| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 12120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.41254125412541254, | |
| "grad_norm": 2.3108015060424805, | |
| "learning_rate": 4.793729372937294e-05, | |
| "loss": 2.2146, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8250825082508251, | |
| "grad_norm": 2.0797488689422607, | |
| "learning_rate": 4.5874587458745876e-05, | |
| "loss": 2.1084, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.2376237623762376, | |
| "grad_norm": 3.0814099311828613, | |
| "learning_rate": 4.3811881188118816e-05, | |
| "loss": 2.0322, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6501650165016502, | |
| "grad_norm": 2.495213508605957, | |
| "learning_rate": 4.174917491749175e-05, | |
| "loss": 2.0345, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.062706270627063, | |
| "grad_norm": 2.1899046897888184, | |
| "learning_rate": 3.968646864686469e-05, | |
| "loss": 2.0033, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.4752475247524752, | |
| "grad_norm": 1.9743722677230835, | |
| "learning_rate": 3.762376237623763e-05, | |
| "loss": 1.9696, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.887788778877888, | |
| "grad_norm": 2.0741982460021973, | |
| "learning_rate": 3.556105610561056e-05, | |
| "loss": 1.9576, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.3003300330033003, | |
| "grad_norm": 2.478909492492676, | |
| "learning_rate": 3.34983498349835e-05, | |
| "loss": 1.9328, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.7128712871287126, | |
| "grad_norm": 1.9436851739883423, | |
| "learning_rate": 3.1435643564356435e-05, | |
| "loss": 1.9219, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.125412541254126, | |
| "grad_norm": 2.5008606910705566, | |
| "learning_rate": 2.9372937293729375e-05, | |
| "loss": 1.9088, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.537953795379538, | |
| "grad_norm": 2.509181499481201, | |
| "learning_rate": 2.731023102310231e-05, | |
| "loss": 1.9007, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.9504950495049505, | |
| "grad_norm": 2.128865957260132, | |
| "learning_rate": 2.5247524752475248e-05, | |
| "loss": 1.8946, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.363036303630363, | |
| "grad_norm": 2.622591972351074, | |
| "learning_rate": 2.3184818481848185e-05, | |
| "loss": 1.8833, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.775577557755776, | |
| "grad_norm": 2.249598264694214, | |
| "learning_rate": 2.1122112211221125e-05, | |
| "loss": 1.8689, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.188118811881188, | |
| "grad_norm": 2.382103681564331, | |
| "learning_rate": 1.905940594059406e-05, | |
| "loss": 1.8549, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.600660066006601, | |
| "grad_norm": 2.8902101516723633, | |
| "learning_rate": 1.6996699669966998e-05, | |
| "loss": 1.8497, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.013201320132013, | |
| "grad_norm": 3.372351884841919, | |
| "learning_rate": 1.4933993399339935e-05, | |
| "loss": 1.8578, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.425742574257426, | |
| "grad_norm": 2.1681647300720215, | |
| "learning_rate": 1.2871287128712873e-05, | |
| "loss": 1.8376, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.838283828382838, | |
| "grad_norm": 2.148872137069702, | |
| "learning_rate": 1.080858085808581e-05, | |
| "loss": 1.846, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.250825082508252, | |
| "grad_norm": 2.4388763904571533, | |
| "learning_rate": 8.745874587458746e-06, | |
| "loss": 1.825, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.663366336633663, | |
| "grad_norm": 3.16158127784729, | |
| "learning_rate": 6.6831683168316835e-06, | |
| "loss": 1.8288, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.075907590759076, | |
| "grad_norm": 2.0919010639190674, | |
| "learning_rate": 4.62046204620462e-06, | |
| "loss": 1.8287, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 9.488448844884488, | |
| "grad_norm": 2.2496068477630615, | |
| "learning_rate": 2.557755775577558e-06, | |
| "loss": 1.8293, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 9.900990099009901, | |
| "grad_norm": 3.0332181453704834, | |
| "learning_rate": 4.950495049504951e-07, | |
| "loss": 1.8216, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 12120, | |
| "total_flos": 1.1470348787122176e+16, | |
| "train_loss": 1.9159377371910775, | |
| "train_runtime": 858.1835, | |
| "train_samples_per_second": 112.948, | |
| "train_steps_per_second": 14.123 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 1.1470348787122176e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |