| { | |
| "best_metric": 0.9309701492537313, | |
| "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-1972", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 4930, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9986744062735468, | |
| "eval_f1": 0.9266697804764128, | |
| "eval_loss": 0.004987192340195179, | |
| "eval_precision": 0.9288389513108615, | |
| "eval_recall": 0.9245107176141659, | |
| "eval_runtime": 13.1382, | |
| "eval_samples_per_second": 528.689, | |
| "eval_steps_per_second": 66.143, | |
| "step": 493 | |
| }, | |
| { | |
| "epoch": 1.0141987829614605, | |
| "grad_norm": 0.12838158011436462, | |
| "learning_rate": 4.4929006085192696e-05, | |
| "loss": 0.018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.998444171573689, | |
| "eval_f1": 0.9146567717996289, | |
| "eval_loss": 0.0057116178795695305, | |
| "eval_precision": 0.9104339796860572, | |
| "eval_recall": 0.918918918918919, | |
| "eval_runtime": 13.1923, | |
| "eval_samples_per_second": 526.518, | |
| "eval_steps_per_second": 65.872, | |
| "step": 986 | |
| }, | |
| { | |
| "epoch": 2.028397565922921, | |
| "grad_norm": 0.014988560229539871, | |
| "learning_rate": 3.98580121703854e-05, | |
| "loss": 0.0044, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9984930092372952, | |
| "eval_f1": 0.9260480452190296, | |
| "eval_loss": 0.00787710677832365, | |
| "eval_precision": 0.9361904761904762, | |
| "eval_recall": 0.9161230195712954, | |
| "eval_runtime": 13.2289, | |
| "eval_samples_per_second": 525.064, | |
| "eval_steps_per_second": 65.69, | |
| "step": 1479 | |
| }, | |
| { | |
| "epoch": 3.0425963488843815, | |
| "grad_norm": 0.24877521395683289, | |
| "learning_rate": 3.47870182555781e-05, | |
| "loss": 0.0023, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9986953367008066, | |
| "eval_f1": 0.9309701492537313, | |
| "eval_loss": 0.005731322802603245, | |
| "eval_precision": 0.9318394024276377, | |
| "eval_recall": 0.9301025163094129, | |
| "eval_runtime": 13.4618, | |
| "eval_samples_per_second": 515.978, | |
| "eval_steps_per_second": 64.553, | |
| "step": 1972 | |
| }, | |
| { | |
| "epoch": 4.056795131845842, | |
| "grad_norm": 0.003771956777200103, | |
| "learning_rate": 2.9716024340770794e-05, | |
| "loss": 0.0014, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9985837077554209, | |
| "eval_f1": 0.9213587715216379, | |
| "eval_loss": 0.007000353187322617, | |
| "eval_precision": 0.9200743494423792, | |
| "eval_recall": 0.9226467847157502, | |
| "eval_runtime": 13.1965, | |
| "eval_samples_per_second": 526.351, | |
| "eval_steps_per_second": 65.851, | |
| "step": 2465 | |
| }, | |
| { | |
| "epoch": 5.070993914807302, | |
| "grad_norm": 0.006697970442473888, | |
| "learning_rate": 2.4645030425963488e-05, | |
| "loss": 0.0008, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9985209164736416, | |
| "eval_f1": 0.9185938945420906, | |
| "eval_loss": 0.008161710575222969, | |
| "eval_precision": 0.9118457300275482, | |
| "eval_recall": 0.9254426840633737, | |
| "eval_runtime": 13.1313, | |
| "eval_samples_per_second": 528.965, | |
| "eval_steps_per_second": 66.178, | |
| "step": 2958 | |
| }, | |
| { | |
| "epoch": 6.085192697768763, | |
| "grad_norm": 0.0909859761595726, | |
| "learning_rate": 1.957403651115619e-05, | |
| "loss": 0.0006, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.998618591800854, | |
| "eval_f1": 0.9281767955801105, | |
| "eval_loss": 0.00739107932895422, | |
| "eval_precision": 0.9171974522292994, | |
| "eval_recall": 0.9394221808014911, | |
| "eval_runtime": 13.1688, | |
| "eval_samples_per_second": 527.459, | |
| "eval_steps_per_second": 65.989, | |
| "step": 3451 | |
| }, | |
| { | |
| "epoch": 7.099391480730223, | |
| "grad_norm": 0.011905203573405743, | |
| "learning_rate": 1.4503042596348884e-05, | |
| "loss": 0.0003, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9985069628554685, | |
| "eval_f1": 0.9232201023731968, | |
| "eval_loss": 0.008491172455251217, | |
| "eval_precision": 0.9219330855018587, | |
| "eval_recall": 0.9245107176141659, | |
| "eval_runtime": 13.2155, | |
| "eval_samples_per_second": 525.595, | |
| "eval_steps_per_second": 65.756, | |
| "step": 3944 | |
| }, | |
| { | |
| "epoch": 8.113590263691684, | |
| "grad_norm": 0.00037644465919584036, | |
| "learning_rate": 9.432048681541583e-06, | |
| "loss": 0.0003, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9985348700918149, | |
| "eval_f1": 0.9233610341643582, | |
| "eval_loss": 0.008594635874032974, | |
| "eval_precision": 0.9149130832570905, | |
| "eval_recall": 0.9319664492078286, | |
| "eval_runtime": 13.5474, | |
| "eval_samples_per_second": 512.72, | |
| "eval_steps_per_second": 64.145, | |
| "step": 4437 | |
| }, | |
| { | |
| "epoch": 9.127789046653144, | |
| "grad_norm": 0.053812168538570404, | |
| "learning_rate": 4.36105476673428e-06, | |
| "loss": 0.0002, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9985418469009014, | |
| "eval_f1": 0.9231481481481483, | |
| "eval_loss": 0.008874327875673771, | |
| "eval_precision": 0.9172033118675254, | |
| "eval_recall": 0.9291705498602051, | |
| "eval_runtime": 13.6739, | |
| "eval_samples_per_second": 507.975, | |
| "eval_steps_per_second": 63.552, | |
| "step": 4930 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 4930, | |
| "total_flos": 1.3595017549497408e+16, | |
| "train_loss": 0.0028862289850114567, | |
| "train_runtime": 1222.2462, | |
| "train_samples_per_second": 258.017, | |
| "train_steps_per_second": 4.034 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4930, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3595017549497408e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |