| { | |
| "best_metric": 0.3475247025489807, | |
| "best_model_checkpoint": "model/MD_SL_distilroberta_stereoset_crowspair/checkpoint-11604", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 11604, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.9569114098586697e-05, | |
| "loss": 0.9727, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.913822819717339e-05, | |
| "loss": 0.6711, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.8707342295760084e-05, | |
| "loss": 0.6215, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.8276456394346776e-05, | |
| "loss": 0.5931, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.7845570492933472e-05, | |
| "loss": 0.5661, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.7414684591520167e-05, | |
| "loss": 0.5487, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.698379869010686e-05, | |
| "loss": 0.511, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_balanced accuracy": 0.8080920694003835, | |
| "eval_f1": 0.7656648771742313, | |
| "eval_loss": 0.4811922311782837, | |
| "eval_precision": 0.7380377322702882, | |
| "eval_recall": 0.8080920694003835, | |
| "eval_runtime": 17.0072, | |
| "eval_samples_per_second": 909.792, | |
| "eval_steps_per_second": 56.917, | |
| "step": 3868 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.6552912788693555e-05, | |
| "loss": 0.4866, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.612202688728025e-05, | |
| "loss": 0.4267, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.5691140985866945e-05, | |
| "loss": 0.3948, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.5260255084453637e-05, | |
| "loss": 0.3941, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.4829369183040333e-05, | |
| "loss": 0.4088, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 1.4398483281627026e-05, | |
| "loss": 0.3999, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.396759738021372e-05, | |
| "loss": 0.3798, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.3536711478800415e-05, | |
| "loss": 0.37, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_balanced accuracy": 0.8496016379296125, | |
| "eval_f1": 0.8261239996996584, | |
| "eval_loss": 0.3630180358886719, | |
| "eval_precision": 0.8068019430624749, | |
| "eval_recall": 0.8496016379296125, | |
| "eval_runtime": 16.2421, | |
| "eval_samples_per_second": 952.648, | |
| "eval_steps_per_second": 59.598, | |
| "step": 7736 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.3105825577387107e-05, | |
| "loss": 0.3334, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.2674939675973803e-05, | |
| "loss": 0.296, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1.2244053774560498e-05, | |
| "loss": 0.2934, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.181316787314719e-05, | |
| "loss": 0.3003, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 1.1382281971733886e-05, | |
| "loss": 0.2823, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.095139607032058e-05, | |
| "loss": 0.2796, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.0520510168907275e-05, | |
| "loss": 0.295, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.008962426749397e-05, | |
| "loss": 0.2815, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_balanced accuracy": 0.8764352768786915, | |
| "eval_f1": 0.8562979050918317, | |
| "eval_loss": 0.3475247025489807, | |
| "eval_precision": 0.8389008188354216, | |
| "eval_recall": 0.8764352768786915, | |
| "eval_runtime": 16.1972, | |
| "eval_samples_per_second": 955.288, | |
| "eval_steps_per_second": 59.763, | |
| "step": 11604 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 23208, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 2776408384051584.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |