| { | |
| "best_metric": 0.0002610796655062586, | |
| "best_model_checkpoint": "convnext-base-384-22k-1k-Kontur-competition-1.3K/checkpoint-135", | |
| "epoch": 14.210526315789474, | |
| "eval_steps": 500, | |
| "global_step": 135, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.5273232460021973, | |
| "eval_runtime": 5.3966, | |
| "eval_samples_per_second": 24.645, | |
| "eval_steps_per_second": 0.927, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 14.016679763793945, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.6611, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.15177780389785767, | |
| "eval_runtime": 4.4294, | |
| "eval_samples_per_second": 30.027, | |
| "eval_steps_per_second": 1.129, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 3.7719967365264893, | |
| "learning_rate": 4.75206611570248e-05, | |
| "loss": 0.2686, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_loss": 0.026649044826626778, | |
| "eval_runtime": 4.3155, | |
| "eval_samples_per_second": 30.819, | |
| "eval_steps_per_second": 1.159, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 9.589371681213379, | |
| "learning_rate": 4.338842975206612e-05, | |
| "loss": 0.0899, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.00659002223983407, | |
| "eval_runtime": 4.3662, | |
| "eval_samples_per_second": 30.461, | |
| "eval_steps_per_second": 1.145, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 24.757675170898438, | |
| "learning_rate": 3.925619834710744e-05, | |
| "loss": 0.0379, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_loss": 0.002536825370043516, | |
| "eval_runtime": 4.318, | |
| "eval_samples_per_second": 30.802, | |
| "eval_steps_per_second": 1.158, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 0.6816350221633911, | |
| "learning_rate": 3.512396694214876e-05, | |
| "loss": 0.0202, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.0019522847142070532, | |
| "eval_runtime": 4.4042, | |
| "eval_samples_per_second": 30.199, | |
| "eval_steps_per_second": 1.135, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "grad_norm": 0.10703104734420776, | |
| "learning_rate": 3.099173553719008e-05, | |
| "loss": 0.0048, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "eval_loss": 0.00098791706841439, | |
| "eval_runtime": 4.3638, | |
| "eval_samples_per_second": 30.478, | |
| "eval_steps_per_second": 1.146, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "grad_norm": 4.492315292358398, | |
| "learning_rate": 2.6859504132231405e-05, | |
| "loss": 0.0056, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.001134931342676282, | |
| "eval_runtime": 4.3514, | |
| "eval_samples_per_second": 30.565, | |
| "eval_steps_per_second": 1.149, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "grad_norm": 0.05526771396398544, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.0011, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "eval_loss": 0.0004670162743423134, | |
| "eval_runtime": 4.379, | |
| "eval_samples_per_second": 30.372, | |
| "eval_steps_per_second": 1.142, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "grad_norm": 2.763624906539917, | |
| "learning_rate": 1.859504132231405e-05, | |
| "loss": 0.0017, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.0014447210123762488, | |
| "eval_runtime": 4.3565, | |
| "eval_samples_per_second": 30.529, | |
| "eval_steps_per_second": 1.148, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "grad_norm": 0.29414886236190796, | |
| "learning_rate": 1.4462809917355372e-05, | |
| "loss": 0.0076, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.95, | |
| "eval_loss": 0.0004167805891484022, | |
| "eval_runtime": 4.5875, | |
| "eval_samples_per_second": 28.992, | |
| "eval_steps_per_second": 1.09, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "grad_norm": 0.02740568295121193, | |
| "learning_rate": 1.0330578512396695e-05, | |
| "loss": 0.0018, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.0003142206114716828, | |
| "eval_runtime": 4.7253, | |
| "eval_samples_per_second": 28.146, | |
| "eval_steps_per_second": 1.058, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "grad_norm": 0.06253942847251892, | |
| "learning_rate": 6.198347107438017e-06, | |
| "loss": 0.0027, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "eval_loss": 0.0002625222550705075, | |
| "eval_runtime": 4.4194, | |
| "eval_samples_per_second": 30.094, | |
| "eval_steps_per_second": 1.131, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "grad_norm": 0.3781050741672516, | |
| "learning_rate": 2.066115702479339e-06, | |
| "loss": 0.0008, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.00026170219643972814, | |
| "eval_runtime": 4.4435, | |
| "eval_samples_per_second": 29.931, | |
| "eval_steps_per_second": 1.125, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "eval_loss": 0.0002610796655062586, | |
| "eval_runtime": 4.4227, | |
| "eval_samples_per_second": 30.072, | |
| "eval_steps_per_second": 1.131, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "step": 135, | |
| "total_flos": 3.9447179555061105e+18, | |
| "train_loss": 0.08187244446534249, | |
| "train_runtime": 1269.7055, | |
| "train_samples_per_second": 14.106, | |
| "train_steps_per_second": 0.106 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 135, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 3.9447179555061105e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |