| { | |
| "best_global_step": 1688, | |
| "best_metric": 0.07286661118268967, | |
| "best_model_checkpoint": "siglip2-finetune-full/checkpoint-1688", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 1688, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5924170616113744, | |
| "grad_norm": 64.00031280517578, | |
| "learning_rate": 1.729404690318701e-06, | |
| "loss": 0.544, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9547777777777777, | |
| "eval_loss": 0.14658287167549133, | |
| "eval_model_preparation_time": 0.0047, | |
| "eval_runtime": 329.9419, | |
| "eval_samples_per_second": 81.833, | |
| "eval_steps_per_second": 10.229, | |
| "step": 844 | |
| }, | |
| { | |
| "epoch": 1.1848341232227488, | |
| "grad_norm": 31.858619689941406, | |
| "learning_rate": 1.428743235117258e-06, | |
| "loss": 0.1177, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7772511848341233, | |
| "grad_norm": 52.434165954589844, | |
| "learning_rate": 1.1280817799158147e-06, | |
| "loss": 0.0904, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9771111111111112, | |
| "eval_loss": 0.07286661118268967, | |
| "eval_model_preparation_time": 0.0047, | |
| "eval_runtime": 326.6893, | |
| "eval_samples_per_second": 82.647, | |
| "eval_steps_per_second": 10.331, | |
| "step": 1688 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3376, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.523096813985792e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |