| { | |
| "best_metric": 0.8943646614708781, | |
| "best_model_checkpoint": "cls_comment-phobert-base-v2-v2.3/checkpoint-1600", | |
| "epoch": 21.875, | |
| "eval_steps": 100, | |
| "global_step": 2100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 2.3460028171539307, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.1884, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_accuracy": 0.7277179236043095, | |
| "eval_f1_score": 0.43708201591573076, | |
| "eval_loss": 1.0905314683914185, | |
| "eval_precision": 0.5016943776884798, | |
| "eval_recall": 0.42288884208445504, | |
| "eval_runtime": 4.9336, | |
| "eval_samples_per_second": 620.848, | |
| "eval_steps_per_second": 9.729, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 9.137602806091309, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0502, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_accuracy": 0.7956252040483186, | |
| "eval_f1_score": 0.5245336258309143, | |
| "eval_loss": 0.9278329014778137, | |
| "eval_precision": 0.5244123326675032, | |
| "eval_recall": 0.5273852871687749, | |
| "eval_runtime": 4.9622, | |
| "eval_samples_per_second": 617.266, | |
| "eval_steps_per_second": 9.673, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 8.277727127075195, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.9031, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.8445968005223637, | |
| "eval_f1_score": 0.5653211085059351, | |
| "eval_loss": 0.7793412208557129, | |
| "eval_precision": 0.7067057672713553, | |
| "eval_recall": 0.589697230912186, | |
| "eval_runtime": 5.0159, | |
| "eval_samples_per_second": 610.664, | |
| "eval_steps_per_second": 9.57, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 3.594104290008545, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7772, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_accuracy": 0.8697355533790402, | |
| "eval_f1_score": 0.6917446533625714, | |
| "eval_loss": 0.7089904546737671, | |
| "eval_precision": 0.7268852070004593, | |
| "eval_recall": 0.6804440791993415, | |
| "eval_runtime": 4.9399, | |
| "eval_samples_per_second": 620.058, | |
| "eval_steps_per_second": 9.717, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "grad_norm": 4.503917694091797, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.6741, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "eval_accuracy": 0.8958537381651975, | |
| "eval_f1_score": 0.8078466525990172, | |
| "eval_loss": 0.648346483707428, | |
| "eval_precision": 0.8797678600731677, | |
| "eval_recall": 0.7817903123125977, | |
| "eval_runtime": 4.9504, | |
| "eval_samples_per_second": 618.741, | |
| "eval_steps_per_second": 9.696, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 4.407541275024414, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.6156, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_accuracy": 0.9079333986287953, | |
| "eval_f1_score": 0.8601153974079928, | |
| "eval_loss": 0.6268225908279419, | |
| "eval_precision": 0.8933059224979765, | |
| "eval_recall": 0.8344643413921417, | |
| "eval_runtime": 4.9864, | |
| "eval_samples_per_second": 614.266, | |
| "eval_steps_per_second": 9.626, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "grad_norm": 5.013394355773926, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.5745, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "eval_accuracy": 0.9160953313744695, | |
| "eval_f1_score": 0.8738968846084525, | |
| "eval_loss": 0.6176044940948486, | |
| "eval_precision": 0.8880590844686177, | |
| "eval_recall": 0.8609982161968145, | |
| "eval_runtime": 5.0009, | |
| "eval_samples_per_second": 612.484, | |
| "eval_steps_per_second": 9.598, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "grad_norm": 4.954068183898926, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.5466, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_accuracy": 0.9180541952334312, | |
| "eval_f1_score": 0.8730278944900508, | |
| "eval_loss": 0.613418698310852, | |
| "eval_precision": 0.8951285463624247, | |
| "eval_recall": 0.8555146035099561, | |
| "eval_runtime": 4.9947, | |
| "eval_samples_per_second": 613.247, | |
| "eval_steps_per_second": 9.61, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "grad_norm": 6.174578666687012, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.5254, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "eval_accuracy": 0.9183806725432583, | |
| "eval_f1_score": 0.8794225212758101, | |
| "eval_loss": 0.6195659637451172, | |
| "eval_precision": 0.898010742334885, | |
| "eval_recall": 0.8634351648847699, | |
| "eval_runtime": 4.9765, | |
| "eval_samples_per_second": 615.487, | |
| "eval_steps_per_second": 9.645, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "grad_norm": 5.367175102233887, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.5059, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "eval_accuracy": 0.9219719229513549, | |
| "eval_f1_score": 0.8799283123746076, | |
| "eval_loss": 0.6174536347389221, | |
| "eval_precision": 0.8925411600987329, | |
| "eval_recall": 0.8702271946202808, | |
| "eval_runtime": 4.9764, | |
| "eval_samples_per_second": 615.501, | |
| "eval_steps_per_second": 9.645, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "grad_norm": 6.586447238922119, | |
| "learning_rate": 8.055555555555557e-06, | |
| "loss": 0.4971, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "eval_accuracy": 0.9226248775710089, | |
| "eval_f1_score": 0.8770727386787929, | |
| "eval_loss": 0.6110312342643738, | |
| "eval_precision": 0.89290929483548, | |
| "eval_recall": 0.8638757728594787, | |
| "eval_runtime": 4.9931, | |
| "eval_samples_per_second": 613.443, | |
| "eval_steps_per_second": 9.613, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 3.8108513355255127, | |
| "learning_rate": 7.77777777777778e-06, | |
| "loss": 0.4872, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_accuracy": 0.9229513548808358, | |
| "eval_f1_score": 0.8833538508725888, | |
| "eval_loss": 0.6191003918647766, | |
| "eval_precision": 0.8914104564156814, | |
| "eval_recall": 0.8774149435732262, | |
| "eval_runtime": 4.9594, | |
| "eval_samples_per_second": 617.612, | |
| "eval_steps_per_second": 9.679, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "grad_norm": 2.2206079959869385, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.4756, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "eval_accuracy": 0.9242572641201436, | |
| "eval_f1_score": 0.892919420274636, | |
| "eval_loss": 0.6239792704582214, | |
| "eval_precision": 0.8938258370033757, | |
| "eval_recall": 0.8931296696096861, | |
| "eval_runtime": 4.9692, | |
| "eval_samples_per_second": 616.403, | |
| "eval_steps_per_second": 9.66, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "grad_norm": 3.2590479850769043, | |
| "learning_rate": 7.222222222222223e-06, | |
| "loss": 0.4737, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "eval_accuracy": 0.92033953640222, | |
| "eval_f1_score": 0.8793750306978456, | |
| "eval_loss": 0.6246046423912048, | |
| "eval_precision": 0.9038460171417751, | |
| "eval_recall": 0.8582419368732399, | |
| "eval_runtime": 5.0236, | |
| "eval_samples_per_second": 609.724, | |
| "eval_steps_per_second": 9.555, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "grad_norm": 4.4676127433776855, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.4626, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "eval_accuracy": 0.9249102187397976, | |
| "eval_f1_score": 0.8889732013227639, | |
| "eval_loss": 0.6267126798629761, | |
| "eval_precision": 0.8955131575427445, | |
| "eval_recall": 0.8842202196754521, | |
| "eval_runtime": 5.0075, | |
| "eval_samples_per_second": 611.678, | |
| "eval_steps_per_second": 9.586, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "grad_norm": 1.5825713872909546, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.4641, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "eval_accuracy": 0.9271955599085864, | |
| "eval_f1_score": 0.8943646614708781, | |
| "eval_loss": 0.6228047013282776, | |
| "eval_precision": 0.9063376648453086, | |
| "eval_recall": 0.8842502417202245, | |
| "eval_runtime": 4.9845, | |
| "eval_samples_per_second": 614.503, | |
| "eval_steps_per_second": 9.63, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.71, | |
| "grad_norm": 3.3820607662200928, | |
| "learning_rate": 6.3888888888888885e-06, | |
| "loss": 0.4562, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 17.71, | |
| "eval_accuracy": 0.9278485145282402, | |
| "eval_f1_score": 0.892329553955837, | |
| "eval_loss": 0.6255541443824768, | |
| "eval_precision": 0.9064974712840742, | |
| "eval_recall": 0.8800313461131938, | |
| "eval_runtime": 5.0301, | |
| "eval_samples_per_second": 608.937, | |
| "eval_steps_per_second": 9.543, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 0.6632604598999023, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 0.4522, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "eval_accuracy": 0.9285014691478942, | |
| "eval_f1_score": 0.8912735294131965, | |
| "eval_loss": 0.6203188300132751, | |
| "eval_precision": 0.8992701751319325, | |
| "eval_recall": 0.8844935194666589, | |
| "eval_runtime": 5.0719, | |
| "eval_samples_per_second": 603.913, | |
| "eval_steps_per_second": 9.464, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "grad_norm": 1.0484055280685425, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.4476, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "eval_accuracy": 0.9262161279791055, | |
| "eval_f1_score": 0.8848887104133892, | |
| "eval_loss": 0.6258328557014465, | |
| "eval_precision": 0.9033804413868713, | |
| "eval_recall": 0.8691664648450299, | |
| "eval_runtime": 5.0588, | |
| "eval_samples_per_second": 605.485, | |
| "eval_steps_per_second": 9.488, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 20.83, | |
| "grad_norm": 4.217446804046631, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.4474, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 20.83, | |
| "eval_accuracy": 0.9271955599085864, | |
| "eval_f1_score": 0.8931632926835013, | |
| "eval_loss": 0.6333132386207581, | |
| "eval_precision": 0.9133975869931371, | |
| "eval_recall": 0.8751018475655369, | |
| "eval_runtime": 4.9412, | |
| "eval_samples_per_second": 619.889, | |
| "eval_steps_per_second": 9.714, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "grad_norm": 2.5882654190063477, | |
| "learning_rate": 5.2777777777777785e-06, | |
| "loss": 0.4421, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "eval_accuracy": 0.9291544237675482, | |
| "eval_f1_score": 0.8929419711474625, | |
| "eval_loss": 0.6339655518531799, | |
| "eval_precision": 0.8975107075254134, | |
| "eval_recall": 0.8896075654961032, | |
| "eval_runtime": 4.9672, | |
| "eval_samples_per_second": 616.645, | |
| "eval_steps_per_second": 9.663, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "step": 2100, | |
| "total_flos": 4711306254898320.0, | |
| "train_loss": 0.5936482783726283, | |
| "train_runtime": 1811.9756, | |
| "train_samples_per_second": 282.565, | |
| "train_steps_per_second": 2.208 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 42, | |
| "save_steps": 100, | |
| "total_flos": 4711306254898320.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |