| { | |
| "best_metric": 1.735720157623291, | |
| "best_model_checkpoint": "models/dehanalkautsar/mbert-uncased-modified_embedding_table-multi/checkpoint-62000", | |
| "epoch": 44.606610911987254, | |
| "eval_steps": 2000, | |
| "global_step": 70000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.273994424532059, | |
| "grad_norm": 14.472808837890625, | |
| "learning_rate": 4.840662842574888e-05, | |
| "loss": 6.7847, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.273994424532059, | |
| "eval_loss": 6.310997486114502, | |
| "eval_runtime": 1.0444, | |
| "eval_samples_per_second": 134.046, | |
| "eval_steps_per_second": 2.872, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.547988849064118, | |
| "grad_norm": 13.92039966583252, | |
| "learning_rate": 4.6813256851497774e-05, | |
| "loss": 5.2141, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.547988849064118, | |
| "eval_loss": 4.0085954666137695, | |
| "eval_runtime": 1.0423, | |
| "eval_samples_per_second": 134.315, | |
| "eval_steps_per_second": 2.878, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.8219832735961767, | |
| "grad_norm": 12.217097282409668, | |
| "learning_rate": 4.521988527724665e-05, | |
| "loss": 3.6444, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.8219832735961767, | |
| "eval_loss": 3.3225996494293213, | |
| "eval_runtime": 1.0422, | |
| "eval_samples_per_second": 134.329, | |
| "eval_steps_per_second": 2.878, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.09557945041816, | |
| "grad_norm": 13.112276077270508, | |
| "learning_rate": 4.362651370299554e-05, | |
| "loss": 3.185, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.09557945041816, | |
| "eval_loss": 2.9885096549987793, | |
| "eval_runtime": 1.0432, | |
| "eval_samples_per_second": 134.208, | |
| "eval_steps_per_second": 2.876, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.369573874950219, | |
| "grad_norm": 9.95456314086914, | |
| "learning_rate": 4.2033142128744425e-05, | |
| "loss": 2.9191, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.369573874950219, | |
| "eval_loss": 2.759176254272461, | |
| "eval_runtime": 1.0435, | |
| "eval_samples_per_second": 134.158, | |
| "eval_steps_per_second": 2.875, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 7.643568299482278, | |
| "grad_norm": 10.692479133605957, | |
| "learning_rate": 4.043977055449331e-05, | |
| "loss": 2.7335, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.643568299482278, | |
| "eval_loss": 2.601854085922241, | |
| "eval_runtime": 1.7999, | |
| "eval_samples_per_second": 77.783, | |
| "eval_steps_per_second": 1.667, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.917562724014337, | |
| "grad_norm": 10.000415802001953, | |
| "learning_rate": 3.8846398980242196e-05, | |
| "loss": 2.6016, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.917562724014337, | |
| "eval_loss": 2.4759533405303955, | |
| "eval_runtime": 2.1495, | |
| "eval_samples_per_second": 65.131, | |
| "eval_steps_per_second": 1.396, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 10.19115890083632, | |
| "grad_norm": 9.746127128601074, | |
| "learning_rate": 3.725302740599108e-05, | |
| "loss": 2.4983, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.19115890083632, | |
| "eval_loss": 2.3779261112213135, | |
| "eval_runtime": 2.1857, | |
| "eval_samples_per_second": 64.052, | |
| "eval_steps_per_second": 1.373, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 11.465153325368378, | |
| "grad_norm": 26.588117599487305, | |
| "learning_rate": 3.565965583173996e-05, | |
| "loss": 2.4195, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.465153325368378, | |
| "eval_loss": 2.34489369392395, | |
| "eval_runtime": 1.0425, | |
| "eval_samples_per_second": 134.292, | |
| "eval_steps_per_second": 2.878, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 12.739147749900438, | |
| "grad_norm": 10.147093772888184, | |
| "learning_rate": 3.4066284257488853e-05, | |
| "loss": 2.355, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.739147749900438, | |
| "eval_loss": 2.2396912574768066, | |
| "eval_runtime": 1.0419, | |
| "eval_samples_per_second": 134.367, | |
| "eval_steps_per_second": 2.879, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 14.012743926722422, | |
| "grad_norm": 9.359902381896973, | |
| "learning_rate": 3.247291268323773e-05, | |
| "loss": 2.3011, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.012743926722422, | |
| "eval_loss": 2.2220568656921387, | |
| "eval_runtime": 1.0421, | |
| "eval_samples_per_second": 134.349, | |
| "eval_steps_per_second": 2.879, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 15.296296296296296, | |
| "grad_norm": 9.719951629638672, | |
| "learning_rate": 4.550106849623215e-05, | |
| "loss": 2.2641, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 15.296296296296296, | |
| "eval_loss": 2.1969993114471436, | |
| "eval_runtime": 1.0459, | |
| "eval_samples_per_second": 133.856, | |
| "eval_steps_per_second": 2.868, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 16.570290720828353, | |
| "grad_norm": 10.155664443969727, | |
| "learning_rate": 4.5126157537584826e-05, | |
| "loss": 2.2145, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 16.570290720828353, | |
| "eval_loss": 2.1572253704071045, | |
| "eval_runtime": 1.0453, | |
| "eval_samples_per_second": 133.93, | |
| "eval_steps_per_second": 2.87, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 17.844285145360413, | |
| "grad_norm": 9.276083946228027, | |
| "learning_rate": 4.4751246578937504e-05, | |
| "loss": 2.1687, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 17.844285145360413, | |
| "eval_loss": 2.1415774822235107, | |
| "eval_runtime": 1.0448, | |
| "eval_samples_per_second": 134.002, | |
| "eval_steps_per_second": 2.871, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 19.1178813221824, | |
| "grad_norm": 9.626545906066895, | |
| "learning_rate": 4.437633562029018e-05, | |
| "loss": 2.1288, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 19.1178813221824, | |
| "eval_loss": 2.0869765281677246, | |
| "eval_runtime": 1.0442, | |
| "eval_samples_per_second": 134.079, | |
| "eval_steps_per_second": 2.873, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 20.391875746714458, | |
| "grad_norm": 9.785582542419434, | |
| "learning_rate": 4.400142466164286e-05, | |
| "loss": 2.0936, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 20.391875746714458, | |
| "eval_loss": 2.0372226238250732, | |
| "eval_runtime": 1.0457, | |
| "eval_samples_per_second": 133.886, | |
| "eval_steps_per_second": 2.869, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 21.665870171246514, | |
| "grad_norm": 10.339402198791504, | |
| "learning_rate": 4.362651370299554e-05, | |
| "loss": 2.0629, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 21.665870171246514, | |
| "eval_loss": 2.0290327072143555, | |
| "eval_runtime": 1.0449, | |
| "eval_samples_per_second": 133.981, | |
| "eval_steps_per_second": 2.871, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 22.939864595778573, | |
| "grad_norm": 33.6297492980957, | |
| "learning_rate": 4.325160274434822e-05, | |
| "loss": 2.0352, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 22.939864595778573, | |
| "eval_loss": 1.9997824430465698, | |
| "eval_runtime": 1.0457, | |
| "eval_samples_per_second": 133.884, | |
| "eval_steps_per_second": 2.869, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 24.21346077260056, | |
| "grad_norm": 9.419952392578125, | |
| "learning_rate": 4.2876691785700895e-05, | |
| "loss": 2.0085, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 24.21346077260056, | |
| "eval_loss": 1.9848532676696777, | |
| "eval_runtime": 1.0456, | |
| "eval_samples_per_second": 133.889, | |
| "eval_steps_per_second": 2.869, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 25.487455197132615, | |
| "grad_norm": 9.297412872314453, | |
| "learning_rate": 4.250178082705358e-05, | |
| "loss": 1.9835, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 25.487455197132615, | |
| "eval_loss": 1.9594355821609497, | |
| "eval_runtime": 1.0451, | |
| "eval_samples_per_second": 133.962, | |
| "eval_steps_per_second": 2.871, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 26.761449621664674, | |
| "grad_norm": 9.560755729675293, | |
| "learning_rate": 4.212686986840625e-05, | |
| "loss": 1.9622, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 26.761449621664674, | |
| "eval_loss": 1.9543912410736084, | |
| "eval_runtime": 1.0446, | |
| "eval_samples_per_second": 134.027, | |
| "eval_steps_per_second": 2.872, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 28.03504579848666, | |
| "grad_norm": 9.696600914001465, | |
| "learning_rate": 4.1751958909758937e-05, | |
| "loss": 1.9423, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 28.03504579848666, | |
| "eval_loss": 1.9044808149337769, | |
| "eval_runtime": 1.0452, | |
| "eval_samples_per_second": 133.942, | |
| "eval_steps_per_second": 2.87, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 29.309040223018716, | |
| "grad_norm": 9.66083812713623, | |
| "learning_rate": 4.1377047951111615e-05, | |
| "loss": 1.9237, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 29.309040223018716, | |
| "eval_loss": 1.872032642364502, | |
| "eval_runtime": 1.0448, | |
| "eval_samples_per_second": 134.0, | |
| "eval_steps_per_second": 2.871, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 30.583034647550775, | |
| "grad_norm": 8.918251991271973, | |
| "learning_rate": 4.100213699246429e-05, | |
| "loss": 1.905, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 30.583034647550775, | |
| "eval_loss": 1.857001781463623, | |
| "eval_runtime": 1.0437, | |
| "eval_samples_per_second": 134.137, | |
| "eval_steps_per_second": 2.874, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 31.857029072082835, | |
| "grad_norm": 9.568062782287598, | |
| "learning_rate": 4.062722603381697e-05, | |
| "loss": 1.8905, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 31.857029072082835, | |
| "eval_loss": 1.8617095947265625, | |
| "eval_runtime": 1.0446, | |
| "eval_samples_per_second": 134.024, | |
| "eval_steps_per_second": 2.872, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 33.14145758661888, | |
| "grad_norm": 9.223485946655273, | |
| "learning_rate": 4.025231507516965e-05, | |
| "loss": 1.8744, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 33.14145758661888, | |
| "eval_loss": 1.8508821725845337, | |
| "eval_runtime": 1.0449, | |
| "eval_samples_per_second": 133.978, | |
| "eval_steps_per_second": 2.871, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 34.41545201115093, | |
| "grad_norm": 9.19514274597168, | |
| "learning_rate": 3.987740411652233e-05, | |
| "loss": 1.8594, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 34.41545201115093, | |
| "eval_loss": 1.807867169380188, | |
| "eval_runtime": 1.0427, | |
| "eval_samples_per_second": 134.272, | |
| "eval_steps_per_second": 2.877, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 35.689446435682996, | |
| "grad_norm": 9.288874626159668, | |
| "learning_rate": 3.9502493157875006e-05, | |
| "loss": 1.845, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 35.689446435682996, | |
| "eval_loss": 1.8352290391921997, | |
| "eval_runtime": 1.0447, | |
| "eval_samples_per_second": 134.011, | |
| "eval_steps_per_second": 2.872, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 36.96344086021505, | |
| "grad_norm": 8.9916353225708, | |
| "learning_rate": 3.9127582199227684e-05, | |
| "loss": 1.8325, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 36.96344086021505, | |
| "eval_loss": 1.8519763946533203, | |
| "eval_runtime": 1.0422, | |
| "eval_samples_per_second": 134.334, | |
| "eval_steps_per_second": 2.879, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 38.237037037037034, | |
| "grad_norm": 9.245351791381836, | |
| "learning_rate": 3.875267124058036e-05, | |
| "loss": 1.8204, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 38.237037037037034, | |
| "eval_loss": 1.792788028717041, | |
| "eval_runtime": 1.0429, | |
| "eval_samples_per_second": 134.237, | |
| "eval_steps_per_second": 2.877, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 39.5110314615691, | |
| "grad_norm": 9.415380477905273, | |
| "learning_rate": 3.837776028193305e-05, | |
| "loss": 1.808, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 39.5110314615691, | |
| "eval_loss": 1.735720157623291, | |
| "eval_runtime": 1.0435, | |
| "eval_samples_per_second": 134.164, | |
| "eval_steps_per_second": 2.875, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 40.78502588610115, | |
| "grad_norm": 9.212890625, | |
| "learning_rate": 3.800284932328572e-05, | |
| "loss": 1.7973, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 40.78502588610115, | |
| "eval_loss": 1.7825311422348022, | |
| "eval_runtime": 1.0432, | |
| "eval_samples_per_second": 134.204, | |
| "eval_steps_per_second": 2.876, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 42.058622062923135, | |
| "grad_norm": 12.76929759979248, | |
| "learning_rate": 3.76279383646384e-05, | |
| "loss": 1.7868, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 42.058622062923135, | |
| "eval_loss": 1.8231488466262817, | |
| "eval_runtime": 1.0438, | |
| "eval_samples_per_second": 134.121, | |
| "eval_steps_per_second": 2.874, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 43.3326164874552, | |
| "grad_norm": 9.313870429992676, | |
| "learning_rate": 3.725302740599108e-05, | |
| "loss": 1.776, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 43.3326164874552, | |
| "eval_loss": 1.7821528911590576, | |
| "eval_runtime": 1.0442, | |
| "eval_samples_per_second": 134.071, | |
| "eval_steps_per_second": 2.873, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 44.606610911987254, | |
| "grad_norm": 9.745814323425293, | |
| "learning_rate": 3.687811644734376e-05, | |
| "loss": 1.7671, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 44.606610911987254, | |
| "eval_loss": 1.7638314962387085, | |
| "eval_runtime": 1.0433, | |
| "eval_samples_per_second": 134.193, | |
| "eval_steps_per_second": 2.876, | |
| "step": 70000 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 266730, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 170, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.714819396263936e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |