| { | |
| "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": 1.4620718955993652, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.2537, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8365739601279842, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.7370312809944153, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 1.0987, | |
| "eval_samples_per_second": 170.195, | |
| "eval_steps_per_second": 2.73, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.4634525775909424, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7093, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8373123307900566, | |
| "eval_f1": 0.0056022408963585435, | |
| "eval_loss": 0.6298087239265442, | |
| "eval_precision": 0.16666666666666666, | |
| "eval_recall": 0.002849002849002849, | |
| "eval_runtime": 0.9814, | |
| "eval_samples_per_second": 190.541, | |
| "eval_steps_per_second": 3.057, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2972607612609863, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.6232, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8417425547624908, | |
| "eval_f1": 0.032171581769437, | |
| "eval_loss": 0.5443252325057983, | |
| "eval_precision": 0.2727272727272727, | |
| "eval_recall": 0.017094017094017096, | |
| "eval_runtime": 1.0287, | |
| "eval_samples_per_second": 181.785, | |
| "eval_steps_per_second": 2.916, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.2424334287643433, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5363, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8624169333005168, | |
| "eval_f1": 0.14977973568281938, | |
| "eval_loss": 0.45594143867492676, | |
| "eval_precision": 0.3300970873786408, | |
| "eval_recall": 0.09686609686609686, | |
| "eval_runtime": 0.9669, | |
| "eval_samples_per_second": 193.397, | |
| "eval_steps_per_second": 3.103, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.3577224016189575, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4591, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8929362539995077, | |
| "eval_f1": 0.3604240282685513, | |
| "eval_loss": 0.3863191306591034, | |
| "eval_precision": 0.4744186046511628, | |
| "eval_recall": 0.2905982905982906, | |
| "eval_runtime": 0.9477, | |
| "eval_samples_per_second": 197.315, | |
| "eval_steps_per_second": 3.165, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2191588878631592, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.387, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9190253507260645, | |
| "eval_f1": 0.5714285714285715, | |
| "eval_loss": 0.32724133133888245, | |
| "eval_precision": 0.5789473684210527, | |
| "eval_recall": 0.5641025641025641, | |
| "eval_runtime": 0.9315, | |
| "eval_samples_per_second": 200.744, | |
| "eval_steps_per_second": 3.22, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.1846706867218018, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3252, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9291164164410534, | |
| "eval_f1": 0.6436170212765958, | |
| "eval_loss": 0.2810536026954651, | |
| "eval_precision": 0.6034912718204489, | |
| "eval_recall": 0.6894586894586895, | |
| "eval_runtime": 0.9504, | |
| "eval_samples_per_second": 196.764, | |
| "eval_steps_per_second": 3.157, | |
| "step": 742 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.0833159685134888, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2874, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9313315284272705, | |
| "eval_f1": 0.655217965653897, | |
| "eval_loss": 0.24546079337596893, | |
| "eval_precision": 0.6108374384236454, | |
| "eval_recall": 0.7065527065527065, | |
| "eval_runtime": 0.9612, | |
| "eval_samples_per_second": 194.541, | |
| "eval_steps_per_second": 3.121, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.9407267570495605, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2588, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9333005168594635, | |
| "eval_f1": 0.6761290322580644, | |
| "eval_loss": 0.22847984731197357, | |
| "eval_precision": 0.6179245283018868, | |
| "eval_recall": 0.7464387464387464, | |
| "eval_runtime": 0.9716, | |
| "eval_samples_per_second": 192.466, | |
| "eval_steps_per_second": 3.088, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.7366420030593872, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2393, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9362539995077529, | |
| "eval_f1": 0.6975546975546976, | |
| "eval_loss": 0.21532489359378815, | |
| "eval_precision": 0.636150234741784, | |
| "eval_recall": 0.7720797720797721, | |
| "eval_runtime": 0.9655, | |
| "eval_samples_per_second": 193.678, | |
| "eval_steps_per_second": 3.107, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.1416951417922974, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.224, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9387152350479941, | |
| "eval_f1": 0.7030456852791879, | |
| "eval_loss": 0.2062499076128006, | |
| "eval_precision": 0.6338672768878718, | |
| "eval_recall": 0.7891737891737892, | |
| "eval_runtime": 0.9619, | |
| "eval_samples_per_second": 194.401, | |
| "eval_steps_per_second": 3.119, | |
| "step": 1166 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.7414947748184204, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2137, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9387152350479941, | |
| "eval_f1": 0.7135549872122762, | |
| "eval_loss": 0.20024912059307098, | |
| "eval_precision": 0.6473317865429234, | |
| "eval_recall": 0.7948717948717948, | |
| "eval_runtime": 0.9544, | |
| "eval_samples_per_second": 195.928, | |
| "eval_steps_per_second": 3.143, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.186489224433899, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2052, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9424070883583558, | |
| "eval_f1": 0.7218628719275549, | |
| "eval_loss": 0.18892288208007812, | |
| "eval_precision": 0.6611374407582938, | |
| "eval_recall": 0.7948717948717948, | |
| "eval_runtime": 0.9502, | |
| "eval_samples_per_second": 196.793, | |
| "eval_steps_per_second": 3.157, | |
| "step": 1378 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.4163442850112915, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2039, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9431454590204282, | |
| "eval_f1": 0.7312661498708009, | |
| "eval_loss": 0.18623687326908112, | |
| "eval_precision": 0.6690307328605201, | |
| "eval_recall": 0.8062678062678063, | |
| "eval_runtime": 0.9553, | |
| "eval_samples_per_second": 195.753, | |
| "eval_steps_per_second": 3.14, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.5289697647094727, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1975, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9431454590204282, | |
| "eval_f1": 0.7319587628865979, | |
| "eval_loss": 0.1867983341217041, | |
| "eval_precision": 0.668235294117647, | |
| "eval_recall": 0.8091168091168092, | |
| "eval_runtime": 0.9597, | |
| "eval_samples_per_second": 194.853, | |
| "eval_steps_per_second": 3.126, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.371168375015259, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1936, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.94265321191238, | |
| "eval_f1": 0.7321428571428572, | |
| "eval_loss": 0.18374690413475037, | |
| "eval_precision": 0.6628175519630485, | |
| "eval_recall": 0.8176638176638177, | |
| "eval_runtime": 0.9485, | |
| "eval_samples_per_second": 197.152, | |
| "eval_steps_per_second": 3.163, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.7330523133277893, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1908, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.94265321191238, | |
| "eval_f1": 0.7305236270753512, | |
| "eval_loss": 0.18250302970409393, | |
| "eval_precision": 0.6620370370370371, | |
| "eval_recall": 0.8148148148148148, | |
| "eval_runtime": 0.9449, | |
| "eval_samples_per_second": 197.907, | |
| "eval_steps_per_second": 3.175, | |
| "step": 1802 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.4619590044021606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1885, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9431454590204282, | |
| "eval_f1": 0.7270408163265305, | |
| "eval_loss": 0.1805543154478073, | |
| "eval_precision": 0.6581986143187067, | |
| "eval_recall": 0.811965811965812, | |
| "eval_runtime": 0.95, | |
| "eval_samples_per_second": 196.84, | |
| "eval_steps_per_second": 3.158, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.7237643003463745, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1877, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9431454590204282, | |
| "eval_f1": 0.7247119078104994, | |
| "eval_loss": 0.17826727032661438, | |
| "eval_precision": 0.6581395348837209, | |
| "eval_recall": 0.8062678062678063, | |
| "eval_runtime": 0.9528, | |
| "eval_samples_per_second": 196.268, | |
| "eval_steps_per_second": 3.149, | |
| "step": 2014 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.3640408515930176, | |
| "learning_rate": 0.0, | |
| "loss": 0.1858, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9433915825744523, | |
| "eval_f1": 0.7279693486590039, | |
| "eval_loss": 0.1783868670463562, | |
| "eval_precision": 0.6597222222222222, | |
| "eval_recall": 0.811965811965812, | |
| "eval_runtime": 0.9505, | |
| "eval_samples_per_second": 196.741, | |
| "eval_steps_per_second": 3.156, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2120, | |
| "total_flos": 907028676246000.0, | |
| "train_loss": 0.35348991987840184, | |
| "train_runtime": 247.3863, | |
| "train_samples_per_second": 136.467, | |
| "train_steps_per_second": 8.57 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 907028676246000.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |