| { | |
| "best_metric": 0.03750571236014366, | |
| "best_model_checkpoint": "doc-topic-model_eval-02_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": 0.9813654208653996, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.09037782996892929, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.0127, | |
| "eval_samples_per_second": 675.12, | |
| "eval_steps_per_second": 2.664, | |
| "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": 0.9813654208653996, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.07024028897285461, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 12.1524, | |
| "eval_samples_per_second": 667.356, | |
| "eval_steps_per_second": 2.633, | |
| "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": 0.9827912184474964, | |
| "eval_f1": 0.1751358695652174, | |
| "eval_loss": 0.05670051649212837, | |
| "eval_precision": 0.8199745547073791, | |
| "eval_recall": 0.09803772436872528, | |
| "eval_runtime": 14.2669, | |
| "eval_samples_per_second": 568.448, | |
| "eval_steps_per_second": 2.243, | |
| "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": 0.9842510310812534, | |
| "eval_f1": 0.33580394500896593, | |
| "eval_loss": 0.049183741211891174, | |
| "eval_precision": 0.7841987716359575, | |
| "eval_recall": 0.21364466078491026, | |
| "eval_runtime": 14.4394, | |
| "eval_samples_per_second": 561.657, | |
| "eval_steps_per_second": 2.216, | |
| "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": 0.9854925804668566, | |
| "eval_f1": 0.46681946036045424, | |
| "eval_loss": 0.045466337352991104, | |
| "eval_precision": 0.7406611570247934, | |
| "eval_recall": 0.3408122908427137, | |
| "eval_runtime": 14.1723, | |
| "eval_samples_per_second": 572.243, | |
| "eval_steps_per_second": 2.258, | |
| "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": 0.9860694190512636, | |
| "eval_f1": 0.5016478223394008, | |
| "eval_loss": 0.042579714208841324, | |
| "eval_precision": 0.7523954372623575, | |
| "eval_recall": 0.3762549437176757, | |
| "eval_runtime": 14.177, | |
| "eval_samples_per_second": 572.053, | |
| "eval_steps_per_second": 2.257, | |
| "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": 0.9865073628413906, | |
| "eval_f1": 0.5336533751347114, | |
| "eval_loss": 0.04068835824728012, | |
| "eval_precision": 0.7496559317368566, | |
| "eval_recall": 0.4142835412229997, | |
| "eval_runtime": 14.2131, | |
| "eval_samples_per_second": 570.6, | |
| "eval_steps_per_second": 2.251, | |
| "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": 0.986619329053106, | |
| "eval_f1": 0.563381584423993, | |
| "eval_loss": 0.03988400474190712, | |
| "eval_precision": 0.7187020648967551, | |
| "eval_recall": 0.463264374809857, | |
| "eval_runtime": 14.2691, | |
| "eval_samples_per_second": 568.36, | |
| "eval_steps_per_second": 2.243, | |
| "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": 0.9869708179202631, | |
| "eval_f1": 0.5653222374580359, | |
| "eval_loss": 0.038919780403375626, | |
| "eval_precision": 0.7471566054243219, | |
| "eval_recall": 0.4546699117736538, | |
| "eval_runtime": 14.2717, | |
| "eval_samples_per_second": 568.258, | |
| "eval_steps_per_second": 2.242, | |
| "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": 0.9873194721997818, | |
| "eval_f1": 0.581505215398288, | |
| "eval_loss": 0.038509998470544815, | |
| "eval_precision": 0.7551937796136557, | |
| "eval_recall": 0.472771524186188, | |
| "eval_runtime": 14.3086, | |
| "eval_samples_per_second": 566.79, | |
| "eval_steps_per_second": 2.236, | |
| "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": 0.9872060886942472, | |
| "eval_f1": 0.6024223739264479, | |
| "eval_loss": 0.03767974302172661, | |
| "eval_precision": 0.7156011300617349, | |
| "eval_recall": 0.5201551566778218, | |
| "eval_runtime": 14.3451, | |
| "eval_samples_per_second": 565.349, | |
| "eval_steps_per_second": 2.231, | |
| "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": 0.9873534872514421, | |
| "eval_f1": 0.5992364697956434, | |
| "eval_loss": 0.03828778117895126, | |
| "eval_precision": 0.7317099923220357, | |
| "eval_recall": 0.5073775479160328, | |
| "eval_runtime": 14.3412, | |
| "eval_samples_per_second": 565.503, | |
| "eval_steps_per_second": 2.231, | |
| "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": 0.9875646640305001, | |
| "eval_f1": 0.6140921885995778, | |
| "eval_loss": 0.03750571236014366, | |
| "eval_precision": 0.7280976220275344, | |
| "eval_recall": 0.5309552783693338, | |
| "eval_runtime": 14.3059, | |
| "eval_samples_per_second": 566.897, | |
| "eval_steps_per_second": 2.237, | |
| "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": 0.9876582054225661, | |
| "eval_f1": 0.6163200563976031, | |
| "eval_loss": 0.03790096566081047, | |
| "eval_precision": 0.7325094260578131, | |
| "eval_recall": 0.5319440219044722, | |
| "eval_runtime": 14.1405, | |
| "eval_samples_per_second": 573.531, | |
| "eval_steps_per_second": 2.263, | |
| "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": 0.9877120625876951, | |
| "eval_f1": 0.6165074309978769, | |
| "eval_loss": 0.03786341845989227, | |
| "eval_precision": 0.7366807610993658, | |
| "eval_recall": 0.530042592029206, | |
| "eval_runtime": 14.1212, | |
| "eval_samples_per_second": 574.314, | |
| "eval_steps_per_second": 2.266, | |
| "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": 0.9873563218390805, | |
| "eval_f1": 0.629849383842994, | |
| "eval_loss": 0.03791099041700363, | |
| "eval_precision": 0.6929608326485894, | |
| "eval_recall": 0.5772741101308184, | |
| "eval_runtime": 14.1983, | |
| "eval_samples_per_second": 571.194, | |
| "eval_steps_per_second": 2.254, | |
| "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": 0.9877219836444293, | |
| "eval_f1": 0.6341484015372271, | |
| "eval_loss": 0.037882279604673386, | |
| "eval_precision": 0.7129427404804862, | |
| "eval_recall": 0.5710374201399452, | |
| "eval_runtime": 14.1368, | |
| "eval_samples_per_second": 573.681, | |
| "eval_steps_per_second": 2.264, | |
| "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": 0.987852374675794, | |
| "eval_f1": 0.6380948359582823, | |
| "eval_loss": 0.03811024874448776, | |
| "eval_precision": 0.7172282866635026, | |
| "eval_recall": 0.5746881655004563, | |
| "eval_runtime": 14.2361, | |
| "eval_samples_per_second": 569.679, | |
| "eval_steps_per_second": 2.248, | |
| "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 | |
| } | |