| { |
| "best_global_step": 4908, |
| "best_metric": 0.6553, |
| "best_model_checkpoint": "output/checkpoint-4908", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 6544, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.3056234718826406, |
| "grad_norm": 22.77391815185547, |
| "learning_rate": 4.255239520958084e-05, |
| "loss": 52.1418, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.6112469437652812, |
| "grad_norm": 15.196788787841797, |
| "learning_rate": 2.383982035928144e-05, |
| "loss": 7.7913, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.9168704156479217, |
| "grad_norm": 13.545781135559082, |
| "learning_rate": 5.127245508982037e-06, |
| "loss": 7.43, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 5.054670810699463, |
| "eval_map": 0.623, |
| "eval_map_50": 0.7756, |
| "eval_map_75": 0.6841, |
| "eval_map_barcode": 0.623, |
| "eval_map_large": 0.6594, |
| "eval_map_medium": 0.6674, |
| "eval_map_small": 0.2648, |
| "eval_mar_1": 0.3542, |
| "eval_mar_10": 0.8117, |
| "eval_mar_100": 0.8521, |
| "eval_mar_100_barcode": 0.8521, |
| "eval_mar_large": 0.8749, |
| "eval_mar_medium": 0.8552, |
| "eval_mar_small": 0.5325, |
| "eval_runtime": 71.7672, |
| "eval_samples_per_second": 39.071, |
| "eval_steps_per_second": 4.891, |
| "step": 1636 |
| }, |
| { |
| "epoch": 1.2224938875305624, |
| "grad_norm": 14.290430068969727, |
| "learning_rate": 3.921954314720812e-05, |
| "loss": 7.2123, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.528117359413203, |
| "grad_norm": 15.990227699279785, |
| "learning_rate": 3.6046954314720814e-05, |
| "loss": 7.0995, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.8337408312958434, |
| "grad_norm": 11.737943649291992, |
| "learning_rate": 3.2874365482233505e-05, |
| "loss": 6.9538, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 4.950796604156494, |
| "eval_map": 0.6297, |
| "eval_map_50": 0.7829, |
| "eval_map_75": 0.6878, |
| "eval_map_barcode": 0.6297, |
| "eval_map_large": 0.7109, |
| "eval_map_medium": 0.6449, |
| "eval_map_small": 0.261, |
| "eval_mar_1": 0.3612, |
| "eval_mar_10": 0.8238, |
| "eval_mar_100": 0.8604, |
| "eval_mar_100_barcode": 0.8604, |
| "eval_mar_large": 0.8794, |
| "eval_mar_medium": 0.8624, |
| "eval_mar_small": 0.5972, |
| "eval_runtime": 67.4554, |
| "eval_samples_per_second": 41.568, |
| "eval_steps_per_second": 5.203, |
| "step": 3272 |
| }, |
| { |
| "epoch": 2.139364303178484, |
| "grad_norm": 14.22537899017334, |
| "learning_rate": 2.9701776649746192e-05, |
| "loss": 6.8515, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.444987775061125, |
| "grad_norm": 19.026344299316406, |
| "learning_rate": 2.6529187817258883e-05, |
| "loss": 6.7831, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.750611246943765, |
| "grad_norm": 13.106438636779785, |
| "learning_rate": 2.3356598984771574e-05, |
| "loss": 6.6733, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 5.005562782287598, |
| "eval_map": 0.6553, |
| "eval_map_50": 0.8073, |
| "eval_map_75": 0.7122, |
| "eval_map_barcode": 0.6553, |
| "eval_map_large": 0.7359, |
| "eval_map_medium": 0.6782, |
| "eval_map_small": 0.2319, |
| "eval_mar_1": 0.3481, |
| "eval_mar_10": 0.8201, |
| "eval_mar_100": 0.8542, |
| "eval_mar_100_barcode": 0.8542, |
| "eval_mar_large": 0.8712, |
| "eval_mar_medium": 0.8573, |
| "eval_mar_small": 0.6107, |
| "eval_runtime": 62.2135, |
| "eval_samples_per_second": 45.071, |
| "eval_steps_per_second": 5.642, |
| "step": 4908 |
| }, |
| { |
| "epoch": 3.056234718826406, |
| "grad_norm": 15.877172470092773, |
| "learning_rate": 2.0184010152284265e-05, |
| "loss": 6.6537, |
| "step": 5000 |
| }, |
| { |
| "epoch": 3.3618581907090466, |
| "grad_norm": 21.57083511352539, |
| "learning_rate": 1.7011421319796956e-05, |
| "loss": 6.5493, |
| "step": 5500 |
| }, |
| { |
| "epoch": 3.667481662591687, |
| "grad_norm": 11.728562355041504, |
| "learning_rate": 1.3838832487309645e-05, |
| "loss": 6.4896, |
| "step": 6000 |
| }, |
| { |
| "epoch": 3.9731051344743276, |
| "grad_norm": 9.994644165039062, |
| "learning_rate": 1.0666243654822336e-05, |
| "loss": 6.5783, |
| "step": 6500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 5.049291610717773, |
| "eval_map": 0.6524, |
| "eval_map_50": 0.808, |
| "eval_map_75": 0.711, |
| "eval_map_barcode": 0.6524, |
| "eval_map_large": 0.7472, |
| "eval_map_medium": 0.6782, |
| "eval_map_small": 0.2391, |
| "eval_mar_1": 0.3581, |
| "eval_mar_10": 0.8268, |
| "eval_mar_100": 0.8629, |
| "eval_mar_100_barcode": 0.8629, |
| "eval_mar_large": 0.8819, |
| "eval_mar_medium": 0.8639, |
| "eval_mar_small": 0.6066, |
| "eval_runtime": 62.2975, |
| "eval_samples_per_second": 45.01, |
| "eval_steps_per_second": 5.634, |
| "step": 6544 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 8180, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "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": 4.357726825476096e+18, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
