{ "best_metric": 1.0, "best_model_checkpoint": "distilbert-base-uncased_finetuned_text_2_disease_cel_v2/checkpoint-501", "epoch": 3.0, "global_step": 501, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.06, "learning_rate": 3.920159680638723e-05, "loss": 3.208, "step": 10 }, { "epoch": 0.12, "learning_rate": 3.840319361277446e-05, "loss": 3.1566, "step": 20 }, { "epoch": 0.18, "learning_rate": 3.7604790419161676e-05, "loss": 3.0327, "step": 30 }, { "epoch": 0.24, "learning_rate": 3.680638722554891e-05, "loss": 2.8296, "step": 40 }, { "epoch": 0.3, "learning_rate": 3.600798403193613e-05, "loss": 2.5667, "step": 50 }, { "epoch": 0.36, "learning_rate": 3.5209580838323356e-05, "loss": 2.3494, "step": 60 }, { "epoch": 0.42, "learning_rate": 3.441117764471058e-05, "loss": 2.0569, "step": 70 }, { "epoch": 0.48, "learning_rate": 3.3612774451097804e-05, "loss": 1.7982, "step": 80 }, { "epoch": 0.54, "learning_rate": 3.2814371257485035e-05, "loss": 1.577, "step": 90 }, { "epoch": 0.6, "learning_rate": 3.201596806387226e-05, "loss": 1.3064, "step": 100 }, { "epoch": 0.66, "learning_rate": 3.1217564870259484e-05, "loss": 1.2182, "step": 110 }, { "epoch": 0.72, "learning_rate": 3.041916167664671e-05, "loss": 0.9891, "step": 120 }, { "epoch": 0.78, "learning_rate": 2.9620758483033936e-05, "loss": 0.872, "step": 130 }, { "epoch": 0.84, "learning_rate": 2.8822355289421164e-05, "loss": 0.7503, "step": 140 }, { "epoch": 0.9, "learning_rate": 2.8023952095808385e-05, "loss": 0.6134, "step": 150 }, { "epoch": 0.96, "learning_rate": 2.7225548902195612e-05, "loss": 0.5425, "step": 160 }, { "epoch": 1.0, "eval_accuracy": 0.963963963963964, "eval_f1": 0.9630884536155822, "eval_loss": 0.36853650212287903, "eval_precision": 0.9698101430349508, "eval_recall": 0.963963963963964, "eval_runtime": 3.8045, "eval_samples_per_second": 175.058, "eval_steps_per_second": 5.52, "step": 167 }, { "epoch": 1.02, "learning_rate": 2.6427145708582837e-05, "loss": 0.4085, "step": 170 }, { "epoch": 1.08, "learning_rate": 2.5628742514970064e-05, "loss": 0.3705, "step": 180 }, { "epoch": 1.14, "learning_rate": 2.4830339321357285e-05, "loss": 0.317, "step": 190 }, { "epoch": 1.2, "learning_rate": 2.4031936127744513e-05, "loss": 0.25, "step": 200 }, { "epoch": 1.26, "learning_rate": 2.323353293413174e-05, "loss": 0.2564, "step": 210 }, { "epoch": 1.32, "learning_rate": 2.2435129740518965e-05, "loss": 0.2334, "step": 220 }, { "epoch": 1.38, "learning_rate": 2.163672654690619e-05, "loss": 0.231, "step": 230 }, { "epoch": 1.44, "learning_rate": 2.0838323353293413e-05, "loss": 0.1942, "step": 240 }, { "epoch": 1.5, "learning_rate": 2.003992015968064e-05, "loss": 0.1782, "step": 250 }, { "epoch": 1.56, "learning_rate": 1.9241516966067865e-05, "loss": 0.1353, "step": 260 }, { "epoch": 1.62, "learning_rate": 1.8443113772455093e-05, "loss": 0.1027, "step": 270 }, { "epoch": 1.68, "learning_rate": 1.7644710578842317e-05, "loss": 0.1048, "step": 280 }, { "epoch": 1.74, "learning_rate": 1.684630738522954e-05, "loss": 0.1052, "step": 290 }, { "epoch": 1.8, "learning_rate": 1.604790419161677e-05, "loss": 0.0849, "step": 300 }, { "epoch": 1.86, "learning_rate": 1.5249500998003994e-05, "loss": 0.0974, "step": 310 }, { "epoch": 1.92, "learning_rate": 1.4451097804391218e-05, "loss": 0.075, "step": 320 }, { "epoch": 1.98, "learning_rate": 1.3652694610778446e-05, "loss": 0.0887, "step": 330 }, { "epoch": 2.0, "eval_accuracy": 0.9984984984984985, "eval_f1": 0.9984973421631103, "eval_loss": 0.042379751801490784, "eval_precision": 0.9985541096652208, "eval_recall": 0.9984984984984985, "eval_runtime": 3.9115, "eval_samples_per_second": 170.269, "eval_steps_per_second": 5.369, "step": 334 }, { "epoch": 2.04, "learning_rate": 1.285429141716567e-05, "loss": 0.0749, "step": 340 }, { "epoch": 2.1, "learning_rate": 1.2055888223552896e-05, "loss": 0.0532, "step": 350 }, { "epoch": 2.16, "learning_rate": 1.125748502994012e-05, "loss": 0.0807, "step": 360 }, { "epoch": 2.22, "learning_rate": 1.0459081836327348e-05, "loss": 0.0712, "step": 370 }, { "epoch": 2.28, "learning_rate": 9.660678642714572e-06, "loss": 0.0575, "step": 380 }, { "epoch": 2.34, "learning_rate": 8.862275449101796e-06, "loss": 0.0533, "step": 390 }, { "epoch": 2.4, "learning_rate": 8.063872255489022e-06, "loss": 0.0612, "step": 400 }, { "epoch": 2.46, "learning_rate": 7.2654690618762484e-06, "loss": 0.0482, "step": 410 }, { "epoch": 2.51, "learning_rate": 6.4670658682634736e-06, "loss": 0.0539, "step": 420 }, { "epoch": 2.57, "learning_rate": 5.668662674650699e-06, "loss": 0.0655, "step": 430 }, { "epoch": 2.63, "learning_rate": 4.870259481037925e-06, "loss": 0.0646, "step": 440 }, { "epoch": 2.69, "learning_rate": 4.07185628742515e-06, "loss": 0.0642, "step": 450 }, { "epoch": 2.75, "learning_rate": 3.2734530938123754e-06, "loss": 0.0448, "step": 460 }, { "epoch": 2.81, "learning_rate": 2.475049900199601e-06, "loss": 0.0394, "step": 470 }, { "epoch": 2.87, "learning_rate": 1.6766467065868263e-06, "loss": 0.0657, "step": 480 }, { "epoch": 2.93, "learning_rate": 8.78243512974052e-07, "loss": 0.0769, "step": 490 }, { "epoch": 2.99, "learning_rate": 7.984031936127744e-08, "loss": 0.0399, "step": 500 }, { "epoch": 3.0, "eval_accuracy": 1.0, "eval_f1": 1.0, "eval_loss": 0.026059171184897423, "eval_precision": 1.0, "eval_recall": 1.0, "eval_runtime": 3.7333, "eval_samples_per_second": 178.395, "eval_steps_per_second": 5.625, "step": 501 }, { "epoch": 3.0, "step": 501, "total_flos": 2118227382190080.0, "train_loss": 0.661047323973117, "train_runtime": 349.1831, "train_samples_per_second": 45.775, "train_steps_per_second": 1.435 }, { "epoch": 3.0, "eval_accuracy": 0.9981231231231231, "eval_f1": 0.9981237024802164, "eval_loss": 0.036395490169525146, "eval_precision": 0.9981332041882663, "eval_recall": 0.9981231231231231, "eval_runtime": 34.4871, "eval_samples_per_second": 154.492, "eval_steps_per_second": 4.842, "step": 501 }, { "epoch": 3.0, "eval_accuracy": 1.0, "eval_f1": 1.0, "eval_loss": 0.026059171184897423, "eval_precision": 1.0, "eval_recall": 1.0, "eval_runtime": 4.0142, "eval_samples_per_second": 165.91, "eval_steps_per_second": 5.231, "step": 501 }, { "epoch": 3.0, "eval_accuracy": 1.0, "eval_f1": 1.0, "eval_loss": 0.025151288136839867, "eval_precision": 1.0, "eval_recall": 1.0, "eval_runtime": 3.787, "eval_samples_per_second": 176.129, "eval_steps_per_second": 5.545, "step": 501 } ], "max_steps": 501, "num_train_epochs": 3, "total_flos": 2118227382190080.0, "trial_name": null, "trial_params": null }