| { | |
| "best_metric": 0.9404261658430337, | |
| "best_model_checkpoint": "finetuned-epitope_attempt7_150M_200k_default_accuracy-finetuned-epitope_attempt7_150M_200k_weighted_accuracy_1.5/checkpoint-1716", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 17160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.708624708624709e-05, | |
| "loss": -991.1524, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 9.417249417249418e-05, | |
| "loss": -2627.8292, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.125874125874126e-05, | |
| "loss": -4819.024, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -7140.48779296875, | |
| "eval_runtime": 61.4312, | |
| "eval_samples_per_second": 74.474, | |
| "eval_steps_per_second": 3.109, | |
| "step": 1716 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.834498834498835e-05, | |
| "loss": -7373.8975, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 8.543123543123544e-05, | |
| "loss": -10190.031, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.251748251748252e-05, | |
| "loss": -13358.039, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -18116.330078125, | |
| "eval_runtime": 61.6525, | |
| "eval_samples_per_second": 74.206, | |
| "eval_steps_per_second": 3.098, | |
| "step": 3432 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 7.960372960372961e-05, | |
| "loss": -16816.902, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 7.66899766899767e-05, | |
| "loss": -20556.736, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7.377622377622378e-05, | |
| "loss": -24214.914, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 7.086247086247087e-05, | |
| "loss": -28116.02, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -31477.326171875, | |
| "eval_runtime": 61.4757, | |
| "eval_samples_per_second": 74.42, | |
| "eval_steps_per_second": 3.107, | |
| "step": 5148 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 6.794871794871795e-05, | |
| "loss": -32552.834, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.503496503496504e-05, | |
| "loss": -36651.288, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 6.212121212121213e-05, | |
| "loss": -40514.324, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -45798.8046875, | |
| "eval_runtime": 61.7173, | |
| "eval_samples_per_second": 74.128, | |
| "eval_steps_per_second": 3.095, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 5.920745920745921e-05, | |
| "loss": -44842.148, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 5.629370629370629e-05, | |
| "loss": -49187.524, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 5.3379953379953385e-05, | |
| "loss": -53412.06, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 5.0466200466200464e-05, | |
| "loss": -57251.756, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -59889.9140625, | |
| "eval_runtime": 61.4135, | |
| "eval_samples_per_second": 74.495, | |
| "eval_steps_per_second": 3.11, | |
| "step": 8580 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.755244755244756e-05, | |
| "loss": -60731.896, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 4.463869463869464e-05, | |
| "loss": -65162.9, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 4.172494172494173e-05, | |
| "loss": -69186.816, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -72761.71875, | |
| "eval_runtime": 61.5973, | |
| "eval_samples_per_second": 74.273, | |
| "eval_steps_per_second": 3.101, | |
| "step": 10296 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.8811188811188816e-05, | |
| "loss": -72763.624, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 3.58974358974359e-05, | |
| "loss": -76222.128, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 3.298368298368298e-05, | |
| "loss": -78716.456, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 3.0069930069930068e-05, | |
| "loss": -82078.184, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -83606.6171875, | |
| "eval_runtime": 61.5161, | |
| "eval_samples_per_second": 74.371, | |
| "eval_steps_per_second": 3.105, | |
| "step": 12012 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 2.715617715617716e-05, | |
| "loss": -84137.776, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 2.4242424242424244e-05, | |
| "loss": -88317.616, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 2.132867132867133e-05, | |
| "loss": -90301.752, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -91807.8203125, | |
| "eval_runtime": 61.5191, | |
| "eval_samples_per_second": 74.367, | |
| "eval_steps_per_second": 3.105, | |
| "step": 13728 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 1.8414918414918416e-05, | |
| "loss": -91880.464, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 1.5501165501165503e-05, | |
| "loss": -94262.232, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 1.2587412587412589e-05, | |
| "loss": -95277.608, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -96907.4296875, | |
| "eval_runtime": 61.3968, | |
| "eval_samples_per_second": 74.515, | |
| "eval_steps_per_second": 3.111, | |
| "step": 15444 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 9.673659673659673e-06, | |
| "loss": -97058.592, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 6.7599067599067605e-06, | |
| "loss": -97791.208, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": -98873.392, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 9.324009324009324e-07, | |
| "loss": -98503.392, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9404261658430337, | |
| "eval_loss": -98637.3671875, | |
| "eval_runtime": 61.2106, | |
| "eval_samples_per_second": 74.742, | |
| "eval_steps_per_second": 3.12, | |
| "step": 17160 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 17160, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.3502108609523e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |