| { | |
| "best_metric": 0.8167628096369188, | |
| "best_model_checkpoint": "finetune_results/omarmomen/structformer_s1_final_with_pos/qqp/checkpoint-7200", | |
| "epoch": 4.532019704433497, | |
| "global_step": 9200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.7194763422012329, | |
| "eval_f1": 0.6927870321345662, | |
| "eval_loss": 0.5358649492263794, | |
| "eval_mcc": 0.44235739892853754, | |
| "eval_runtime": 52.1084, | |
| "eval_samples_per_second": 516.02, | |
| "eval_steps_per_second": 64.519, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.876847290640394e-05, | |
| "loss": 0.5893, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 0.737550675868988, | |
| "eval_f1": 0.702575125384583, | |
| "eval_loss": 0.5323818325996399, | |
| "eval_mcc": 0.47036503965291415, | |
| "eval_runtime": 52.0801, | |
| "eval_samples_per_second": 516.301, | |
| "eval_steps_per_second": 64.554, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.753694581280788e-05, | |
| "loss": 0.5313, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_accuracy": 0.7599018216133118, | |
| "eval_f1": 0.7463061930210626, | |
| "eval_loss": 0.48686277866363525, | |
| "eval_mcc": 0.5356846095627209, | |
| "eval_runtime": 52.0592, | |
| "eval_samples_per_second": 516.508, | |
| "eval_steps_per_second": 64.58, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.630541871921182e-05, | |
| "loss": 0.474, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_accuracy": 0.7835918068885803, | |
| "eval_f1": 0.7604659778537027, | |
| "eval_loss": 0.4482875466346741, | |
| "eval_mcc": 0.5694046959973369, | |
| "eval_runtime": 52.074, | |
| "eval_samples_per_second": 516.361, | |
| "eval_steps_per_second": 64.562, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.507389162561577e-05, | |
| "loss": 0.4415, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7969058156013489, | |
| "eval_f1": 0.7559111428954542, | |
| "eval_loss": 0.42860299348831177, | |
| "eval_mcc": 0.5821443580249482, | |
| "eval_runtime": 52.0834, | |
| "eval_samples_per_second": 516.269, | |
| "eval_steps_per_second": 64.55, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_accuracy": 0.8098106980323792, | |
| "eval_f1": 0.7829556064850182, | |
| "eval_loss": 0.41030365228652954, | |
| "eval_mcc": 0.6158151672626542, | |
| "eval_runtime": 52.0798, | |
| "eval_samples_per_second": 516.304, | |
| "eval_steps_per_second": 64.555, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.384236453201971e-05, | |
| "loss": 0.3721, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_accuracy": 0.813975989818573, | |
| "eval_f1": 0.787672977332541, | |
| "eval_loss": 0.40312352776527405, | |
| "eval_mcc": 0.6242547952081453, | |
| "eval_runtime": 52.0405, | |
| "eval_samples_per_second": 516.693, | |
| "eval_steps_per_second": 64.603, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 4.261083743842365e-05, | |
| "loss": 0.3587, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_accuracy": 0.8195172548294067, | |
| "eval_f1": 0.788013803346001, | |
| "eval_loss": 0.39645758271217346, | |
| "eval_mcc": 0.6310785500997798, | |
| "eval_runtime": 52.0877, | |
| "eval_samples_per_second": 516.225, | |
| "eval_steps_per_second": 64.545, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 0.3525, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_accuracy": 0.8241659998893738, | |
| "eval_f1": 0.7917364108889084, | |
| "eval_loss": 0.38623449206352234, | |
| "eval_mcc": 0.639616031231923, | |
| "eval_runtime": 52.0731, | |
| "eval_samples_per_second": 516.37, | |
| "eval_steps_per_second": 64.563, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 4.014778325123153e-05, | |
| "loss": 0.3471, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_accuracy": 0.8224552869796753, | |
| "eval_f1": 0.8010999083409716, | |
| "eval_loss": 0.3869820833206177, | |
| "eval_mcc": 0.6453100501165008, | |
| "eval_runtime": 52.0844, | |
| "eval_samples_per_second": 516.258, | |
| "eval_steps_per_second": 64.549, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.8234965801239014, | |
| "eval_f1": 0.8041271151465125, | |
| "eval_loss": 0.44252586364746094, | |
| "eval_mcc": 0.6496465352226999, | |
| "eval_runtime": 52.0674, | |
| "eval_samples_per_second": 516.427, | |
| "eval_steps_per_second": 64.57, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.891625615763547e-05, | |
| "loss": 0.2559, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_accuracy": 0.8278850317001343, | |
| "eval_f1": 0.8037652645861602, | |
| "eval_loss": 0.41496795415878296, | |
| "eval_mcc": 0.6527458760331785, | |
| "eval_runtime": 52.0529, | |
| "eval_samples_per_second": 516.571, | |
| "eval_steps_per_second": 64.588, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.768472906403941e-05, | |
| "loss": 0.2519, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_accuracy": 0.8265833854675293, | |
| "eval_f1": 0.8060558166618143, | |
| "eval_loss": 0.4093920886516571, | |
| "eval_mcc": 0.6540712291224197, | |
| "eval_runtime": 52.0628, | |
| "eval_samples_per_second": 516.472, | |
| "eval_steps_per_second": 64.576, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 3.645320197044335e-05, | |
| "loss": 0.2532, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.8206701874732971, | |
| "eval_f1": 0.8046666126549461, | |
| "eval_loss": 0.41375023126602173, | |
| "eval_mcc": 0.6489224586140616, | |
| "eval_runtime": 52.0877, | |
| "eval_samples_per_second": 516.225, | |
| "eval_steps_per_second": 64.545, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 3.522167487684729e-05, | |
| "loss": 0.2564, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.8307858109474182, | |
| "eval_f1": 0.8087592468056489, | |
| "eval_loss": 0.38956916332244873, | |
| "eval_mcc": 0.660341552225514, | |
| "eval_runtime": 52.063, | |
| "eval_samples_per_second": 516.47, | |
| "eval_steps_per_second": 64.576, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "eval_accuracy": 0.8343188762664795, | |
| "eval_f1": 0.8126025322845244, | |
| "eval_loss": 0.4872523546218872, | |
| "eval_mcc": 0.6673564826893487, | |
| "eval_runtime": 52.0098, | |
| "eval_samples_per_second": 516.998, | |
| "eval_steps_per_second": 64.642, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 3.399014778325123e-05, | |
| "loss": 0.1843, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_accuracy": 0.8294841647148132, | |
| "eval_f1": 0.8107639605431507, | |
| "eval_loss": 0.4589603543281555, | |
| "eval_mcc": 0.6617755486673744, | |
| "eval_runtime": 51.9587, | |
| "eval_samples_per_second": 517.508, | |
| "eval_steps_per_second": 64.705, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.1746, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "eval_accuracy": 0.8393394947052002, | |
| "eval_f1": 0.8167628096369188, | |
| "eval_loss": 0.43506747484207153, | |
| "eval_mcc": 0.6759774095179583, | |
| "eval_runtime": 52.0115, | |
| "eval_samples_per_second": 516.981, | |
| "eval_steps_per_second": 64.64, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 3.152709359605912e-05, | |
| "loss": 0.1745, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_accuracy": 0.8354717493057251, | |
| "eval_f1": 0.8156513042753563, | |
| "eval_loss": 0.43602147698402405, | |
| "eval_mcc": 0.671683544076007, | |
| "eval_runtime": 51.9689, | |
| "eval_samples_per_second": 517.405, | |
| "eval_steps_per_second": 64.693, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 3.0295566502463057e-05, | |
| "loss": 0.1819, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_accuracy": 0.8390792012214661, | |
| "eval_f1": 0.8161774077063597, | |
| "eval_loss": 0.432695209980011, | |
| "eval_mcc": 0.6751676671615486, | |
| "eval_runtime": 51.9416, | |
| "eval_samples_per_second": 517.678, | |
| "eval_steps_per_second": 64.727, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_accuracy": 0.8347651362419128, | |
| "eval_f1": 0.8147437768419297, | |
| "eval_loss": 0.5000645518302917, | |
| "eval_mcc": 0.670111604659899, | |
| "eval_runtime": 51.9514, | |
| "eval_samples_per_second": 517.58, | |
| "eval_steps_per_second": 64.714, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 2.9064039408866993e-05, | |
| "loss": 0.1327, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "eval_accuracy": 0.8399345278739929, | |
| "eval_f1": 0.814802065404475, | |
| "eval_loss": 0.5149998068809509, | |
| "eval_mcc": 0.6748319203929715, | |
| "eval_runtime": 51.959, | |
| "eval_samples_per_second": 517.504, | |
| "eval_steps_per_second": 64.705, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 2.7832512315270936e-05, | |
| "loss": 0.12, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_accuracy": 0.8342816829681396, | |
| "eval_f1": 0.8136032795114196, | |
| "eval_loss": 0.5296575427055359, | |
| "eval_mcc": 0.6684337154989034, | |
| "eval_runtime": 51.9583, | |
| "eval_samples_per_second": 517.511, | |
| "eval_steps_per_second": 64.706, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "step": 9200, | |
| "total_flos": 9.212089774816051e+16, | |
| "train_loss": 0.29893846636233123, | |
| "train_runtime": 6512.0588, | |
| "train_samples_per_second": 373.919, | |
| "train_steps_per_second": 3.117 | |
| } | |
| ], | |
| "max_steps": 20300, | |
| "num_train_epochs": 10, | |
| "total_flos": 9.212089774816051e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |