| { | |
| "best_metric": 0.9384684432613534, | |
| "best_model_checkpoint": "esm2_t30_150M_UR50D-finetuned-epitope_attempt7_150M_200k_default_accuracy/checkpoint-4576", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 22880, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.781468531468531e-05, | |
| "loss": 0.2387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.562937062937063e-05, | |
| "loss": 0.2219, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 9.344405594405596e-05, | |
| "loss": 0.2248, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.125874125874126e-05, | |
| "loss": 0.2252, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.93836177956272, | |
| "eval_loss": 0.22232767939567566, | |
| "eval_runtime": 73.981, | |
| "eval_samples_per_second": 61.84, | |
| "eval_steps_per_second": 3.447, | |
| "step": 2288 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.907342657342657e-05, | |
| "loss": 0.2174, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 8.688811188811189e-05, | |
| "loss": 0.2174, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 8.47027972027972e-05, | |
| "loss": 0.2133, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.251748251748252e-05, | |
| "loss": 0.2153, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 8.033216783216784e-05, | |
| "loss": 0.2129, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9384684432613534, | |
| "eval_loss": 0.21357476711273193, | |
| "eval_runtime": 72.5716, | |
| "eval_samples_per_second": 63.041, | |
| "eval_steps_per_second": 3.514, | |
| "step": 4576 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 7.814685314685315e-05, | |
| "loss": 0.2041, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 7.596153846153846e-05, | |
| "loss": 0.1996, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7.377622377622378e-05, | |
| "loss": 0.2019, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 7.15909090909091e-05, | |
| "loss": 0.2018, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9380660067945291, | |
| "eval_loss": 0.21539297699928284, | |
| "eval_runtime": 72.5454, | |
| "eval_samples_per_second": 63.064, | |
| "eval_steps_per_second": 3.515, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 6.940559440559441e-05, | |
| "loss": 0.1987, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.722027972027972e-05, | |
| "loss": 0.1793, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.503496503496504e-05, | |
| "loss": 0.1827, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 6.284965034965036e-05, | |
| "loss": 0.1834, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 6.066433566433567e-05, | |
| "loss": 0.1846, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9373786185144478, | |
| "eval_loss": 0.2264455258846283, | |
| "eval_runtime": 72.7083, | |
| "eval_samples_per_second": 62.923, | |
| "eval_steps_per_second": 3.507, | |
| "step": 9152 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 5.8479020979020984e-05, | |
| "loss": 0.1619, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 5.629370629370629e-05, | |
| "loss": 0.1514, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 5.4108391608391606e-05, | |
| "loss": 0.1605, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 5.192307692307693e-05, | |
| "loss": 0.1556, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9329569702150639, | |
| "eval_loss": 0.2557878792285919, | |
| "eval_runtime": 72.6531, | |
| "eval_samples_per_second": 62.97, | |
| "eval_steps_per_second": 3.51, | |
| "step": 11440 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 4.973776223776224e-05, | |
| "loss": 0.1522, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.755244755244756e-05, | |
| "loss": 0.1183, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 4.5367132867132865e-05, | |
| "loss": 0.13, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "learning_rate": 4.318181818181819e-05, | |
| "loss": 0.1314, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 4.0996503496503495e-05, | |
| "loss": 0.1335, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9281792609700007, | |
| "eval_loss": 0.29685506224632263, | |
| "eval_runtime": 72.5857, | |
| "eval_samples_per_second": 63.029, | |
| "eval_steps_per_second": 3.513, | |
| "step": 13728 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.8811188811188816e-05, | |
| "loss": 0.1169, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.6625874125874124e-05, | |
| "loss": 0.1083, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 3.4440559440559445e-05, | |
| "loss": 0.1126, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 3.225524475524476e-05, | |
| "loss": 0.1098, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 3.0069930069930068e-05, | |
| "loss": 0.1142, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9246155598117979, | |
| "eval_loss": 0.3515949547290802, | |
| "eval_runtime": 72.4969, | |
| "eval_samples_per_second": 63.106, | |
| "eval_steps_per_second": 3.517, | |
| "step": 16016 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 2.7884615384615386e-05, | |
| "loss": 0.0948, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 2.5699300699300697e-05, | |
| "loss": 0.0956, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 2.3513986013986015e-05, | |
| "loss": 0.0955, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 2.132867132867133e-05, | |
| "loss": 0.0988, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.925055096695538, | |
| "eval_loss": 0.3799491226673126, | |
| "eval_runtime": 72.5847, | |
| "eval_samples_per_second": 63.03, | |
| "eval_steps_per_second": 3.513, | |
| "step": 18304 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 1.9143356643356645e-05, | |
| "loss": 0.092, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 1.695804195804196e-05, | |
| "loss": 0.0822, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 1.4772727272727274e-05, | |
| "loss": 0.0854, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 1.2587412587412589e-05, | |
| "loss": 0.0885, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 1.0402097902097904e-05, | |
| "loss": 0.0842, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9202073088851892, | |
| "eval_loss": 0.4125804305076599, | |
| "eval_runtime": 72.5474, | |
| "eval_samples_per_second": 63.062, | |
| "eval_steps_per_second": 3.515, | |
| "step": 20592 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 8.216783216783217e-06, | |
| "loss": 0.0761, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 6.031468531468531e-06, | |
| "loss": 0.074, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.0758, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 1.6608391608391608e-06, | |
| "loss": 0.0747, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9171542537431486, | |
| "eval_loss": 0.4537607729434967, | |
| "eval_runtime": 72.8015, | |
| "eval_samples_per_second": 62.842, | |
| "eval_steps_per_second": 3.503, | |
| "step": 22880 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 22880, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.257616148755644e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |