm": 0.515450656414032,
|
||
|
|
"learning_rate": 3.8011695906432747e-06,
|
||
|
|
"loss": 0.6847426414489746,
|
||
|
|
"mean_token_accuracy": 0.7844431973993778,
|
||
|
|
"num_tokens": 1579041.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6448297386057675,
|
||
|
|
"epoch": 0.9956076134699854,
|
||
|
|
"grad_norm": 0.5102537274360657,
|
||
|
|
"learning_rate": 8.771929824561404e-07,
|
||
|
|
"loss": 0.7077066898345947,
|
||
|
|
"mean_token_accuracy": 0.7873889222741127,
|
||
|
|
"num_tokens": 1629312.0,
|
||
|
|
"step": 340
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 342,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"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": 1.4179475073792e+16,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|