| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 3180, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6351848244667053, | |
| "learning_rate": 1.8006289308176103e-05, | |
| "loss": 0.3969, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5796774193548387, | |
| "eval_loss": 0.19846348464488983, | |
| "eval_runtime": 1.3983, | |
| "eval_samples_per_second": 2216.976, | |
| "eval_steps_per_second": 46.485, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.533223032951355, | |
| "learning_rate": 1.6012578616352204e-05, | |
| "loss": 0.165, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8316129032258065, | |
| "eval_loss": 0.09574232250452042, | |
| "eval_runtime": 1.4234, | |
| "eval_samples_per_second": 2177.865, | |
| "eval_steps_per_second": 45.665, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.4442247450351715, | |
| "learning_rate": 1.4018867924528304e-05, | |
| "loss": 0.1023, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8812903225806452, | |
| "eval_loss": 0.06304766982793808, | |
| "eval_runtime": 1.6221, | |
| "eval_samples_per_second": 1911.159, | |
| "eval_steps_per_second": 40.073, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.38984864950180054, | |
| "learning_rate": 1.2025157232704403e-05, | |
| "loss": 0.0763, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9009677419354839, | |
| "eval_loss": 0.04724794998764992, | |
| "eval_runtime": 1.82, | |
| "eval_samples_per_second": 1703.326, | |
| "eval_steps_per_second": 35.715, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.2871840000152588, | |
| "learning_rate": 1.0031446540880504e-05, | |
| "loss": 0.0622, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.91, | |
| "eval_loss": 0.03912010416388512, | |
| "eval_runtime": 1.6351, | |
| "eval_samples_per_second": 1895.862, | |
| "eval_steps_per_second": 39.752, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.3613884747028351, | |
| "learning_rate": 8.037735849056606e-06, | |
| "loss": 0.054, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9167741935483871, | |
| "eval_loss": 0.03382793441414833, | |
| "eval_runtime": 1.64, | |
| "eval_samples_per_second": 1890.201, | |
| "eval_steps_per_second": 39.633, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.2407248318195343, | |
| "learning_rate": 6.044025157232704e-06, | |
| "loss": 0.0489, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9229032258064516, | |
| "eval_loss": 0.031094178557395935, | |
| "eval_runtime": 1.452, | |
| "eval_samples_per_second": 2134.921, | |
| "eval_steps_per_second": 44.764, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.26211392879486084, | |
| "learning_rate": 4.0503144654088055e-06, | |
| "loss": 0.0457, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.925483870967742, | |
| "eval_loss": 0.028951287269592285, | |
| "eval_runtime": 1.6426, | |
| "eval_samples_per_second": 1887.24, | |
| "eval_steps_per_second": 39.571, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.26027989387512207, | |
| "learning_rate": 2.056603773584906e-06, | |
| "loss": 0.0434, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9280645161290323, | |
| "eval_loss": 0.027912667021155357, | |
| "eval_runtime": 1.6391, | |
| "eval_samples_per_second": 1891.318, | |
| "eval_steps_per_second": 39.657, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.968553459119496, | |
| "grad_norm": 0.2062142789363861, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0423, | |
| "step": 3170 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 825254092458012.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6561589068494422, | |
| "num_train_epochs": 10, | |
| "temperature": 11 | |
| } | |
| } | |