| { |
| "best_global_step": 200, |
| "best_metric": 0.9924812030075187, |
| "best_model_checkpoint": "./vit-base-beans-model-finetuned/checkpoint-200", |
| "epoch": 4.0, |
| "eval_steps": 50, |
| "global_step": 260, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.15384615384615385, |
| "grad_norm": 0.046927113085985184, |
| "learning_rate": 0.00019230769230769233, |
| "loss": 0.0203, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.3076923076923077, |
| "grad_norm": 9.78127670288086, |
| "learning_rate": 0.00018461538461538463, |
| "loss": 0.0936, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.46153846153846156, |
| "grad_norm": 0.15273386240005493, |
| "learning_rate": 0.00017692307692307693, |
| "loss": 0.1531, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6153846153846154, |
| "grad_norm": 0.059062909334897995, |
| "learning_rate": 0.00016923076923076923, |
| "loss": 0.0808, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7692307692307693, |
| "grad_norm": 0.04636870697140694, |
| "learning_rate": 0.00016153846153846155, |
| "loss": 0.0492, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.7692307692307693, |
| "eval_accuracy": 0.9774436090225563, |
| "eval_f1": 0.9774379108058134, |
| "eval_loss": 0.1117313951253891, |
| "eval_precision": 0.9775956574956096, |
| "eval_recall": 0.9774436090225563, |
| "eval_runtime": 1.2431, |
| "eval_samples_per_second": 106.99, |
| "eval_steps_per_second": 13.675, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9230769230769231, |
| "grad_norm": 0.03537413850426674, |
| "learning_rate": 0.00015384615384615385, |
| "loss": 0.1339, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.0769230769230769, |
| "grad_norm": 8.806337356567383, |
| "learning_rate": 0.00014615384615384615, |
| "loss": 0.0454, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.2307692307692308, |
| "grad_norm": 10.037283897399902, |
| "learning_rate": 0.00013846153846153847, |
| "loss": 0.0988, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.3846153846153846, |
| "grad_norm": 0.030385395511984825, |
| "learning_rate": 0.00013076923076923077, |
| "loss": 0.0539, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.5384615384615383, |
| "grad_norm": 0.03275859355926514, |
| "learning_rate": 0.0001230769230769231, |
| "loss": 0.0432, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.5384615384615383, |
| "eval_accuracy": 0.9624060150375939, |
| "eval_f1": 0.9619558041418667, |
| "eval_loss": 0.14281876385211945, |
| "eval_precision": 0.9650282168005047, |
| "eval_recall": 0.9624060150375939, |
| "eval_runtime": 1.6279, |
| "eval_samples_per_second": 81.699, |
| "eval_steps_per_second": 10.443, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.6923076923076923, |
| "grad_norm": 0.030820081010460854, |
| "learning_rate": 0.00011538461538461538, |
| "loss": 0.0186, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.8461538461538463, |
| "grad_norm": 0.027867253869771957, |
| "learning_rate": 0.0001076923076923077, |
| "loss": 0.0827, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.03597959503531456, |
| "learning_rate": 0.0001, |
| "loss": 0.0319, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.1538461538461537, |
| "grad_norm": 0.12349524348974228, |
| "learning_rate": 9.230769230769232e-05, |
| "loss": 0.0368, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.3076923076923075, |
| "grad_norm": 0.3923969566822052, |
| "learning_rate": 8.461538461538461e-05, |
| "loss": 0.0119, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.3076923076923075, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_f1": 0.9849605066094567, |
| "eval_loss": 0.07355950772762299, |
| "eval_precision": 0.9851258581235698, |
| "eval_recall": 0.9849624060150376, |
| "eval_runtime": 1.2553, |
| "eval_samples_per_second": 105.955, |
| "eval_steps_per_second": 13.543, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.4615384615384617, |
| "grad_norm": 5.8568878173828125, |
| "learning_rate": 7.692307692307693e-05, |
| "loss": 0.0635, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.6153846153846154, |
| "grad_norm": 0.03230699151754379, |
| "learning_rate": 6.923076923076924e-05, |
| "loss": 0.0067, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.769230769230769, |
| "grad_norm": 0.02458355203270912, |
| "learning_rate": 6.153846153846155e-05, |
| "loss": 0.0102, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.9230769230769234, |
| "grad_norm": 0.017980379983782768, |
| "learning_rate": 5.384615384615385e-05, |
| "loss": 0.0055, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.076923076923077, |
| "grad_norm": 0.017595401033759117, |
| "learning_rate": 4.615384615384616e-05, |
| "loss": 0.0038, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.076923076923077, |
| "eval_accuracy": 0.9924812030075187, |
| "eval_f1": 0.9924802319630927, |
| "eval_loss": 0.04042869806289673, |
| "eval_precision": 0.9926482873851294, |
| "eval_recall": 0.9924812030075187, |
| "eval_runtime": 1.2785, |
| "eval_samples_per_second": 104.031, |
| "eval_steps_per_second": 13.297, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.230769230769231, |
| "grad_norm": 0.016184426844120026, |
| "learning_rate": 3.846153846153846e-05, |
| "loss": 0.0036, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.3846153846153846, |
| "grad_norm": 0.016925645992159843, |
| "learning_rate": 3.0769230769230774e-05, |
| "loss": 0.0037, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.5384615384615383, |
| "grad_norm": 0.018014749512076378, |
| "learning_rate": 2.307692307692308e-05, |
| "loss": 0.003, |
| "step": 230 |
| }, |
| { |
| "epoch": 3.6923076923076925, |
| "grad_norm": 0.016459761187434196, |
| "learning_rate": 1.5384615384615387e-05, |
| "loss": 0.003, |
| "step": 240 |
| }, |
| { |
| "epoch": 3.8461538461538463, |
| "grad_norm": 0.015728654339909554, |
| "learning_rate": 7.692307692307694e-06, |
| "loss": 0.0046, |
| "step": 250 |
| }, |
| { |
| "epoch": 3.8461538461538463, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_f1": 0.9850440384831379, |
| "eval_loss": 0.04719630256295204, |
| "eval_precision": 0.9856162144491664, |
| "eval_recall": 0.9849624060150376, |
| "eval_runtime": 1.2361, |
| "eval_samples_per_second": 107.601, |
| "eval_steps_per_second": 13.753, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.015585985034704208, |
| "learning_rate": 0.0, |
| "loss": 0.0029, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 260, |
| "total_flos": 3.205097416476426e+17, |
| "train_loss": 0.04095530418249277, |
| "train_runtime": 294.0083, |
| "train_samples_per_second": 14.068, |
| "train_steps_per_second": 0.884 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 260, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.205097416476426e+17, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
