| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.062061786651611, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.566, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_f1": 0.6402562480227776, | |
| "eval_loss": 0.5205540657043457, | |
| "eval_precision": 0.6483653398896937, | |
| "eval_recall": 0.6353427895981087, | |
| "eval_runtime": 5.1909, | |
| "eval_samples_per_second": 76.865, | |
| "eval_steps_per_second": 9.632, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.104482173919678, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5117, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7343358395989975, | |
| "eval_f1": 0.69391211208893, | |
| "eval_loss": 0.5062463879585266, | |
| "eval_precision": 0.6880119239984399, | |
| "eval_recall": 0.7045371885797418, | |
| "eval_runtime": 5.0488, | |
| "eval_samples_per_second": 79.028, | |
| "eval_steps_per_second": 9.903, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.243982791900635, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4804, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7669172932330827, | |
| "eval_f1": 0.7152209115816456, | |
| "eval_loss": 0.46674054861068726, | |
| "eval_precision": 0.7182055749128919, | |
| "eval_recall": 0.7125841062011276, | |
| "eval_runtime": 5.0519, | |
| "eval_samples_per_second": 78.98, | |
| "eval_steps_per_second": 9.897, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.68826961517334, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4345, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7919799498746867, | |
| "eval_f1": 0.744501107107864, | |
| "eval_loss": 0.43496260046958923, | |
| "eval_precision": 0.7494180559924504, | |
| "eval_recall": 0.7403164211674851, | |
| "eval_runtime": 5.0763, | |
| "eval_samples_per_second": 78.601, | |
| "eval_steps_per_second": 9.85, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.9247703552246094, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4081, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7944862155388471, | |
| "eval_f1": 0.7660995138690305, | |
| "eval_loss": 0.43371620774269104, | |
| "eval_precision": 0.7565013111888113, | |
| "eval_recall": 0.7845971994908165, | |
| "eval_runtime": 5.0744, | |
| "eval_samples_per_second": 78.631, | |
| "eval_steps_per_second": 9.853, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.43408203125, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3793, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_f1": 0.7753378378378378, | |
| "eval_loss": 0.39230969548225403, | |
| "eval_precision": 0.7857142857142857, | |
| "eval_recall": 0.7673213311511184, | |
| "eval_runtime": 5.0665, | |
| "eval_samples_per_second": 78.752, | |
| "eval_steps_per_second": 9.869, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.72346031665802, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3665, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7933776044839771, | |
| "eval_loss": 0.3765198886394501, | |
| "eval_precision": 0.7949020208205757, | |
| "eval_recall": 0.7919167121294781, | |
| "eval_runtime": 5.0594, | |
| "eval_samples_per_second": 78.863, | |
| "eval_steps_per_second": 9.883, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.3123555183410645, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3471, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.796615353247018, | |
| "eval_loss": 0.368134468793869, | |
| "eval_precision": 0.8088983050847458, | |
| "eval_recall": 0.7872340425531914, | |
| "eval_runtime": 5.0543, | |
| "eval_samples_per_second": 78.942, | |
| "eval_steps_per_second": 9.892, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.656528949737549, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3498, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.790357364116157, | |
| "eval_loss": 0.3676688075065613, | |
| "eval_precision": 0.8023956975228161, | |
| "eval_recall": 0.7811874886342971, | |
| "eval_runtime": 5.0959, | |
| "eval_samples_per_second": 78.299, | |
| "eval_steps_per_second": 9.812, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.076303482055664, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3282, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.7917273014868713, | |
| "eval_loss": 0.363395094871521, | |
| "eval_precision": 0.8074456774536514, | |
| "eval_recall": 0.780460083651573, | |
| "eval_runtime": 5.0935, | |
| "eval_samples_per_second": 78.335, | |
| "eval_steps_per_second": 9.816, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.567991256713867, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3149, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8065409159159159, | |
| "eval_loss": 0.3537313938140869, | |
| "eval_precision": 0.8180272108843537, | |
| "eval_recall": 0.7975541007455902, | |
| "eval_runtime": 5.0734, | |
| "eval_samples_per_second": 78.645, | |
| "eval_steps_per_second": 9.855, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 11.14825439453125, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3092, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8167483159828537, | |
| "eval_loss": 0.3528764247894287, | |
| "eval_precision": 0.8201621387462095, | |
| "eval_recall": 0.8136024731769412, | |
| "eval_runtime": 5.0609, | |
| "eval_samples_per_second": 78.84, | |
| "eval_steps_per_second": 9.88, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.867825031280518, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.3135, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8121903546212454, | |
| "eval_loss": 0.3471047580242157, | |
| "eval_precision": 0.8331751305173232, | |
| "eval_recall": 0.7978723404255319, | |
| "eval_runtime": 5.0541, | |
| "eval_samples_per_second": 78.945, | |
| "eval_steps_per_second": 9.893, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 12.051921844482422, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3103, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8269335415335841, | |
| "eval_loss": 0.3426941931247711, | |
| "eval_precision": 0.8430382253911666, | |
| "eval_recall": 0.8149663575195489, | |
| "eval_runtime": 5.0514, | |
| "eval_samples_per_second": 78.988, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.2898627519607544, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2974, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8297847585805701, | |
| "eval_loss": 0.33716997504234314, | |
| "eval_precision": 0.8385357006491028, | |
| "eval_recall": 0.8224677214038916, | |
| "eval_runtime": 5.0651, | |
| "eval_samples_per_second": 78.774, | |
| "eval_steps_per_second": 9.871, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.9146409034729, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2905, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8386324041811847, | |
| "eval_loss": 0.3345378339290619, | |
| "eval_precision": 0.8487869670976828, | |
| "eval_recall": 0.830287324968176, | |
| "eval_runtime": 5.0595, | |
| "eval_samples_per_second": 78.862, | |
| "eval_steps_per_second": 9.882, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.737354278564453, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2895, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8269335415335841, | |
| "eval_loss": 0.3339170217514038, | |
| "eval_precision": 0.8430382253911666, | |
| "eval_recall": 0.8149663575195489, | |
| "eval_runtime": 5.0983, | |
| "eval_samples_per_second": 78.261, | |
| "eval_steps_per_second": 9.807, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.6233842372894287, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2922, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8386324041811847, | |
| "eval_loss": 0.3318663537502289, | |
| "eval_precision": 0.8487869670976828, | |
| "eval_recall": 0.830287324968176, | |
| "eval_runtime": 5.0967, | |
| "eval_samples_per_second": 78.285, | |
| "eval_steps_per_second": 9.81, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.824616432189941, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2843, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8269335415335841, | |
| "eval_loss": 0.3319249749183655, | |
| "eval_precision": 0.8430382253911666, | |
| "eval_recall": 0.8149663575195489, | |
| "eval_runtime": 5.0765, | |
| "eval_samples_per_second": 78.598, | |
| "eval_steps_per_second": 9.849, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.412130832672119, | |
| "learning_rate": 0.0, | |
| "loss": 0.287, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8279052989013229, | |
| "eval_loss": 0.33124828338623047, | |
| "eval_precision": 0.8414113428943938, | |
| "eval_recall": 0.8174668121476631, | |
| "eval_runtime": 5.0634, | |
| "eval_samples_per_second": 78.8, | |
| "eval_steps_per_second": 9.875, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7597037114448000.0, | |
| "train_loss": 0.35801424432973394, | |
| "train_runtime": 1955.1531, | |
| "train_samples_per_second": 37.214, | |
| "train_steps_per_second": 1.248 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7597037114448000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |