| { | |
| "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": 5.431522846221924, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5653, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_f1": 0.6333236742973052, | |
| "eval_loss": 0.5245500802993774, | |
| "eval_precision": 0.6540233301136786, | |
| "eval_recall": 0.6256592107655937, | |
| "eval_runtime": 4.6575, | |
| "eval_samples_per_second": 85.668, | |
| "eval_steps_per_second": 10.735, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.8715531826019287, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5167, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_f1": 0.6854598540145985, | |
| "eval_loss": 0.521543025970459, | |
| "eval_precision": 0.6804511278195489, | |
| "eval_recall": 0.6934897254046191, | |
| "eval_runtime": 5.0566, | |
| "eval_samples_per_second": 78.907, | |
| "eval_steps_per_second": 9.888, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.720804691314697, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4984, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7443609022556391, | |
| "eval_f1": 0.6916257501363885, | |
| "eval_loss": 0.49747392535209656, | |
| "eval_precision": 0.6916257501363885, | |
| "eval_recall": 0.6916257501363885, | |
| "eval_runtime": 5.0553, | |
| "eval_samples_per_second": 78.927, | |
| "eval_steps_per_second": 9.891, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.2620692253112793, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4765, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7418546365914787, | |
| "eval_f1": 0.6619339448031918, | |
| "eval_loss": 0.4854496121406555, | |
| "eval_precision": 0.6836634025138848, | |
| "eval_recall": 0.6523458810692853, | |
| "eval_runtime": 5.0451, | |
| "eval_samples_per_second": 79.087, | |
| "eval_steps_per_second": 9.911, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.330729007720947, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4797, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7719298245614035, | |
| "eval_f1": 0.7320072332730561, | |
| "eval_loss": 0.485201358795166, | |
| "eval_precision": 0.7269805119926199, | |
| "eval_recall": 0.7386342971449354, | |
| "eval_runtime": 5.0467, | |
| "eval_samples_per_second": 79.062, | |
| "eval_steps_per_second": 9.907, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.9019949436187744, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.4668, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7669172932330827, | |
| "eval_f1": 0.7195344091304183, | |
| "eval_loss": 0.47377943992614746, | |
| "eval_precision": 0.7189969238192895, | |
| "eval_recall": 0.7200854700854701, | |
| "eval_runtime": 5.0582, | |
| "eval_samples_per_second": 78.882, | |
| "eval_steps_per_second": 9.885, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 14.114830017089844, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.4622, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7719298245614035, | |
| "eval_f1": 0.7295157072938161, | |
| "eval_loss": 0.47686928510665894, | |
| "eval_precision": 0.7260557184750733, | |
| "eval_recall": 0.7336333878887071, | |
| "eval_runtime": 5.045, | |
| "eval_samples_per_second": 79.088, | |
| "eval_steps_per_second": 9.911, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.223310947418213, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4621, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7493734335839599, | |
| "eval_f1": 0.6685826300750881, | |
| "eval_loss": 0.46253928542137146, | |
| "eval_precision": 0.6949044585987261, | |
| "eval_recall": 0.6576650300054556, | |
| "eval_runtime": 5.052, | |
| "eval_samples_per_second": 78.979, | |
| "eval_steps_per_second": 9.897, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.2552032470703125, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.4561, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7769423558897243, | |
| "eval_f1": 0.7199470035725271, | |
| "eval_loss": 0.4609311521053314, | |
| "eval_precision": 0.7310853634383045, | |
| "eval_recall": 0.7121749408983451, | |
| "eval_runtime": 5.0567, | |
| "eval_samples_per_second": 78.905, | |
| "eval_steps_per_second": 9.888, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.524580240249634, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4519, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7669172932330827, | |
| "eval_f1": 0.6822078533807219, | |
| "eval_loss": 0.4608353078365326, | |
| "eval_precision": 0.7252321981424149, | |
| "eval_recall": 0.6675759228950718, | |
| "eval_runtime": 5.0558, | |
| "eval_samples_per_second": 78.919, | |
| "eval_steps_per_second": 9.89, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.3073718547821045, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.4413, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7694235588972431, | |
| "eval_f1": 0.7079992363497518, | |
| "eval_loss": 0.4543740451335907, | |
| "eval_precision": 0.7214646464646465, | |
| "eval_recall": 0.6993544280778323, | |
| "eval_runtime": 5.0512, | |
| "eval_samples_per_second": 78.992, | |
| "eval_steps_per_second": 9.899, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.039628982543945, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4449, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7412841546534773, | |
| "eval_loss": 0.4569094777107239, | |
| "eval_precision": 0.7401260504201681, | |
| "eval_recall": 0.7424986361156574, | |
| "eval_runtime": 5.0491, | |
| "eval_samples_per_second": 79.023, | |
| "eval_steps_per_second": 9.903, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.8723957538604736, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.4506, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7644110275689223, | |
| "eval_f1": 0.6821309919316564, | |
| "eval_loss": 0.4527250826358795, | |
| "eval_precision": 0.7196598101265823, | |
| "eval_recall": 0.6683033278777959, | |
| "eval_runtime": 5.0532, | |
| "eval_samples_per_second": 78.959, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 2.8429393768310547, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.4446, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7794486215538847, | |
| "eval_f1": 0.7120834426659669, | |
| "eval_loss": 0.4487856924533844, | |
| "eval_precision": 0.7379122870605291, | |
| "eval_recall": 0.69894526277505, | |
| "eval_runtime": 5.0594, | |
| "eval_samples_per_second": 78.864, | |
| "eval_steps_per_second": 9.883, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.688943386077881, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.4426, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7869674185463659, | |
| "eval_f1": 0.7355039968804835, | |
| "eval_loss": 0.44907739758491516, | |
| "eval_precision": 0.7435604353145727, | |
| "eval_recall": 0.7292689579923622, | |
| "eval_runtime": 5.0481, | |
| "eval_samples_per_second": 79.039, | |
| "eval_steps_per_second": 9.905, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.842677593231201, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4409, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7719298245614035, | |
| "eval_f1": 0.7068474127421138, | |
| "eval_loss": 0.44651278853416443, | |
| "eval_precision": 0.725706313219393, | |
| "eval_recall": 0.696126568466994, | |
| "eval_runtime": 5.0391, | |
| "eval_samples_per_second": 79.18, | |
| "eval_steps_per_second": 9.922, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.7264163494110107, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.4348, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7869674185463659, | |
| "eval_f1": 0.7355039968804835, | |
| "eval_loss": 0.4473975598812103, | |
| "eval_precision": 0.7435604353145727, | |
| "eval_recall": 0.7292689579923622, | |
| "eval_runtime": 5.0762, | |
| "eval_samples_per_second": 78.602, | |
| "eval_steps_per_second": 9.85, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.3701038360595703, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4478, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7301509908776345, | |
| "eval_loss": 0.4460136294364929, | |
| "eval_precision": 0.7407832589871425, | |
| "eval_recall": 0.7224949990907438, | |
| "eval_runtime": 5.1002, | |
| "eval_samples_per_second": 78.233, | |
| "eval_steps_per_second": 9.804, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.103198289871216, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4382, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7869674185463659, | |
| "eval_f1": 0.7309977236133474, | |
| "eval_loss": 0.44484180212020874, | |
| "eval_precision": 0.7447157190635452, | |
| "eval_recall": 0.7217675941080197, | |
| "eval_runtime": 5.061, | |
| "eval_samples_per_second": 78.838, | |
| "eval_steps_per_second": 9.879, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 6.061123847961426, | |
| "learning_rate": 0.0, | |
| "loss": 0.4313, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7869674185463659, | |
| "eval_f1": 0.7325336550973572, | |
| "eval_loss": 0.445127934217453, | |
| "eval_precision": 0.7442562883739354, | |
| "eval_recall": 0.7242680487361338, | |
| "eval_runtime": 5.049, | |
| "eval_samples_per_second": 79.025, | |
| "eval_steps_per_second": 9.903, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7588990440528000.0, | |
| "train_loss": 0.4626342241881324, | |
| "train_runtime": 1944.0727, | |
| "train_samples_per_second": 37.427, | |
| "train_steps_per_second": 1.255 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7588990440528000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |