| { | |
| "best_metric": 0.03650856390595436, | |
| "best_model_checkpoint": "doc-topic-model_eval-04_train-00/checkpoint-14000", | |
| "epoch": 9.368836291913215, | |
| "eval_steps": 1000, | |
| "global_step": 19000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2465483234714004, | |
| "grad_norm": 0.32839319109916687, | |
| "learning_rate": 1.9950690335305722e-05, | |
| "loss": 0.1668, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "grad_norm": 0.4088590443134308, | |
| "learning_rate": 1.9901380670611442e-05, | |
| "loss": 0.0929, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4930966469428008, | |
| "eval_accuracy": 0.981496987239155, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.09063737094402313, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.1099, | |
| "eval_samples_per_second": 669.945, | |
| "eval_steps_per_second": 2.642, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7396449704142012, | |
| "grad_norm": 0.33391374349594116, | |
| "learning_rate": 1.9852071005917162e-05, | |
| "loss": 0.0874, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "grad_norm": 0.3162483870983124, | |
| "learning_rate": 1.980276134122288e-05, | |
| "loss": 0.0785, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9861932938856016, | |
| "eval_accuracy": 0.981496987239155, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.0703761950135231, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.2009, | |
| "eval_samples_per_second": 664.948, | |
| "eval_steps_per_second": 2.623, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.232741617357002, | |
| "grad_norm": 0.31390729546546936, | |
| "learning_rate": 1.9753451676528602e-05, | |
| "loss": 0.0684, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "grad_norm": 0.40165144205093384, | |
| "learning_rate": 1.9704142011834322e-05, | |
| "loss": 0.0622, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "eval_accuracy": 0.982317296916684, | |
| "eval_f1": 0.10215092439392849, | |
| "eval_loss": 0.057249072939157486, | |
| "eval_precision": 0.8442330558858502, | |
| "eval_recall": 0.05436447166921899, | |
| "eval_runtime": 14.4254, | |
| "eval_samples_per_second": 562.413, | |
| "eval_steps_per_second": 2.218, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.725838264299803, | |
| "grad_norm": 0.4135937988758087, | |
| "learning_rate": 1.965483234714004e-05, | |
| "loss": 0.0584, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "grad_norm": 0.324859082698822, | |
| "learning_rate": 1.9605522682445763e-05, | |
| "loss": 0.0542, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.972386587771203, | |
| "eval_accuracy": 0.9842951074690683, | |
| "eval_f1": 0.338958793010913, | |
| "eval_loss": 0.04988710954785347, | |
| "eval_precision": 0.7662442706929091, | |
| "eval_recall": 0.21761102603369065, | |
| "eval_runtime": 14.4271, | |
| "eval_samples_per_second": 562.343, | |
| "eval_steps_per_second": 2.218, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.2189349112426036, | |
| "grad_norm": 0.3279891312122345, | |
| "learning_rate": 1.9556213017751483e-05, | |
| "loss": 0.0488, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "grad_norm": 0.4424910843372345, | |
| "learning_rate": 1.95069033530572e-05, | |
| "loss": 0.048, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.465483234714004, | |
| "eval_accuracy": 0.9853463506136738, | |
| "eval_f1": 0.4277731673582296, | |
| "eval_loss": 0.045854322612285614, | |
| "eval_precision": 0.7708873379860419, | |
| "eval_recall": 0.2960183767228178, | |
| "eval_runtime": 14.537, | |
| "eval_samples_per_second": 558.095, | |
| "eval_steps_per_second": 2.201, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.712031558185404, | |
| "grad_norm": 0.3125520646572113, | |
| "learning_rate": 1.9457593688362923e-05, | |
| "loss": 0.0453, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "grad_norm": 0.4686921536922455, | |
| "learning_rate": 1.940828402366864e-05, | |
| "loss": 0.0436, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.9585798816568047, | |
| "eval_accuracy": 0.9862800018701361, | |
| "eval_f1": 0.5134646302250804, | |
| "eval_loss": 0.04293818399310112, | |
| "eval_precision": 0.7466393921683226, | |
| "eval_recall": 0.39127105666156203, | |
| "eval_runtime": 14.3533, | |
| "eval_samples_per_second": 565.238, | |
| "eval_steps_per_second": 2.229, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.2051282051282053, | |
| "grad_norm": 0.3562796711921692, | |
| "learning_rate": 1.935897435897436e-05, | |
| "loss": 0.0412, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "grad_norm": 0.24857360124588013, | |
| "learning_rate": 1.930966469428008e-05, | |
| "loss": 0.0384, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.4516765285996054, | |
| "eval_accuracy": 0.9867872054358622, | |
| "eval_f1": 0.551246270811279, | |
| "eval_loss": 0.041109804064035416, | |
| "eval_precision": 0.7417767417767418, | |
| "eval_recall": 0.43859111791730476, | |
| "eval_runtime": 14.583, | |
| "eval_samples_per_second": 556.331, | |
| "eval_steps_per_second": 2.194, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.698224852071006, | |
| "grad_norm": 0.47940972447395325, | |
| "learning_rate": 1.92603550295858e-05, | |
| "loss": 0.0394, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "grad_norm": 0.39106041193008423, | |
| "learning_rate": 1.921104536489152e-05, | |
| "loss": 0.0385, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.9447731755424065, | |
| "eval_accuracy": 0.9868410426858554, | |
| "eval_f1": 0.5391028185788012, | |
| "eval_loss": 0.0395812913775444, | |
| "eval_precision": 0.7659334461364918, | |
| "eval_recall": 0.4159264931087289, | |
| "eval_runtime": 14.388, | |
| "eval_samples_per_second": 563.873, | |
| "eval_steps_per_second": 2.224, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.191321499013807, | |
| "grad_norm": 0.4470931887626648, | |
| "learning_rate": 1.916173570019724e-05, | |
| "loss": 0.0342, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "grad_norm": 0.33709824085235596, | |
| "learning_rate": 1.911242603550296e-05, | |
| "loss": 0.0343, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.437869822485207, | |
| "eval_accuracy": 0.9870167221332019, | |
| "eval_f1": 0.5622014141028091, | |
| "eval_loss": 0.039211541414260864, | |
| "eval_precision": 0.7474593495934959, | |
| "eval_recall": 0.45053598774885145, | |
| "eval_runtime": 14.475, | |
| "eval_samples_per_second": 560.482, | |
| "eval_steps_per_second": 2.211, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.684418145956608, | |
| "grad_norm": 0.3310143053531647, | |
| "learning_rate": 1.906311637080868e-05, | |
| "loss": 0.0343, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "grad_norm": 0.5156289935112, | |
| "learning_rate": 1.90138067061144e-05, | |
| "loss": 0.0343, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.930966469428008, | |
| "eval_accuracy": 0.9872320711331749, | |
| "eval_f1": 0.5746649046630168, | |
| "eval_loss": 0.038305457681417465, | |
| "eval_precision": 0.7490157480314961, | |
| "eval_recall": 0.466156202143951, | |
| "eval_runtime": 14.4319, | |
| "eval_samples_per_second": 562.156, | |
| "eval_steps_per_second": 2.217, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.177514792899408, | |
| "grad_norm": 0.3030126988887787, | |
| "learning_rate": 1.896459566074951e-05, | |
| "loss": 0.0314, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "grad_norm": 0.49180835485458374, | |
| "learning_rate": 1.891528599605523e-05, | |
| "loss": 0.0304, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.424063116370808, | |
| "eval_accuracy": 0.9873284114752682, | |
| "eval_f1": 0.5883273497192304, | |
| "eval_loss": 0.03808252140879631, | |
| "eval_precision": 0.7374798061389337, | |
| "eval_recall": 0.48935681470137826, | |
| "eval_runtime": 14.3093, | |
| "eval_samples_per_second": 566.975, | |
| "eval_steps_per_second": 2.236, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.670611439842209, | |
| "grad_norm": 0.4685238003730774, | |
| "learning_rate": 1.886597633136095e-05, | |
| "loss": 0.0297, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "grad_norm": 0.46912193298339844, | |
| "learning_rate": 1.881666666666667e-05, | |
| "loss": 0.0299, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.9171597633136095, | |
| "eval_accuracy": 0.9876826039094344, | |
| "eval_f1": 0.6116322701688556, | |
| "eval_loss": 0.036726485937833786, | |
| "eval_precision": 0.7340767746086211, | |
| "eval_recall": 0.5241960183767228, | |
| "eval_runtime": 14.4271, | |
| "eval_samples_per_second": 562.346, | |
| "eval_steps_per_second": 2.218, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.16370808678501, | |
| "grad_norm": 0.3353117108345032, | |
| "learning_rate": 1.876735700197239e-05, | |
| "loss": 0.0276, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "grad_norm": 0.2552751898765564, | |
| "learning_rate": 1.8718145956607497e-05, | |
| "loss": 0.0265, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.410256410256411, | |
| "eval_accuracy": 0.9876018480344445, | |
| "eval_f1": 0.6156616452193773, | |
| "eval_loss": 0.03736021742224693, | |
| "eval_precision": 0.7219075084972706, | |
| "eval_recall": 0.536676875957121, | |
| "eval_runtime": 14.3916, | |
| "eval_samples_per_second": 563.73, | |
| "eval_steps_per_second": 2.224, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.65680473372781, | |
| "grad_norm": 0.2560911774635315, | |
| "learning_rate": 1.8668836291913217e-05, | |
| "loss": 0.0273, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "grad_norm": 0.4542127549648285, | |
| "learning_rate": 1.8619526627218937e-05, | |
| "loss": 0.0261, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.903353057199211, | |
| "eval_accuracy": 0.9879177876857208, | |
| "eval_f1": 0.6178869074289811, | |
| "eval_loss": 0.03650856390595436, | |
| "eval_precision": 0.7447612875351047, | |
| "eval_recall": 0.527947932618683, | |
| "eval_runtime": 14.6033, | |
| "eval_samples_per_second": 555.561, | |
| "eval_steps_per_second": 2.191, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.149901380670611, | |
| "grad_norm": 0.2716215252876282, | |
| "learning_rate": 1.8570216962524657e-05, | |
| "loss": 0.025, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "grad_norm": 0.2959195077419281, | |
| "learning_rate": 1.8520907297830377e-05, | |
| "loss": 0.0236, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.396449704142012, | |
| "eval_accuracy": 0.9877236902317977, | |
| "eval_f1": 0.6227524054159955, | |
| "eval_loss": 0.037361323833465576, | |
| "eval_precision": 0.7217680896155011, | |
| "eval_recall": 0.5476263399693722, | |
| "eval_runtime": 14.3637, | |
| "eval_samples_per_second": 564.828, | |
| "eval_steps_per_second": 2.228, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.642998027613412, | |
| "grad_norm": 0.44261014461517334, | |
| "learning_rate": 1.8471696252465485e-05, | |
| "loss": 0.0225, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "grad_norm": 0.721012532711029, | |
| "learning_rate": 1.8422386587771205e-05, | |
| "loss": 0.0236, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.889546351084813, | |
| "eval_accuracy": 0.9879602907778208, | |
| "eval_f1": 0.6263301380705303, | |
| "eval_loss": 0.037222035229206085, | |
| "eval_precision": 0.7355918198719272, | |
| "eval_recall": 0.5453292496171516, | |
| "eval_runtime": 14.4785, | |
| "eval_samples_per_second": 560.347, | |
| "eval_steps_per_second": 2.21, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.136094674556213, | |
| "grad_norm": 0.2590983510017395, | |
| "learning_rate": 1.8373076923076926e-05, | |
| "loss": 0.0219, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.382642998027613, | |
| "grad_norm": 0.6246200203895569, | |
| "learning_rate": 1.8323767258382646e-05, | |
| "loss": 0.0215, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.382642998027613, | |
| "eval_accuracy": 0.9878738678238842, | |
| "eval_f1": 0.6325820991629105, | |
| "eval_loss": 0.03760357201099396, | |
| "eval_precision": 0.7198827552515877, | |
| "eval_recall": 0.5641653905053599, | |
| "eval_runtime": 14.3741, | |
| "eval_samples_per_second": 564.417, | |
| "eval_steps_per_second": 2.226, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.629191321499015, | |
| "grad_norm": 0.46473428606987, | |
| "learning_rate": 1.8274457593688362e-05, | |
| "loss": 0.0204, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 8.875739644970414, | |
| "grad_norm": 0.5553109049797058, | |
| "learning_rate": 1.822524654832347e-05, | |
| "loss": 0.0216, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.875739644970414, | |
| "eval_accuracy": 0.9878001957975776, | |
| "eval_f1": 0.632181453163043, | |
| "eval_loss": 0.03807205334305763, | |
| "eval_precision": 0.7149067722925321, | |
| "eval_recall": 0.5666156202143952, | |
| "eval_runtime": 14.3955, | |
| "eval_samples_per_second": 563.577, | |
| "eval_steps_per_second": 2.223, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.122287968441814, | |
| "grad_norm": 0.3819935917854309, | |
| "learning_rate": 1.8175936883629194e-05, | |
| "loss": 0.0194, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.368836291913215, | |
| "grad_norm": 0.5073143243789673, | |
| "learning_rate": 1.8126627218934914e-05, | |
| "loss": 0.0177, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 9.368836291913215, | |
| "eval_accuracy": 0.987988626172554, | |
| "eval_f1": 0.6384956506907726, | |
| "eval_loss": 0.03773576766252518, | |
| "eval_precision": 0.7204580446497305, | |
| "eval_recall": 0.5732771822358346, | |
| "eval_runtime": 14.3771, | |
| "eval_samples_per_second": 564.3, | |
| "eval_steps_per_second": 2.226, | |
| "step": 19000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 202800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 1000, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 430075849523328.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |