| { | |
| "best_metric": 0.18233025074005127, | |
| "best_model_checkpoint": "finetuned-Leukemia-cell/checkpoint-200", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 268, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019253731343283584, | |
| "loss": 1.7951, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00018507462686567165, | |
| "loss": 1.3572, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00017761194029850748, | |
| "loss": 0.98, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00017014925373134328, | |
| "loss": 0.9394, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00016268656716417911, | |
| "loss": 0.7973, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00015522388059701495, | |
| "loss": 0.8591, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00014776119402985075, | |
| "loss": 0.6357, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00014029850746268658, | |
| "loss": 0.6522, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0001328358208955224, | |
| "loss": 0.495, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00012537313432835822, | |
| "loss": 0.4862, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_accuracy": 0.9172932330827067, | |
| "eval_loss": 0.4092223644256592, | |
| "eval_runtime": 173.193, | |
| "eval_samples_per_second": 1.536, | |
| "eval_steps_per_second": 0.196, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00011791044776119405, | |
| "loss": 0.5047, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00011044776119402987, | |
| "loss": 0.3878, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00010298507462686569, | |
| "loss": 0.43, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.552238805970149e-05, | |
| "loss": 0.3441, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.805970149253732e-05, | |
| "loss": 0.3271, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 8.059701492537314e-05, | |
| "loss": 0.3415, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 7.313432835820896e-05, | |
| "loss": 0.293, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 6.567164179104478e-05, | |
| "loss": 0.2864, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5.82089552238806e-05, | |
| "loss": 0.2873, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 5.074626865671642e-05, | |
| "loss": 0.2095, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9548872180451128, | |
| "eval_loss": 0.18233025074005127, | |
| "eval_runtime": 2.578, | |
| "eval_samples_per_second": 103.183, | |
| "eval_steps_per_second": 13.189, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.328358208955224e-05, | |
| "loss": 0.1822, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 3.582089552238806e-05, | |
| "loss": 0.228, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.835820895522388e-05, | |
| "loss": 0.1377, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 2.0895522388059702e-05, | |
| "loss": 0.2099, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 1.3432835820895523e-05, | |
| "loss": 0.1576, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 5.970149253731343e-06, | |
| "loss": 0.1394, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 268, | |
| "total_flos": 3.292007238315049e+17, | |
| "train_loss": 0.5091555158593761, | |
| "train_runtime": 958.3438, | |
| "train_samples_per_second": 4.433, | |
| "train_steps_per_second": 0.28 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 268, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 3.292007238315049e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |