| { |
| "best_metric": 0.037108249962329865, |
| "best_model_checkpoint": "doc-topic-model_eval-03_train-04/checkpoint-13000", |
| "epoch": 8.871365204534253, |
| "eval_steps": 1000, |
| "global_step": 18000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2464268112370626, |
| "grad_norm": 0.36145079135894775, |
| "learning_rate": 1.9950714637752587e-05, |
| "loss": 0.1655, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4928536224741252, |
| "grad_norm": 0.3764462471008301, |
| "learning_rate": 1.990142927550518e-05, |
| "loss": 0.0934, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4928536224741252, |
| "eval_accuracy": 0.981439120143997, |
| "eval_f1": 0.0, |
| "eval_loss": 0.09006954729557037, |
| "eval_precision": 0.0, |
| "eval_recall": 0.0, |
| "eval_runtime": 12.2138, |
| "eval_samples_per_second": 664.003, |
| "eval_steps_per_second": 2.62, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.7392804337111878, |
| "grad_norm": 0.3353608548641205, |
| "learning_rate": 1.9852143913257764e-05, |
| "loss": 0.0875, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.9857072449482503, |
| "grad_norm": 0.3118976652622223, |
| "learning_rate": 1.980285855101035e-05, |
| "loss": 0.0778, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.9857072449482503, |
| "eval_accuracy": 0.981439120143997, |
| "eval_f1": 0.0, |
| "eval_loss": 0.07006971538066864, |
| "eval_precision": 0.0, |
| "eval_recall": 0.0, |
| "eval_runtime": 12.1184, |
| "eval_samples_per_second": 669.231, |
| "eval_steps_per_second": 2.641, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.232134056185313, |
| "grad_norm": 0.36931222677230835, |
| "learning_rate": 1.975357318876294e-05, |
| "loss": 0.0682, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.4785608674223756, |
| "grad_norm": 0.3769352436065674, |
| "learning_rate": 1.9704287826515527e-05, |
| "loss": 0.0618, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.4785608674223756, |
| "eval_accuracy": 0.9827784628031238, |
| "eval_f1": 0.16665523626637405, |
| "eval_loss": 0.056604694575071335, |
| "eval_precision": 0.8181818181818182, |
| "eval_recall": 0.09277642028100183, |
| "eval_runtime": 14.5143, |
| "eval_samples_per_second": 558.758, |
| "eval_steps_per_second": 2.205, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.724987678659438, |
| "grad_norm": 0.3530617356300354, |
| "learning_rate": 1.9655002464268113e-05, |
| "loss": 0.0565, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.9714144898965007, |
| "grad_norm": 0.3648615777492523, |
| "learning_rate": 1.96057171020207e-05, |
| "loss": 0.0535, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.9714144898965007, |
| "eval_accuracy": 0.9843729183497031, |
| "eval_f1": 0.3357831325301205, |
| "eval_loss": 0.04898000508546829, |
| "eval_precision": 0.7953767123287672, |
| "eval_recall": 0.21281307269395236, |
| "eval_runtime": 14.4304, |
| "eval_samples_per_second": 562.008, |
| "eval_steps_per_second": 2.218, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.2178413011335634, |
| "grad_norm": 0.4629921317100525, |
| "learning_rate": 1.955643173977329e-05, |
| "loss": 0.0479, |
| "step": 4500 |
| }, |
| { |
| "epoch": 2.464268112370626, |
| "grad_norm": 0.38679325580596924, |
| "learning_rate": 1.9507146377525875e-05, |
| "loss": 0.0473, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.464268112370626, |
| "eval_accuracy": 0.9854939977606758, |
| "eval_f1": 0.4636025365546879, |
| "eval_loss": 0.045349542051553726, |
| "eval_precision": 0.7390142021720969, |
| "eval_recall": 0.33773671350030543, |
| "eval_runtime": 14.4922, |
| "eval_samples_per_second": 559.61, |
| "eval_steps_per_second": 2.208, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.7106949236076883, |
| "grad_norm": 0.35379377007484436, |
| "learning_rate": 1.9457861015278464e-05, |
| "loss": 0.0442, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.957121734844751, |
| "grad_norm": 0.3872750997543335, |
| "learning_rate": 1.9408575653031053e-05, |
| "loss": 0.0436, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.957121734844751, |
| "eval_accuracy": 0.9860013889479428, |
| "eval_f1": 0.49589138978206504, |
| "eval_loss": 0.04238178953528404, |
| "eval_precision": 0.7477297214098815, |
| "eval_recall": 0.3709529627367135, |
| "eval_runtime": 14.4803, |
| "eval_samples_per_second": 560.073, |
| "eval_steps_per_second": 2.21, |
| "step": 6000 |
| }, |
| { |
| "epoch": 3.2035485460818136, |
| "grad_norm": 0.4312683939933777, |
| "learning_rate": 1.9359290290783638e-05, |
| "loss": 0.0399, |
| "step": 6500 |
| }, |
| { |
| "epoch": 3.449975357318876, |
| "grad_norm": 0.46780943870544434, |
| "learning_rate": 1.9310004928536227e-05, |
| "loss": 0.0389, |
| "step": 7000 |
| }, |
| { |
| "epoch": 3.449975357318876, |
| "eval_accuracy": 0.9864747650835495, |
| "eval_f1": 0.530156073063857, |
| "eval_loss": 0.04048691317439079, |
| "eval_precision": 0.7462231462231462, |
| "eval_recall": 0.4111178985949908, |
| "eval_runtime": 14.4385, |
| "eval_samples_per_second": 561.691, |
| "eval_steps_per_second": 2.216, |
| "step": 7000 |
| }, |
| { |
| "epoch": 3.696402168555939, |
| "grad_norm": 0.3296278417110443, |
| "learning_rate": 1.9260719566288815e-05, |
| "loss": 0.0379, |
| "step": 7500 |
| }, |
| { |
| "epoch": 3.9428289797930014, |
| "grad_norm": 0.22207264602184296, |
| "learning_rate": 1.92114342040414e-05, |
| "loss": 0.0376, |
| "step": 8000 |
| }, |
| { |
| "epoch": 3.9428289797930014, |
| "eval_accuracy": 0.9865031109599331, |
| "eval_f1": 0.5572962670261726, |
| "eval_loss": 0.0397622175514698, |
| "eval_precision": 0.7122994652406417, |
| "eval_recall": 0.45769700671960906, |
| "eval_runtime": 14.5013, |
| "eval_samples_per_second": 559.261, |
| "eval_steps_per_second": 2.207, |
| "step": 8000 |
| }, |
| { |
| "epoch": 4.189255791030064, |
| "grad_norm": 0.22761479020118713, |
| "learning_rate": 1.916214884179399e-05, |
| "loss": 0.034, |
| "step": 8500 |
| }, |
| { |
| "epoch": 4.435682602267127, |
| "grad_norm": 0.4061233699321747, |
| "learning_rate": 1.9112863479546578e-05, |
| "loss": 0.0339, |
| "step": 9000 |
| }, |
| { |
| "epoch": 4.435682602267127, |
| "eval_accuracy": 0.9867610584350242, |
| "eval_f1": 0.5566472067967155, |
| "eval_loss": 0.03901992738246918, |
| "eval_precision": 0.7354822526025335, |
| "eval_recall": 0.44777031154551006, |
| "eval_runtime": 14.7269, |
| "eval_samples_per_second": 550.694, |
| "eval_steps_per_second": 2.173, |
| "step": 9000 |
| }, |
| { |
| "epoch": 4.68210941350419, |
| "grad_norm": 0.31547975540161133, |
| "learning_rate": 1.9063578117299164e-05, |
| "loss": 0.0328, |
| "step": 9500 |
| }, |
| { |
| "epoch": 4.928536224741252, |
| "grad_norm": 0.307452529668808, |
| "learning_rate": 1.9014292755051752e-05, |
| "loss": 0.0337, |
| "step": 10000 |
| }, |
| { |
| "epoch": 4.928536224741252, |
| "eval_accuracy": 0.9871479796476608, |
| "eval_f1": 0.5749507827880379, |
| "eval_loss": 0.038141537457704544, |
| "eval_precision": 0.7444768147608642, |
| "eval_recall": 0.46831093463653023, |
| "eval_runtime": 14.4462, |
| "eval_samples_per_second": 561.395, |
| "eval_steps_per_second": 2.215, |
| "step": 10000 |
| }, |
| { |
| "epoch": 5.1749630359783145, |
| "grad_norm": 0.3740977942943573, |
| "learning_rate": 1.8965007392804338e-05, |
| "loss": 0.0303, |
| "step": 10500 |
| }, |
| { |
| "epoch": 5.421389847215377, |
| "grad_norm": 0.37522149085998535, |
| "learning_rate": 1.891582060128142e-05, |
| "loss": 0.0295, |
| "step": 11000 |
| }, |
| { |
| "epoch": 5.421389847215377, |
| "eval_accuracy": 0.9872840398543022, |
| "eval_f1": 0.6044091710758377, |
| "eval_loss": 0.037459347397089005, |
| "eval_precision": 0.7151502504173622, |
| "eval_recall": 0.523365913255956, |
| "eval_runtime": 14.4586, |
| "eval_samples_per_second": 560.913, |
| "eval_steps_per_second": 2.213, |
| "step": 11000 |
| }, |
| { |
| "epoch": 5.667816658452439, |
| "grad_norm": 0.3089052736759186, |
| "learning_rate": 1.886653523903401e-05, |
| "loss": 0.029, |
| "step": 11500 |
| }, |
| { |
| "epoch": 5.914243469689502, |
| "grad_norm": 0.3211393356323242, |
| "learning_rate": 1.8817249876786596e-05, |
| "loss": 0.0305, |
| "step": 12000 |
| }, |
| { |
| "epoch": 5.914243469689502, |
| "eval_accuracy": 0.9872145924571623, |
| "eval_f1": 0.5947621400655856, |
| "eval_loss": 0.03825980797410011, |
| "eval_precision": 0.7223131478450627, |
| "eval_recall": 0.5054978619425778, |
| "eval_runtime": 14.4529, |
| "eval_samples_per_second": 561.132, |
| "eval_steps_per_second": 2.214, |
| "step": 12000 |
| }, |
| { |
| "epoch": 6.160670280926565, |
| "grad_norm": 0.16942404210567474, |
| "learning_rate": 1.8767964514539182e-05, |
| "loss": 0.0275, |
| "step": 12500 |
| }, |
| { |
| "epoch": 6.407097092163627, |
| "grad_norm": 0.3766990303993225, |
| "learning_rate": 1.8718777723016266e-05, |
| "loss": 0.0254, |
| "step": 13000 |
| }, |
| { |
| "epoch": 6.407097092163627, |
| "eval_accuracy": 0.9874838782828068, |
| "eval_f1": 0.6108834545053977, |
| "eval_loss": 0.037108249962329865, |
| "eval_precision": 0.7221585581831441, |
| "eval_recall": 0.5293219303604154, |
| "eval_runtime": 14.4405, |
| "eval_samples_per_second": 561.615, |
| "eval_steps_per_second": 2.216, |
| "step": 13000 |
| }, |
| { |
| "epoch": 6.65352390340069, |
| "grad_norm": 0.5648388266563416, |
| "learning_rate": 1.8669492360768852e-05, |
| "loss": 0.0264, |
| "step": 13500 |
| }, |
| { |
| "epoch": 6.899950714637752, |
| "grad_norm": 0.4413442015647888, |
| "learning_rate": 1.862020699852144e-05, |
| "loss": 0.0273, |
| "step": 14000 |
| }, |
| { |
| "epoch": 6.899950714637752, |
| "eval_accuracy": 0.9876822994174922, |
| "eval_f1": 0.6180619644034279, |
| "eval_loss": 0.037435684353113174, |
| "eval_precision": 0.7280256755357698, |
| "eval_recall": 0.5369578497251069, |
| "eval_runtime": 14.6502, |
| "eval_samples_per_second": 553.575, |
| "eval_steps_per_second": 2.184, |
| "step": 14000 |
| }, |
| { |
| "epoch": 7.146377525874815, |
| "grad_norm": 0.36695852875709534, |
| "learning_rate": 1.857092163627403e-05, |
| "loss": 0.0234, |
| "step": 14500 |
| }, |
| { |
| "epoch": 7.392804337111878, |
| "grad_norm": 0.36567145586013794, |
| "learning_rate": 1.852173484475111e-05, |
| "loss": 0.0228, |
| "step": 15000 |
| }, |
| { |
| "epoch": 7.392804337111878, |
| "eval_accuracy": 0.9875632467366809, |
| "eval_f1": 0.6112956810631229, |
| "eval_loss": 0.0375823937356472, |
| "eval_precision": 0.7279248865914126, |
| "eval_recall": 0.5268784361637141, |
| "eval_runtime": 14.4327, |
| "eval_samples_per_second": 561.917, |
| "eval_steps_per_second": 2.217, |
| "step": 15000 |
| }, |
| { |
| "epoch": 7.639231148348941, |
| "grad_norm": 0.5829640626907349, |
| "learning_rate": 1.84724494825037e-05, |
| "loss": 0.0236, |
| "step": 15500 |
| }, |
| { |
| "epoch": 7.885657959586003, |
| "grad_norm": 0.470248281955719, |
| "learning_rate": 1.8423164120256285e-05, |
| "loss": 0.0235, |
| "step": 16000 |
| }, |
| { |
| "epoch": 7.885657959586003, |
| "eval_accuracy": 0.9873634083081764, |
| "eval_f1": 0.6297034637428358, |
| "eval_loss": 0.037557560950517654, |
| "eval_precision": 0.6903114186851211, |
| "eval_recall": 0.5788790470372633, |
| "eval_runtime": 14.5611, |
| "eval_samples_per_second": 556.963, |
| "eval_steps_per_second": 2.198, |
| "step": 16000 |
| }, |
| { |
| "epoch": 8.132084770823065, |
| "grad_norm": 0.48524007201194763, |
| "learning_rate": 1.8373878758008873e-05, |
| "loss": 0.0217, |
| "step": 16500 |
| }, |
| { |
| "epoch": 8.378511582060128, |
| "grad_norm": 0.3671877384185791, |
| "learning_rate": 1.8324593395761462e-05, |
| "loss": 0.0208, |
| "step": 17000 |
| }, |
| { |
| "epoch": 8.378511582060128, |
| "eval_accuracy": 0.9876043482574373, |
| "eval_f1": 0.6323047170604558, |
| "eval_loss": 0.03765055537223816, |
| "eval_precision": 0.7034611786716558, |
| "eval_recall": 0.5742211362248014, |
| "eval_runtime": 14.6933, |
| "eval_samples_per_second": 551.952, |
| "eval_steps_per_second": 2.178, |
| "step": 17000 |
| }, |
| { |
| "epoch": 8.62493839329719, |
| "grad_norm": 0.2823368012905121, |
| "learning_rate": 1.8275308033514047e-05, |
| "loss": 0.0202, |
| "step": 17500 |
| }, |
| { |
| "epoch": 8.871365204534253, |
| "grad_norm": 0.374859094619751, |
| "learning_rate": 1.8226121241991132e-05, |
| "loss": 0.0204, |
| "step": 18000 |
| }, |
| { |
| "epoch": 8.871365204534253, |
| "eval_accuracy": 0.9877517468146322, |
| "eval_f1": 0.6369212671204101, |
| "eval_loss": 0.03813798725605011, |
| "eval_precision": 0.7080141976461797, |
| "eval_recall": 0.5788026878436163, |
| "eval_runtime": 14.6486, |
| "eval_samples_per_second": 553.637, |
| "eval_steps_per_second": 2.185, |
| "step": 18000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 202900, |
| "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": 408291319078542.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
