| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.87527352297593, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00437636761487965, | |
| "eval_loss": 2.1039679050445557, | |
| "eval_runtime": 5.4829, | |
| "eval_samples_per_second": 17.692, | |
| "eval_steps_per_second": 8.937, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0437636761487965, | |
| "grad_norm": 1.9867570400238037, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8878, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.087527352297593, | |
| "grad_norm": 1.723362684249878, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4308, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13129102844638948, | |
| "grad_norm": 1.2347863912582397, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3851, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.175054704595186, | |
| "grad_norm": 0.8125748038291931, | |
| "learning_rate": 0.0002, | |
| "loss": 0.324, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2188183807439825, | |
| "grad_norm": 1.0239754915237427, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3551, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2188183807439825, | |
| "eval_loss": 0.35808104276657104, | |
| "eval_runtime": 3.9367, | |
| "eval_samples_per_second": 24.64, | |
| "eval_steps_per_second": 12.447, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.26258205689277897, | |
| "grad_norm": 0.9813992381095886, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3551, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3063457330415755, | |
| "grad_norm": 1.1051058769226074, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3701, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.350109409190372, | |
| "grad_norm": 0.8475605249404907, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2976, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3938730853391685, | |
| "grad_norm": 1.0010513067245483, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2987, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.437636761487965, | |
| "grad_norm": 1.0572749376296997, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3285, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.437636761487965, | |
| "eval_loss": 0.31449094414711, | |
| "eval_runtime": 3.9469, | |
| "eval_samples_per_second": 24.576, | |
| "eval_steps_per_second": 12.415, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4814004376367615, | |
| "grad_norm": 1.2032049894332886, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2868, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5251641137855579, | |
| "grad_norm": 0.9359709024429321, | |
| "learning_rate": 0.0002, | |
| "loss": 0.308, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5689277899343544, | |
| "grad_norm": 1.3596349954605103, | |
| "learning_rate": 0.0002, | |
| "loss": 0.352, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.612691466083151, | |
| "grad_norm": 0.8330109715461731, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2522, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6564551422319475, | |
| "grad_norm": 1.0199737548828125, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2819, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6564551422319475, | |
| "eval_loss": 0.2861652076244354, | |
| "eval_runtime": 3.9483, | |
| "eval_samples_per_second": 24.567, | |
| "eval_steps_per_second": 12.41, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.700218818380744, | |
| "grad_norm": 0.7601907849311829, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2765, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7439824945295405, | |
| "grad_norm": 1.2764638662338257, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2546, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.787746170678337, | |
| "grad_norm": 0.7702401876449585, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2627, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8315098468271335, | |
| "grad_norm": 0.6173223853111267, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2493, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.87527352297593, | |
| "grad_norm": 0.5508137941360474, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2761, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.87527352297593, | |
| "eval_loss": 0.27655091881752014, | |
| "eval_runtime": 3.938, | |
| "eval_samples_per_second": 24.632, | |
| "eval_steps_per_second": 12.443, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.25076769947648e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |