| { | |
| "best_metric": 1.88290536403656, | |
| "best_model_checkpoint": "$../logs/03-08-2024T09-55-11-detailclip-b32/checkpoint-58", | |
| "epoch": 0.5043478260869565, | |
| "eval_steps": 58, | |
| "global_step": 58, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05217391304347826, | |
| "grad_norm": 2.7094666957855225, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.7099, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10434782608695652, | |
| "grad_norm": 2.6887893676757812, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.6948, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1565217391304348, | |
| "grad_norm": 2.56785249710083, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7276, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20869565217391303, | |
| "grad_norm": 2.413269281387329, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5937, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 2.0822973251342773, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4702, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3130434782608696, | |
| "grad_norm": 1.9383469820022583, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.4579, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3652173913043478, | |
| "grad_norm": 1.6925252676010132, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.4028, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.41739130434782606, | |
| "grad_norm": 1.5693070888519287, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.3466, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.46956521739130436, | |
| "grad_norm": 1.8624712228775024, | |
| "learning_rate": 9.999673735634258e-05, | |
| "loss": 0.31, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5043478260869565, | |
| "eval_loss": 1.88290536403656, | |
| "eval_runtime": 371.0087, | |
| "eval_samples_per_second": 6.021, | |
| "eval_steps_per_second": 0.049, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.5043478260869565, | |
| "neg_increments": 0.40031376906933136, | |
| "positive_increments": 0.6000771337979885, | |
| "rate_increments": 0.60599292530573, | |
| "step": 58 | |
| } | |
| ], | |
| "logging_steps": 6, | |
| "max_steps": 1150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 58, | |
| "total_flos": 1318444128207360.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |