| { | |
| "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": 5.065453052520752, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5633, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7167919799498746, | |
| "eval_f1": 0.6483759017352311, | |
| "eval_loss": 0.5099548697471619, | |
| "eval_precision": 0.653631914482581, | |
| "eval_recall": 0.6446172031278414, | |
| "eval_runtime": 5.1271, | |
| "eval_samples_per_second": 77.822, | |
| "eval_steps_per_second": 9.752, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.7954251766204834, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5083, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6886704119850188, | |
| "eval_loss": 0.4999238848686218, | |
| "eval_precision": 0.6825396825396826, | |
| "eval_recall": 0.704946353882524, | |
| "eval_runtime": 5.0551, | |
| "eval_samples_per_second": 78.931, | |
| "eval_steps_per_second": 9.891, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.972048759460449, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4904, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7619047619047619, | |
| "eval_f1": 0.7090966301102831, | |
| "eval_loss": 0.4595041573047638, | |
| "eval_precision": 0.7119991289198606, | |
| "eval_recall": 0.7065375522822331, | |
| "eval_runtime": 5.1087, | |
| "eval_samples_per_second": 78.103, | |
| "eval_steps_per_second": 9.787, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.7911946773529053, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4644, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7919799498746867, | |
| "eval_f1": 0.7357932107383621, | |
| "eval_loss": 0.4287411570549011, | |
| "eval_precision": 0.7520170859041291, | |
| "eval_recall": 0.7253136933987998, | |
| "eval_runtime": 5.0546, | |
| "eval_samples_per_second": 78.937, | |
| "eval_steps_per_second": 9.892, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.300946593284607, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4439, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_f1": 0.720337003079755, | |
| "eval_loss": 0.43993374705314636, | |
| "eval_precision": 0.7126763217400668, | |
| "eval_recall": 0.7394526277505, | |
| "eval_runtime": 5.045, | |
| "eval_samples_per_second": 79.089, | |
| "eval_steps_per_second": 9.911, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.9654462337493896, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4241, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8220551378446115, | |
| "eval_f1": 0.7837101170434504, | |
| "eval_loss": 0.402724027633667, | |
| "eval_precision": 0.7859649122807018, | |
| "eval_recall": 0.7815966539370796, | |
| "eval_runtime": 5.056, | |
| "eval_samples_per_second": 78.917, | |
| "eval_steps_per_second": 9.889, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.654602527618408, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.4092, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8070175438596491, | |
| "eval_f1": 0.7742565963981571, | |
| "eval_loss": 0.4018576741218567, | |
| "eval_precision": 0.7674435230197312, | |
| "eval_recall": 0.7834606292053101, | |
| "eval_runtime": 5.0686, | |
| "eval_samples_per_second": 78.72, | |
| "eval_steps_per_second": 9.865, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.236440181732178, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3891, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7919126261290201, | |
| "eval_loss": 0.3805297911167145, | |
| "eval_precision": 0.7911966946136679, | |
| "eval_recall": 0.7926441171122023, | |
| "eval_runtime": 5.0705, | |
| "eval_samples_per_second": 78.691, | |
| "eval_steps_per_second": 9.861, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.970191478729248, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3777, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7919126261290201, | |
| "eval_loss": 0.3789067566394806, | |
| "eval_precision": 0.7911966946136679, | |
| "eval_recall": 0.7926441171122023, | |
| "eval_runtime": 5.0553, | |
| "eval_samples_per_second": 78.927, | |
| "eval_steps_per_second": 9.891, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.090290069580078, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.369, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8395989974937343, | |
| "eval_f1": 0.8055318630437431, | |
| "eval_loss": 0.37584108114242554, | |
| "eval_precision": 0.8071187997550521, | |
| "eval_recall": 0.8040098199672667, | |
| "eval_runtime": 5.101, | |
| "eval_samples_per_second": 78.22, | |
| "eval_steps_per_second": 9.802, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.1066460609436035, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3531, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7983530530947143, | |
| "eval_loss": 0.38051342964172363, | |
| "eval_precision": 0.7932554744525547, | |
| "eval_recall": 0.8044189852700492, | |
| "eval_runtime": 5.11, | |
| "eval_samples_per_second": 78.082, | |
| "eval_steps_per_second": 9.785, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.612831115722656, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3486, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.8026866442779643, | |
| "eval_loss": 0.38006365299224854, | |
| "eval_precision": 0.795995733394834, | |
| "eval_recall": 0.8111929441716675, | |
| "eval_runtime": 5.0674, | |
| "eval_samples_per_second": 78.738, | |
| "eval_steps_per_second": 9.867, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.0057244300842285, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.3472, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8090546545634092, | |
| "eval_loss": 0.3675023913383484, | |
| "eval_precision": 0.8098422078713293, | |
| "eval_recall": 0.808283324240771, | |
| "eval_runtime": 5.0557, | |
| "eval_samples_per_second": 78.92, | |
| "eval_steps_per_second": 9.89, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 9.116528511047363, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3379, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8039756622954537, | |
| "eval_loss": 0.36543166637420654, | |
| "eval_precision": 0.8032299897460643, | |
| "eval_recall": 0.8047372249499909, | |
| "eval_runtime": 5.051, | |
| "eval_samples_per_second": 78.994, | |
| "eval_steps_per_second": 9.899, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.9552446007728577, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.3353, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8127416435111035, | |
| "eval_loss": 0.37026751041412354, | |
| "eval_precision": 0.8079618768328446, | |
| "eval_recall": 0.8182851427532278, | |
| "eval_runtime": 5.0565, | |
| "eval_samples_per_second": 78.908, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.738326549530029, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3213, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8076965854743632, | |
| "eval_loss": 0.3708532154560089, | |
| "eval_precision": 0.8018925518925519, | |
| "eval_recall": 0.8147390434624477, | |
| "eval_runtime": 5.0546, | |
| "eval_samples_per_second": 78.937, | |
| "eval_steps_per_second": 9.892, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.96878719329834, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.3214, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8058772015179527, | |
| "eval_loss": 0.36412617564201355, | |
| "eval_precision": 0.8024057525004439, | |
| "eval_recall": 0.8097381342062193, | |
| "eval_runtime": 5.0478, | |
| "eval_samples_per_second": 79.044, | |
| "eval_steps_per_second": 9.905, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.617929697036743, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3225, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8058772015179527, | |
| "eval_loss": 0.36399707198143005, | |
| "eval_precision": 0.8024057525004439, | |
| "eval_recall": 0.8097381342062193, | |
| "eval_runtime": 5.0536, | |
| "eval_samples_per_second": 78.953, | |
| "eval_steps_per_second": 9.894, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.830294132232666, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.3159, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8033602150537634, | |
| "eval_loss": 0.3648917078971863, | |
| "eval_precision": 0.7993107104984093, | |
| "eval_recall": 0.8079650845608293, | |
| "eval_runtime": 5.1284, | |
| "eval_samples_per_second": 77.802, | |
| "eval_steps_per_second": 9.75, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 6.41981840133667, | |
| "learning_rate": 0.0, | |
| "loss": 0.3195, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8033602150537634, | |
| "eval_loss": 0.36415907740592957, | |
| "eval_precision": 0.7993107104984093, | |
| "eval_recall": 0.8079650845608293, | |
| "eval_runtime": 5.106, | |
| "eval_samples_per_second": 78.143, | |
| "eval_steps_per_second": 9.792, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7590599775312000.0, | |
| "train_loss": 0.38811437106523355, | |
| "train_runtime": 1952.7564, | |
| "train_samples_per_second": 37.26, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7590599775312000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |