| { |
| "best_global_step": 3312, |
| "best_metric": 0.26380330324172974, |
| "best_model_checkpoint": "metaclip-2-image-classification/checkpoint-3312", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 3312, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.6038647342995169, |
| "grad_norm": 10.075688362121582, |
| "learning_rate": 0.00017247087676272227, |
| "loss": 0.6110287475585937, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7589396971642186, |
| "eval_loss": 0.4971259534358978, |
| "eval_model_preparation_time": 0.0022, |
| "eval_runtime": 329.8093, |
| "eval_samples_per_second": 80.298, |
| "eval_steps_per_second": 10.039, |
| "step": 828 |
| }, |
| { |
| "epoch": 1.2077294685990339, |
| "grad_norm": 3.758291721343994, |
| "learning_rate": 0.00014181483752299202, |
| "loss": 0.5085984802246094, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.8115942028985508, |
| "grad_norm": 8.271088600158691, |
| "learning_rate": 0.00011115879828326181, |
| "loss": 0.4401962585449219, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8178076501906884, |
| "eval_loss": 0.40332305431365967, |
| "eval_model_preparation_time": 0.0022, |
| "eval_runtime": 330.4601, |
| "eval_samples_per_second": 80.14, |
| "eval_steps_per_second": 10.019, |
| "step": 1656 |
| }, |
| { |
| "epoch": 2.4154589371980677, |
| "grad_norm": 3.3510007858276367, |
| "learning_rate": 8.050275904353158e-05, |
| "loss": 0.37929122924804687, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8700298304572744, |
| "eval_loss": 0.30450552701950073, |
| "eval_model_preparation_time": 0.0022, |
| "eval_runtime": 338.3909, |
| "eval_samples_per_second": 78.262, |
| "eval_steps_per_second": 9.785, |
| "step": 2484 |
| }, |
| { |
| "epoch": 3.0193236714975846, |
| "grad_norm": 3.468003273010254, |
| "learning_rate": 4.984671980380135e-05, |
| "loss": 0.35336199951171876, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.6231884057971016, |
| "grad_norm": 3.735905885696411, |
| "learning_rate": 1.9190680564071123e-05, |
| "loss": 0.3137332763671875, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8917796322168938, |
| "eval_loss": 0.26380330324172974, |
| "eval_model_preparation_time": 0.0022, |
| "eval_runtime": 323.8225, |
| "eval_samples_per_second": 81.782, |
| "eval_steps_per_second": 10.225, |
| "step": 3312 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 3312, |
| "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": 2.065722074755154e+18, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
