| { | |
| "best_global_step": 344, | |
| "best_metric": 0.8545918367346939, | |
| "best_model_checkpoint": "../results\\checkpoint-344", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 516, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11627906976744186, | |
| "grad_norm": 3.748704671859741, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.7308, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23255813953488372, | |
| "grad_norm": 4.270473957061768, | |
| "learning_rate": 2.5333333333333338e-06, | |
| "loss": 0.711, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3488372093023256, | |
| "grad_norm": 3.427567481994629, | |
| "learning_rate": 3.866666666666667e-06, | |
| "loss": 0.6948, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.46511627906976744, | |
| "grad_norm": 2.623924732208252, | |
| "learning_rate": 5.2e-06, | |
| "loss": 0.6786, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5813953488372093, | |
| "grad_norm": 3.8385608196258545, | |
| "learning_rate": 6.533333333333334e-06, | |
| "loss": 0.6393, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6976744186046512, | |
| "grad_norm": 6.035278797149658, | |
| "learning_rate": 7.866666666666667e-06, | |
| "loss": 0.6001, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.813953488372093, | |
| "grad_norm": 12.787239074707031, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.5629, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9302325581395349, | |
| "grad_norm": 5.702328681945801, | |
| "learning_rate": 1.0533333333333333e-05, | |
| "loss": 0.5313, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7, | |
| "eval_f1": 0.7009200572480064, | |
| "eval_loss": 0.5837136507034302, | |
| "eval_precision": 0.7089869281045751, | |
| "eval_recall": 0.7, | |
| "eval_runtime": 41.9795, | |
| "eval_samples_per_second": 1.667, | |
| "eval_steps_per_second": 0.071, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.0465116279069768, | |
| "grad_norm": 9.588857650756836, | |
| "learning_rate": 1.186666666666667e-05, | |
| "loss": 0.5636, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1627906976744187, | |
| "grad_norm": 11.674090385437012, | |
| "learning_rate": 1.3200000000000002e-05, | |
| "loss": 0.5236, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2790697674418605, | |
| "grad_norm": 6.487068176269531, | |
| "learning_rate": 1.4533333333333335e-05, | |
| "loss": 0.4575, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3953488372093024, | |
| "grad_norm": 7.364016532897949, | |
| "learning_rate": 1.586666666666667e-05, | |
| "loss": 0.436, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5116279069767442, | |
| "grad_norm": 3.313822031021118, | |
| "learning_rate": 1.72e-05, | |
| "loss": 0.4582, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.627906976744186, | |
| "grad_norm": 9.990741729736328, | |
| "learning_rate": 1.8533333333333334e-05, | |
| "loss": 0.4913, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.744186046511628, | |
| "grad_norm": 3.886936664581299, | |
| "learning_rate": 1.9866666666666667e-05, | |
| "loss": 0.4572, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8604651162790697, | |
| "grad_norm": 20.496986389160156, | |
| "learning_rate": 1.9601769911504427e-05, | |
| "loss": 0.4328, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9767441860465116, | |
| "grad_norm": 4.413027286529541, | |
| "learning_rate": 1.915929203539823e-05, | |
| "loss": 0.4989, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7, | |
| "eval_f1": 0.6979758740543855, | |
| "eval_loss": 0.5208796858787537, | |
| "eval_precision": 0.7318027210884354, | |
| "eval_recall": 0.7, | |
| "eval_runtime": 40.632, | |
| "eval_samples_per_second": 1.723, | |
| "eval_steps_per_second": 0.074, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.0930232558139537, | |
| "grad_norm": 4.327018737792969, | |
| "learning_rate": 1.8716814159292035e-05, | |
| "loss": 0.4098, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2093023255813953, | |
| "grad_norm": 3.8148539066314697, | |
| "learning_rate": 1.8274336283185843e-05, | |
| "loss": 0.4117, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.3255813953488373, | |
| "grad_norm": 10.718646049499512, | |
| "learning_rate": 1.7831858407079646e-05, | |
| "loss": 0.4328, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.441860465116279, | |
| "grad_norm": 4.208929538726807, | |
| "learning_rate": 1.7389380530973454e-05, | |
| "loss": 0.3792, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.558139534883721, | |
| "grad_norm": 7.7923479080200195, | |
| "learning_rate": 1.6946902654867258e-05, | |
| "loss": 0.3999, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.6744186046511627, | |
| "grad_norm": 4.283344745635986, | |
| "learning_rate": 1.6504424778761065e-05, | |
| "loss": 0.3568, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.7906976744186047, | |
| "grad_norm": 5.532216548919678, | |
| "learning_rate": 1.606194690265487e-05, | |
| "loss": 0.4055, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.9069767441860463, | |
| "grad_norm": 3.663770914077759, | |
| "learning_rate": 1.5619469026548676e-05, | |
| "loss": 0.3422, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7571428571428571, | |
| "eval_f1": 0.7574933003504432, | |
| "eval_loss": 0.4062376022338867, | |
| "eval_precision": 0.7581531954887218, | |
| "eval_recall": 0.7571428571428571, | |
| "eval_runtime": 39.0639, | |
| "eval_samples_per_second": 1.792, | |
| "eval_steps_per_second": 0.077, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 3.0232558139534884, | |
| "grad_norm": 4.565627098083496, | |
| "learning_rate": 1.517699115044248e-05, | |
| "loss": 0.3082, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.13953488372093, | |
| "grad_norm": 14.768342971801758, | |
| "learning_rate": 1.4734513274336284e-05, | |
| "loss": 0.3585, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.255813953488372, | |
| "grad_norm": 6.304065704345703, | |
| "learning_rate": 1.429203539823009e-05, | |
| "loss": 0.3781, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.3720930232558137, | |
| "grad_norm": 8.55212688446045, | |
| "learning_rate": 1.3849557522123895e-05, | |
| "loss": 0.3061, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.488372093023256, | |
| "grad_norm": 10.989974975585938, | |
| "learning_rate": 1.34070796460177e-05, | |
| "loss": 0.3507, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.604651162790698, | |
| "grad_norm": 8.045317649841309, | |
| "learning_rate": 1.2964601769911504e-05, | |
| "loss": 0.3536, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.7209302325581395, | |
| "grad_norm": 4.802742958068848, | |
| "learning_rate": 1.2522123893805312e-05, | |
| "loss": 0.3107, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.8372093023255816, | |
| "grad_norm": 2.4994394779205322, | |
| "learning_rate": 1.2079646017699115e-05, | |
| "loss": 0.3118, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.953488372093023, | |
| "grad_norm": 7.95784854888916, | |
| "learning_rate": 1.1637168141592921e-05, | |
| "loss": 0.305, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8545918367346939, | |
| "eval_loss": 0.34065258502960205, | |
| "eval_precision": 0.8655238095238094, | |
| "eval_recall": 0.8571428571428571, | |
| "eval_runtime": 39.7031, | |
| "eval_samples_per_second": 1.763, | |
| "eval_steps_per_second": 0.076, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 4.069767441860465, | |
| "grad_norm": 8.617964744567871, | |
| "learning_rate": 1.1194690265486727e-05, | |
| "loss": 0.3376, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.186046511627907, | |
| "grad_norm": 13.136701583862305, | |
| "learning_rate": 1.0752212389380532e-05, | |
| "loss": 0.3829, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.3023255813953485, | |
| "grad_norm": 3.352994203567505, | |
| "learning_rate": 1.0309734513274336e-05, | |
| "loss": 0.3315, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.4186046511627906, | |
| "grad_norm": 4.276778697967529, | |
| "learning_rate": 9.867256637168142e-06, | |
| "loss": 0.2992, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.534883720930233, | |
| "grad_norm": 4.883634090423584, | |
| "learning_rate": 9.424778761061947e-06, | |
| "loss": 0.2683, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.651162790697675, | |
| "grad_norm": 7.341738700866699, | |
| "learning_rate": 8.982300884955753e-06, | |
| "loss": 0.262, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.767441860465116, | |
| "grad_norm": 8.630583763122559, | |
| "learning_rate": 8.539823008849558e-06, | |
| "loss": 0.2658, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.883720930232558, | |
| "grad_norm": 11.391676902770996, | |
| "learning_rate": 8.097345132743364e-06, | |
| "loss": 0.3288, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 44.6805419921875, | |
| "learning_rate": 7.654867256637168e-06, | |
| "loss": 0.314, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7714285714285715, | |
| "eval_f1": 0.7714285714285715, | |
| "eval_loss": 0.43682464957237244, | |
| "eval_precision": 0.7714285714285715, | |
| "eval_recall": 0.7714285714285715, | |
| "eval_runtime": 38.7605, | |
| "eval_samples_per_second": 1.806, | |
| "eval_steps_per_second": 0.077, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.116279069767442, | |
| "grad_norm": 3.780876398086548, | |
| "learning_rate": 7.212389380530974e-06, | |
| "loss": 0.3121, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.232558139534884, | |
| "grad_norm": 4.6331000328063965, | |
| "learning_rate": 6.76991150442478e-06, | |
| "loss": 0.3032, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.348837209302325, | |
| "grad_norm": 3.838736057281494, | |
| "learning_rate": 6.3274336283185845e-06, | |
| "loss": 0.1971, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.465116279069767, | |
| "grad_norm": 7.0051469802856445, | |
| "learning_rate": 5.88495575221239e-06, | |
| "loss": 0.2727, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.5813953488372094, | |
| "grad_norm": 6.933923721313477, | |
| "learning_rate": 5.442477876106195e-06, | |
| "loss": 0.2669, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.6976744186046515, | |
| "grad_norm": 4.192832946777344, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2535, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.813953488372093, | |
| "grad_norm": 4.221975803375244, | |
| "learning_rate": 4.557522123893805e-06, | |
| "loss": 0.2963, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.930232558139535, | |
| "grad_norm": 7.972716331481934, | |
| "learning_rate": 4.115044247787611e-06, | |
| "loss": 0.2709, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8142857142857143, | |
| "eval_f1": 0.814553700267986, | |
| "eval_loss": 0.3767072856426239, | |
| "eval_precision": 0.8151550751879698, | |
| "eval_recall": 0.8142857142857143, | |
| "eval_runtime": 40.2015, | |
| "eval_samples_per_second": 1.741, | |
| "eval_steps_per_second": 0.075, | |
| "step": 516 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 602, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 2, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 2 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4137555984396360.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |