| { | |
| "best_metric": 0.9248917833931436, | |
| "best_model_checkpoint": "esm2_t30_150M_UR50D-finetuned-epitope_attempt7_150M_200k_weighted_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": -8049.9655, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.562937062937063e-05, | |
| "loss": -23246.98, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 9.344405594405596e-05, | |
| "loss": -44018.38, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.125874125874126e-05, | |
| "loss": -67728.424, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8359226421802908, | |
| "eval_loss": -98722.5546875, | |
| "eval_runtime": 73.589, | |
| "eval_samples_per_second": 62.17, | |
| "eval_steps_per_second": 3.465, | |
| "step": 2288 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.907342657342657e-05, | |
| "loss": -95669.496, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 8.688811188811189e-05, | |
| "loss": -126227.712, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 8.47027972027972e-05, | |
| "loss": -163074.608, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.251748251748252e-05, | |
| "loss": -200426.112, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 8.033216783216784e-05, | |
| "loss": -237543.184, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9248917833931436, | |
| "eval_loss": -266875.46875, | |
| "eval_runtime": 73.356, | |
| "eval_samples_per_second": 62.367, | |
| "eval_steps_per_second": 3.476, | |
| "step": 4576 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 7.814685314685315e-05, | |
| "loss": -280275.456, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 7.596153846153846e-05, | |
| "loss": -323276.256, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7.377622377622378e-05, | |
| "loss": -368565.056, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 7.15909090909091e-05, | |
| "loss": -415972.096, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5617423803932594, | |
| "eval_loss": -476670.71875, | |
| "eval_runtime": 72.4457, | |
| "eval_samples_per_second": 63.151, | |
| "eval_steps_per_second": 3.52, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 6.940559440559441e-05, | |
| "loss": -462468.448, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.722027972027972e-05, | |
| "loss": -515124.992, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.503496503496504e-05, | |
| "loss": -561827.584, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 6.284965034965036e-05, | |
| "loss": -608227.776, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 6.066433566433567e-05, | |
| "loss": -657890.112, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.14814150639170484, | |
| "eval_loss": -704394.6875, | |
| "eval_runtime": 72.4631, | |
| "eval_samples_per_second": 63.136, | |
| "eval_steps_per_second": 3.519, | |
| "step": 9152 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 5.8479020979020984e-05, | |
| "loss": -708670.976, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 5.629370629370629e-05, | |
| "loss": -758351.232, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 5.4108391608391606e-05, | |
| "loss": -812264.64, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 5.192307692307693e-05, | |
| "loss": -862053.248, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8705106128760192, | |
| "eval_loss": -930341.0625, | |
| "eval_runtime": 73.6448, | |
| "eval_samples_per_second": 62.123, | |
| "eval_steps_per_second": 3.463, | |
| "step": 11440 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 4.973776223776224e-05, | |
| "loss": -914762.624, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.755244755244756e-05, | |
| "loss": -952473.472, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 4.5367132867132865e-05, | |
| "loss": -998544.192, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "learning_rate": 4.318181818181819e-05, | |
| "loss": -1049669.696, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 4.0996503496503495e-05, | |
| "loss": -1093231.488, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9136241297193604, | |
| "eval_loss": -1137862.25, | |
| "eval_runtime": 73.362, | |
| "eval_samples_per_second": 62.362, | |
| "eval_steps_per_second": 3.476, | |
| "step": 13728 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.8811188811188816e-05, | |
| "loss": -1124165.632, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.6625874125874124e-05, | |
| "loss": -1166674.048, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 3.4440559440559445e-05, | |
| "loss": -1219658.112, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 3.225524475524476e-05, | |
| "loss": -1264105.6, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 3.0069930069930068e-05, | |
| "loss": -1293580.032, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8931229224892093, | |
| "eval_loss": -1313222.375, | |
| "eval_runtime": 73.2253, | |
| "eval_samples_per_second": 62.478, | |
| "eval_steps_per_second": 3.482, | |
| "step": 16016 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 2.7884615384615386e-05, | |
| "loss": -1324955.904, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 2.5699300699300697e-05, | |
| "loss": -1357552.384, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 2.3513986013986015e-05, | |
| "loss": -1387682.816, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 2.132867132867133e-05, | |
| "loss": -1407950.848, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.14780610726157203, | |
| "eval_loss": -1446230.625, | |
| "eval_runtime": 73.7459, | |
| "eval_samples_per_second": 62.037, | |
| "eval_steps_per_second": 3.458, | |
| "step": 18304 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 1.9143356643356645e-05, | |
| "loss": -1438207.744, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 1.695804195804196e-05, | |
| "loss": -1459738.368, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 1.4772727272727274e-05, | |
| "loss": -1474812.288, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 1.2587412587412589e-05, | |
| "loss": -1490503.296, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 1.0402097902097904e-05, | |
| "loss": -1523451.904, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6420831335064248, | |
| "eval_loss": -1529061.625, | |
| "eval_runtime": 73.2387, | |
| "eval_samples_per_second": 62.467, | |
| "eval_steps_per_second": 3.482, | |
| "step": 20592 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 8.216783216783217e-06, | |
| "loss": -1525384.704, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 6.031468531468531e-06, | |
| "loss": -1541364.48, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": -1543961.344, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 1.6608391608391608e-06, | |
| "loss": -1551367.424, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5396024202483918, | |
| "eval_loss": -1557180.125, | |
| "eval_runtime": 73.3202, | |
| "eval_samples_per_second": 62.398, | |
| "eval_steps_per_second": 3.478, | |
| "step": 22880 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 22880, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.253600112923489e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |