| { | |
| "best_global_step": 4396, | |
| "best_metric": 0.9813332557678223, | |
| "best_model_checkpoint": "siglip2-finetune-full/checkpoint-4396", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 4396, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22747952684258416, | |
| "grad_norm": 30.234283447265625, | |
| "learning_rate": 1.7929130234698574e-06, | |
| "loss": 1.2741, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4549590536851683, | |
| "grad_norm": 22.959794998168945, | |
| "learning_rate": 1.5628163828808099e-06, | |
| "loss": 1.1086, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6824385805277525, | |
| "grad_norm": 22.508281707763672, | |
| "learning_rate": 1.3327197422917624e-06, | |
| "loss": 1.0646, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9099181073703366, | |
| "grad_norm": 23.277908325195312, | |
| "learning_rate": 1.102623101702715e-06, | |
| "loss": 1.0494, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5876782642078173, | |
| "eval_loss": 1.0199397802352905, | |
| "eval_model_preparation_time": 0.0025, | |
| "eval_runtime": 1231.0032, | |
| "eval_samples_per_second": 57.133, | |
| "eval_steps_per_second": 7.142, | |
| "step": 2198 | |
| }, | |
| { | |
| "epoch": 1.1373976342129208, | |
| "grad_norm": 25.28997802734375, | |
| "learning_rate": 8.725264611136677e-07, | |
| "loss": 1.0211, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.364877161055505, | |
| "grad_norm": 23.57412338256836, | |
| "learning_rate": 6.424298205246204e-07, | |
| "loss": 0.9994, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5923566878980893, | |
| "grad_norm": 31.955904006958008, | |
| "learning_rate": 4.123331799355729e-07, | |
| "loss": 0.9899, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8198362147406733, | |
| "grad_norm": 23.125858306884766, | |
| "learning_rate": 1.8223653934652553e-07, | |
| "loss": 0.9761, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6032190641395686, | |
| "eval_loss": 0.9813332557678223, | |
| "eval_model_preparation_time": 0.0025, | |
| "eval_runtime": 1219.9593, | |
| "eval_samples_per_second": 57.65, | |
| "eval_steps_per_second": 7.207, | |
| "step": 4396 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4396, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 1.1781704027466584e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |