| { | |
| "best_metric": 0.8089887640449438, | |
| "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-16/checkpoint-240", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 240, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.2602354288101196, | |
| "learning_rate": 6.163228300166189e-06, | |
| "loss": 0.7018, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.1513686180114746, | |
| "learning_rate": 1.2326456600332378e-05, | |
| "loss": 0.6815, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.21480131149292, | |
| "learning_rate": 1.8489684900498564e-05, | |
| "loss": 0.6818, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.8887011408805847, | |
| "learning_rate": 2.4652913200664756e-05, | |
| "loss": 0.6556, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.8671120405197144, | |
| "learning_rate": 3.081614150083095e-05, | |
| "loss": 0.6284, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.6801092028617859, | |
| "learning_rate": 3.697936980099713e-05, | |
| "loss": 0.6166, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 2.9359517097473145, | |
| "learning_rate": 4.314259810116332e-05, | |
| "loss": 0.5407, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 1.6087604761123657, | |
| "learning_rate": 4.930582640132951e-05, | |
| "loss": 0.6755, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.7170913219451904, | |
| "learning_rate": 5.54690547014957e-05, | |
| "loss": 0.6359, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7614002227783203, | |
| "eval_runtime": 1.3801, | |
| "eval_samples_per_second": 46.373, | |
| "eval_steps_per_second": 5.797, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 4.872888565063477, | |
| "learning_rate": 5.889307042381025e-05, | |
| "loss": 0.7115, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.9400704503059387, | |
| "learning_rate": 5.820826727934734e-05, | |
| "loss": 0.7024, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.040548086166382, | |
| "learning_rate": 5.752346413488443e-05, | |
| "loss": 0.4111, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.7833982706069946, | |
| "learning_rate": 5.6838660990421516e-05, | |
| "loss": 0.5595, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 1.4219136238098145, | |
| "learning_rate": 5.615385784595861e-05, | |
| "loss": 0.4831, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": Infinity, | |
| "learning_rate": 5.560601533038828e-05, | |
| "loss": 0.7837, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 2.069016218185425, | |
| "learning_rate": 5.492121218592537e-05, | |
| "loss": 0.62, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 1.6181252002716064, | |
| "learning_rate": 5.423640904146246e-05, | |
| "loss": 0.5911, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.8916351795196533, | |
| "learning_rate": 5.355160589699955e-05, | |
| "loss": 0.5905, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 3.4919614791870117, | |
| "learning_rate": 5.286680275253664e-05, | |
| "loss": 0.5972, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.74, | |
| "eval_loss": 0.6621341705322266, | |
| "eval_runtime": 1.3734, | |
| "eval_samples_per_second": 46.601, | |
| "eval_steps_per_second": 5.825, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.276584267616272, | |
| "learning_rate": 5.231896023696631e-05, | |
| "loss": 0.5767, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 1.1047593355178833, | |
| "learning_rate": 5.163415709250341e-05, | |
| "loss": 0.4715, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 1.551737666130066, | |
| "learning_rate": 5.0949353948040496e-05, | |
| "loss": 0.5687, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.3372799158096313, | |
| "learning_rate": 5.0264550803577585e-05, | |
| "loss": 0.5399, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 4.043832778930664, | |
| "learning_rate": 4.9716708288007255e-05, | |
| "loss": 0.5178, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 5.607926845550537, | |
| "learning_rate": 4.9031905143544343e-05, | |
| "loss": 0.4201, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 10.094707489013672, | |
| "learning_rate": 4.834710199908144e-05, | |
| "loss": 0.6339, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 2.790902614593506, | |
| "learning_rate": 4.766229885461853e-05, | |
| "loss": 0.3564, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 8.226430892944336, | |
| "learning_rate": 4.6977495710155616e-05, | |
| "loss": 0.2918, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.72, | |
| "eval_loss": 0.8946094512939453, | |
| "eval_runtime": 1.3813, | |
| "eval_samples_per_second": 46.334, | |
| "eval_steps_per_second": 5.792, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 3.521246910095215, | |
| "learning_rate": 4.6292692565692705e-05, | |
| "loss": 0.4787, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": Infinity, | |
| "learning_rate": 4.574485005012238e-05, | |
| "loss": 0.3851, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 3.6489925384521484, | |
| "learning_rate": 4.506004690565947e-05, | |
| "loss": 0.2152, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 9.47097396850586, | |
| "learning_rate": 4.451220439008914e-05, | |
| "loss": 0.6503, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 1.4761028289794922, | |
| "learning_rate": 4.382740124562623e-05, | |
| "loss": 0.177, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 15.07903003692627, | |
| "learning_rate": 4.314259810116332e-05, | |
| "loss": 0.3834, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 98.74039459228516, | |
| "learning_rate": 4.245779495670041e-05, | |
| "loss": 0.816, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.6765172481536865, | |
| "learning_rate": 4.17729918122375e-05, | |
| "loss": 0.3177, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 1.5212448835372925, | |
| "learning_rate": 4.108818866777459e-05, | |
| "loss": 0.3593, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 27.127269744873047, | |
| "learning_rate": 4.040338552331168e-05, | |
| "loss": 0.3041, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.7692307692307693, | |
| "eval_loss": 1.0041449069976807, | |
| "eval_runtime": 1.3873, | |
| "eval_samples_per_second": 46.134, | |
| "eval_steps_per_second": 5.767, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 0.7140729427337646, | |
| "learning_rate": 3.9718582378848774e-05, | |
| "loss": 0.2064, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 4.367918491363525, | |
| "learning_rate": 3.903377923438586e-05, | |
| "loss": 0.365, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "grad_norm": 0.4135078489780426, | |
| "learning_rate": 3.834897608992295e-05, | |
| "loss": 0.3944, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "grad_norm": 88.720458984375, | |
| "learning_rate": 3.766417294546004e-05, | |
| "loss": 0.3774, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 1.5230985879898071, | |
| "learning_rate": 3.697936980099713e-05, | |
| "loss": 0.0844, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "grad_norm": Infinity, | |
| "learning_rate": 3.6431527285426805e-05, | |
| "loss": 0.3702, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "grad_norm": 39.243804931640625, | |
| "learning_rate": 3.5746724140963894e-05, | |
| "loss": 0.3381, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 0.23127637803554535, | |
| "learning_rate": 3.506192099650098e-05, | |
| "loss": 0.03, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "grad_norm": 23.879899978637695, | |
| "learning_rate": 3.451407848093066e-05, | |
| "loss": 0.1969, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.2918482720851898, | |
| "learning_rate": 3.382927533646775e-05, | |
| "loss": 0.2145, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.8089887640449438, | |
| "eval_loss": 1.1479806900024414, | |
| "eval_runtime": 1.3695, | |
| "eval_samples_per_second": 46.731, | |
| "eval_steps_per_second": 5.841, | |
| "step": 240 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3675891927575280.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 5.916699168159541e-05, | |
| "per_device_train_batch_size": 4 | |
| } | |
| } | |