| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9938398357289527, | |
| "eval_steps": 500, | |
| "global_step": 121, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04106776180698152, | |
| "grad_norm": 61.824676513671875, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 4.1884, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08213552361396304, | |
| "grad_norm": 10.049797058105469, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.3499, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12320328542094455, | |
| "grad_norm": 6.188166618347168, | |
| "learning_rate": 1.9983081582712684e-05, | |
| "loss": 0.5711, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16427104722792607, | |
| "grad_norm": 5.229387283325195, | |
| "learning_rate": 1.9793406217655516e-05, | |
| "loss": 0.3699, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2053388090349076, | |
| "grad_norm": 5.0885233879089355, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.2933, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2464065708418891, | |
| "grad_norm": 7.334043025970459, | |
| "learning_rate": 1.880201391180111e-05, | |
| "loss": 0.243, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2874743326488706, | |
| "grad_norm": 4.5426764488220215, | |
| "learning_rate": 1.802123192755044e-05, | |
| "loss": 0.242, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.32854209445585214, | |
| "grad_norm": 3.701889991760254, | |
| "learning_rate": 1.7071067811865477e-05, | |
| "loss": 0.169, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.36960985626283366, | |
| "grad_norm": 2.314136505126953, | |
| "learning_rate": 1.5971585917027864e-05, | |
| "loss": 0.1295, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4106776180698152, | |
| "grad_norm": 1.4874415397644043, | |
| "learning_rate": 1.4746003697476406e-05, | |
| "loss": 0.1007, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4517453798767967, | |
| "grad_norm": 1.0777748823165894, | |
| "learning_rate": 1.342020143325669e-05, | |
| "loss": 0.084, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.4928131416837782, | |
| "grad_norm": 1.0137399435043335, | |
| "learning_rate": 1.2022175723320382e-05, | |
| "loss": 0.0777, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5338809034907598, | |
| "grad_norm": 1.8608695268630981, | |
| "learning_rate": 1.0581448289104759e-05, | |
| "loss": 0.0761, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5749486652977412, | |
| "grad_norm": 1.3916422128677368, | |
| "learning_rate": 9.128442572523418e-06, | |
| "loss": 0.0744, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6160164271047228, | |
| "grad_norm": 1.410448431968689, | |
| "learning_rate": 7.6938412925756e-06, | |
| "loss": 0.0732, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6570841889117043, | |
| "grad_norm": 0.7402985095977783, | |
| "learning_rate": 6.3079385268731575e-06, | |
| "loss": 0.0708, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6981519507186859, | |
| "grad_norm": 0.5348497629165649, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.0705, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.7392197125256673, | |
| "grad_norm": 0.5039529204368591, | |
| "learning_rate": 3.797645087317401e-06, | |
| "loss": 0.0696, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7802874743326489, | |
| "grad_norm": 0.39937153458595276, | |
| "learning_rate": 2.726263584269513e-06, | |
| "loss": 0.0685, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.8213552361396304, | |
| "grad_norm": 0.40543800592422485, | |
| "learning_rate": 1.808479557110081e-06, | |
| "loss": 0.0682, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8624229979466119, | |
| "grad_norm": 0.3837692439556122, | |
| "learning_rate": 1.0636735967658785e-06, | |
| "loss": 0.0678, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.9034907597535934, | |
| "grad_norm": 0.3506762981414795, | |
| "learning_rate": 5.075735642696611e-07, | |
| "loss": 0.0676, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.944558521560575, | |
| "grad_norm": 0.3370087444782257, | |
| "learning_rate": 1.519224698779198e-07, | |
| "loss": 0.0676, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.9856262833675564, | |
| "grad_norm": 0.30759868025779724, | |
| "learning_rate": 4.230499177994007e-09, | |
| "loss": 0.0677, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9938398357289527, | |
| "step": 121, | |
| "total_flos": 2.0357450272079872e+17, | |
| "train_loss": 0.3584386111537287, | |
| "train_runtime": 8351.8216, | |
| "train_samples_per_second": 8.381, | |
| "train_steps_per_second": 0.014 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 121, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.0357450272079872e+17, | |
| "train_batch_size": 18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |