| { | |
| "best_metric": 0.8981937602627258, | |
| "best_model_checkpoint": "dinov2-small-imagenet1k-1-layer-finetuned-noh\\checkpoint-69", | |
| "epoch": 9.577777777777778, | |
| "eval_steps": 500, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 159.195556640625, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.6375, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 104.19737243652344, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.5315, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.24794745484400657, | |
| "eval_loss": 1.2674404382705688, | |
| "eval_runtime": 77.3353, | |
| "eval_samples_per_second": 7.875, | |
| "eval_steps_per_second": 0.504, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.3111111111111111, | |
| "grad_norm": 71.8785400390625, | |
| "learning_rate": 4.797979797979798e-05, | |
| "loss": 0.6264, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7555555555555555, | |
| "grad_norm": 95.69356536865234, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.4629, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7881773399014779, | |
| "eval_loss": 0.5133813619613647, | |
| "eval_runtime": 70.8468, | |
| "eval_samples_per_second": 8.596, | |
| "eval_steps_per_second": 0.55, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.1777777777777776, | |
| "grad_norm": 193.81893920898438, | |
| "learning_rate": 4.292929292929293e-05, | |
| "loss": 0.3744, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.6222222222222222, | |
| "grad_norm": 42.527183532714844, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.4368, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8981937602627258, | |
| "eval_loss": 0.3057605028152466, | |
| "eval_runtime": 67.9327, | |
| "eval_samples_per_second": 8.965, | |
| "eval_steps_per_second": 0.574, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0444444444444443, | |
| "grad_norm": 112.69408416748047, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.4417, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.488888888888889, | |
| "grad_norm": 51.07742691040039, | |
| "learning_rate": 3.535353535353535e-05, | |
| "loss": 0.3974, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.9333333333333336, | |
| "grad_norm": 114.08860778808594, | |
| "learning_rate": 3.282828282828283e-05, | |
| "loss": 0.4123, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.41477170586586, | |
| "eval_runtime": 67.1988, | |
| "eval_samples_per_second": 9.063, | |
| "eval_steps_per_second": 0.58, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "grad_norm": 57.1572265625, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.2962, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 73.67029571533203, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3301, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8735632183908046, | |
| "eval_loss": 0.35200509428977966, | |
| "eval_runtime": 67.892, | |
| "eval_samples_per_second": 8.97, | |
| "eval_steps_per_second": 0.574, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.222222222222222, | |
| "grad_norm": 72.7608871459961, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.2722, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.666666666666667, | |
| "grad_norm": 54.74464797973633, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.2907, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8440065681444991, | |
| "eval_loss": 0.441454142332077, | |
| "eval_runtime": 67.5009, | |
| "eval_samples_per_second": 9.022, | |
| "eval_steps_per_second": 0.578, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.088888888888889, | |
| "grad_norm": 79.50174713134766, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.2261, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.533333333333333, | |
| "grad_norm": 89.56454467773438, | |
| "learning_rate": 1.7676767676767676e-05, | |
| "loss": 0.2502, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.977777777777778, | |
| "grad_norm": 181.8550567626953, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.2809, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7520525451559934, | |
| "eval_loss": 0.5785913467407227, | |
| "eval_runtime": 74.2006, | |
| "eval_samples_per_second": 8.207, | |
| "eval_steps_per_second": 0.526, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 53.69293212890625, | |
| "learning_rate": 1.2626262626262628e-05, | |
| "loss": 0.2154, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.844444444444444, | |
| "grad_norm": 48.94661331176758, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.2243, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8752052545155994, | |
| "eval_loss": 0.4724029004573822, | |
| "eval_runtime": 70.7275, | |
| "eval_samples_per_second": 8.611, | |
| "eval_steps_per_second": 0.551, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.266666666666667, | |
| "grad_norm": 52.60847854614258, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.1612, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.71111111111111, | |
| "grad_norm": 54.77922058105469, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.1968, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8702791461412152, | |
| "eval_loss": 0.5452057719230652, | |
| "eval_runtime": 69.5328, | |
| "eval_samples_per_second": 8.758, | |
| "eval_steps_per_second": 0.561, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.133333333333333, | |
| "grad_norm": 76.0404281616211, | |
| "learning_rate": 2.5252525252525253e-06, | |
| "loss": 0.1506, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "grad_norm": 52.908023834228516, | |
| "learning_rate": 0.0, | |
| "loss": 0.1601, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "eval_accuracy": 0.8440065681444991, | |
| "eval_loss": 0.5386197566986084, | |
| "eval_runtime": 71.1162, | |
| "eval_samples_per_second": 8.563, | |
| "eval_steps_per_second": 0.548, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "step": 220, | |
| "total_flos": 2.7968752299762893e+17, | |
| "train_loss": 0.335261548649181, | |
| "train_runtime": 2511.8857, | |
| "train_samples_per_second": 5.729, | |
| "train_steps_per_second": 0.088 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 220, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.7968752299762893e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |