| { |
| "best_global_step": 6000, |
| "best_metric": 0.25916826725006104, |
| "best_model_checkpoint": "siglip2-image-classification/checkpoint-6000", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 6000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 3.479196786880493, |
| "learning_rate": 0.00018490756302521008, |
| "loss": 0.5047473754882813, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 4.285048007965088, |
| "learning_rate": 0.00016810084033613447, |
| "loss": 0.42015097045898436, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.655521869659424, |
| "learning_rate": 0.00015129411764705882, |
| "loss": 0.38898434448242186, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8312134754211069, |
| "eval_loss": 0.37709423899650574, |
| "eval_model_preparation_time": 0.0019, |
| "eval_runtime": 283.3202, |
| "eval_samples_per_second": 112.943, |
| "eval_steps_per_second": 14.118, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 1.3074887990951538, |
| "learning_rate": 0.0001344873949579832, |
| "loss": 0.3693521728515625, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 1.2879441976547241, |
| "learning_rate": 0.00011768067226890757, |
| "loss": 0.3484758605957031, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.5275274515151978, |
| "learning_rate": 0.00010087394957983194, |
| "loss": 0.34061029052734376, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8569642801337541, |
| "eval_loss": 0.3550371527671814, |
| "eval_model_preparation_time": 0.0019, |
| "eval_runtime": 282.8579, |
| "eval_samples_per_second": 113.127, |
| "eval_steps_per_second": 14.141, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 4.5269551277160645, |
| "learning_rate": 8.406722689075631e-05, |
| "loss": 0.31870977783203125, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 1.9202643632888794, |
| "learning_rate": 6.726050420168067e-05, |
| "loss": 0.30601947021484377, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.203441858291626, |
| "learning_rate": 5.045378151260505e-05, |
| "loss": 0.2938027648925781, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8811525360167505, |
| "eval_loss": 0.2799703776836395, |
| "eval_model_preparation_time": 0.0019, |
| "eval_runtime": 287.094, |
| "eval_samples_per_second": 111.458, |
| "eval_steps_per_second": 13.933, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 3.2569119930267334, |
| "learning_rate": 3.364705882352941e-05, |
| "loss": 0.2793130798339844, |
| "step": 5000 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 3.2401018142700195, |
| "learning_rate": 1.6840336134453782e-05, |
| "loss": 0.26365960693359375, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.6921696662902832, |
| "learning_rate": 3.3613445378151266e-08, |
| "loss": 0.2508096923828125, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8935591737241789, |
| "eval_loss": 0.25916826725006104, |
| "eval_model_preparation_time": 0.0019, |
| "eval_runtime": 286.6404, |
| "eval_samples_per_second": 111.635, |
| "eval_steps_per_second": 13.955, |
| "step": 6000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 6000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "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": 1.6080385153725334e+19, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
