| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7648299932479858, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.1483, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8448836627470603, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6899715065956116, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.8894, | |
| "eval_samples_per_second": 210.248, | |
| "eval_steps_per_second": 3.373, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.372132420539856, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6875, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8463847885914436, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.5737118721008301, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.926, | |
| "eval_samples_per_second": 201.946, | |
| "eval_steps_per_second": 3.24, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2157526016235352, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.5874, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8633975481611208, | |
| "eval_f1": 0.10071942446043167, | |
| "eval_loss": 0.4661254584789276, | |
| "eval_precision": 0.2692307692307692, | |
| "eval_recall": 0.061946902654867256, | |
| "eval_runtime": 0.9773, | |
| "eval_samples_per_second": 191.347, | |
| "eval_steps_per_second": 3.07, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.093184232711792, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4729, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.898173630222667, | |
| "eval_f1": 0.37722419928825623, | |
| "eval_loss": 0.3599094748497009, | |
| "eval_precision": 0.47533632286995514, | |
| "eval_recall": 0.31268436578171094, | |
| "eval_runtime": 0.9352, | |
| "eval_samples_per_second": 199.961, | |
| "eval_steps_per_second": 3.208, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.716677665710449, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3692, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9246935201401051, | |
| "eval_f1": 0.5917496443812232, | |
| "eval_loss": 0.29404327273368835, | |
| "eval_precision": 0.5714285714285714, | |
| "eval_recall": 0.6135693215339233, | |
| "eval_runtime": 0.899, | |
| "eval_samples_per_second": 208.018, | |
| "eval_steps_per_second": 3.337, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.1444728374481201, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3058, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9334500875656743, | |
| "eval_f1": 0.6621621621621622, | |
| "eval_loss": 0.2527407705783844, | |
| "eval_precision": 0.6109725685785536, | |
| "eval_recall": 0.7227138643067846, | |
| "eval_runtime": 0.8914, | |
| "eval_samples_per_second": 209.792, | |
| "eval_steps_per_second": 3.366, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.998134434223175, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2636, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.937453089817363, | |
| "eval_f1": 0.6954177897574124, | |
| "eval_loss": 0.22462092339992523, | |
| "eval_precision": 0.6401985111662531, | |
| "eval_recall": 0.7610619469026548, | |
| "eval_runtime": 0.9047, | |
| "eval_samples_per_second": 206.701, | |
| "eval_steps_per_second": 3.316, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.480947732925415, | |
| "learning_rate": 3e-05, | |
| "loss": 0.24, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9417062797097824, | |
| "eval_f1": 0.7241379310344828, | |
| "eval_loss": 0.20909550786018372, | |
| "eval_precision": 0.6578313253012048, | |
| "eval_recall": 0.8053097345132744, | |
| "eval_runtime": 0.9189, | |
| "eval_samples_per_second": 203.503, | |
| "eval_steps_per_second": 3.265, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.5677202939987183, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2228, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.940205153865399, | |
| "eval_f1": 0.716883116883117, | |
| "eval_loss": 0.1985715925693512, | |
| "eval_precision": 0.6403712296983759, | |
| "eval_recall": 0.8141592920353983, | |
| "eval_runtime": 0.9225, | |
| "eval_samples_per_second": 202.716, | |
| "eval_steps_per_second": 3.252, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.242694854736328, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2105, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9417062797097824, | |
| "eval_f1": 0.733245729303548, | |
| "eval_loss": 0.1821284145116806, | |
| "eval_precision": 0.6611374407582938, | |
| "eval_recall": 0.8230088495575221, | |
| "eval_runtime": 0.925, | |
| "eval_samples_per_second": 202.159, | |
| "eval_steps_per_second": 3.243, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.5337995290756226, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2007, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9432074055541656, | |
| "eval_f1": 0.7394736842105263, | |
| "eval_loss": 0.17939399182796478, | |
| "eval_precision": 0.667458432304038, | |
| "eval_recall": 0.8289085545722714, | |
| "eval_runtime": 0.9211, | |
| "eval_samples_per_second": 203.028, | |
| "eval_steps_per_second": 3.257, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.7884016036987305, | |
| "learning_rate": 2e-05, | |
| "loss": 0.195, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9429572179134351, | |
| "eval_f1": 0.7392996108949417, | |
| "eval_loss": 0.18079817295074463, | |
| "eval_precision": 0.6597222222222222, | |
| "eval_recall": 0.8407079646017699, | |
| "eval_runtime": 0.9116, | |
| "eval_samples_per_second": 205.139, | |
| "eval_steps_per_second": 3.291, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.4956753253936768, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.19, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9459594696022017, | |
| "eval_f1": 0.7463479415670652, | |
| "eval_loss": 0.16896921396255493, | |
| "eval_precision": 0.678743961352657, | |
| "eval_recall": 0.8289085545722714, | |
| "eval_runtime": 0.9045, | |
| "eval_samples_per_second": 206.741, | |
| "eval_steps_per_second": 3.317, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.4634425640106201, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1835, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9477107830873155, | |
| "eval_f1": 0.751336898395722, | |
| "eval_loss": 0.16314250230789185, | |
| "eval_precision": 0.687041564792176, | |
| "eval_recall": 0.8289085545722714, | |
| "eval_runtime": 0.9091, | |
| "eval_samples_per_second": 205.694, | |
| "eval_steps_per_second": 3.3, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.0593953132629395, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1821, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9472104078058544, | |
| "eval_f1": 0.753968253968254, | |
| "eval_loss": 0.16711552441120148, | |
| "eval_precision": 0.6834532374100719, | |
| "eval_recall": 0.8407079646017699, | |
| "eval_runtime": 0.9129, | |
| "eval_samples_per_second": 204.846, | |
| "eval_steps_per_second": 3.286, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.1103074550628662, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1774, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9472104078058544, | |
| "eval_f1": 0.7647831800262812, | |
| "eval_loss": 0.16675202548503876, | |
| "eval_precision": 0.6895734597156398, | |
| "eval_recall": 0.8584070796460177, | |
| "eval_runtime": 0.913, | |
| "eval_samples_per_second": 204.819, | |
| "eval_steps_per_second": 3.286, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.5177829265594482, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1764, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9477107830873155, | |
| "eval_f1": 0.7602649006622517, | |
| "eval_loss": 0.16347847878932953, | |
| "eval_precision": 0.6899038461538461, | |
| "eval_recall": 0.8466076696165191, | |
| "eval_runtime": 0.9243, | |
| "eval_samples_per_second": 202.326, | |
| "eval_steps_per_second": 3.246, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.8699701428413391, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1729, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9472104078058544, | |
| "eval_f1": 0.7611548556430445, | |
| "eval_loss": 0.16536261141300201, | |
| "eval_precision": 0.6855791962174941, | |
| "eval_recall": 0.855457227138643, | |
| "eval_runtime": 0.9101, | |
| "eval_samples_per_second": 205.482, | |
| "eval_steps_per_second": 3.297, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.786510705947876, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1726, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9477107830873155, | |
| "eval_f1": 0.7621550591327202, | |
| "eval_loss": 0.16280074417591095, | |
| "eval_precision": 0.6872037914691943, | |
| "eval_recall": 0.855457227138643, | |
| "eval_runtime": 0.9163, | |
| "eval_samples_per_second": 204.074, | |
| "eval_steps_per_second": 3.274, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.0649765729904175, | |
| "learning_rate": 0.0, | |
| "loss": 0.1684, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9472104078058544, | |
| "eval_f1": 0.7595269382391592, | |
| "eval_loss": 0.16256052255630493, | |
| "eval_precision": 0.6848341232227488, | |
| "eval_recall": 0.8525073746312685, | |
| "eval_runtime": 0.9134, | |
| "eval_samples_per_second": 204.735, | |
| "eval_steps_per_second": 3.285, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2120, | |
| "total_flos": 904262544144960.0, | |
| "train_loss": 0.3163445589677343, | |
| "train_runtime": 244.3626, | |
| "train_samples_per_second": 138.074, | |
| "train_steps_per_second": 8.676 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 904262544144960.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |