| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.315789473684211, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 29.249841690063477, | |
| "learning_rate": 5.0000000000000004e-08, | |
| "loss": 2.3522, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 28.9044246673584, | |
| "learning_rate": 1.0000000000000001e-07, | |
| "loss": 2.3252, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 29.364099502563477, | |
| "learning_rate": 1.5000000000000002e-07, | |
| "loss": 2.3516, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 29.600059509277344, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 2.3531, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 28.67375373840332, | |
| "learning_rate": 2.5000000000000004e-07, | |
| "loss": 2.2996, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 28.520565032958984, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 2.261, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 28.520565032958984, | |
| "learning_rate": 3.5000000000000004e-07, | |
| "loss": 2.4103, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.1578947368421053, | |
| "grad_norm": 50.07780075073242, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 2.3406, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.3157894736842106, | |
| "grad_norm": 27.654495239257812, | |
| "learning_rate": 4.5000000000000003e-07, | |
| "loss": 2.2567, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.4736842105263157, | |
| "grad_norm": 29.456050872802734, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 2.2952, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.631578947368421, | |
| "grad_norm": 29.20940399169922, | |
| "learning_rate": 5.5e-07, | |
| "loss": 2.2942, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.7894736842105263, | |
| "grad_norm": 29.31597328186035, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 2.1754, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.9473684210526314, | |
| "grad_norm": 28.444318771362305, | |
| "learning_rate": 6.5e-07, | |
| "loss": 2.1927, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 28.444318771362305, | |
| "learning_rate": 7.000000000000001e-07, | |
| "loss": 2.1262, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.1578947368421053, | |
| "grad_norm": 47.70970153808594, | |
| "learning_rate": 7.5e-07, | |
| "loss": 2.1372, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.3157894736842106, | |
| "grad_norm": 27.695999145507812, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 2.0525, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.473684210526316, | |
| "grad_norm": 27.79825210571289, | |
| "learning_rate": 8.500000000000001e-07, | |
| "loss": 2.0136, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 26.69138526916504, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 1.8875, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.7894736842105265, | |
| "grad_norm": 26.22272300720215, | |
| "learning_rate": 9.500000000000001e-07, | |
| "loss": 1.7887, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.9473684210526314, | |
| "grad_norm": 25.795120239257812, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.6733, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 40.42267990112305, | |
| "learning_rate": 1.0500000000000001e-06, | |
| "loss": 1.6464, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 24.54705810546875, | |
| "learning_rate": 1.1e-06, | |
| "loss": 1.5408, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.3157894736842106, | |
| "grad_norm": 24.196022033691406, | |
| "learning_rate": 1.1500000000000002e-06, | |
| "loss": 1.3596, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.473684210526316, | |
| "grad_norm": 24.2416934967041, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 1.2173, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.6315789473684212, | |
| "grad_norm": 24.014326095581055, | |
| "learning_rate": 1.25e-06, | |
| "loss": 1.0318, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.7894736842105265, | |
| "grad_norm": 23.31243324279785, | |
| "learning_rate": 1.3e-06, | |
| "loss": 0.8277, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.9473684210526314, | |
| "grad_norm": 20.742502212524414, | |
| "learning_rate": 1.3500000000000002e-06, | |
| "loss": 0.6301, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 20.742502212524414, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 0.4823, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.157894736842105, | |
| "grad_norm": 28.484569549560547, | |
| "learning_rate": 1.45e-06, | |
| "loss": 0.4436, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.315789473684211, | |
| "grad_norm": 12.672039031982422, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.3486, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 36, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 6, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7584034368286884e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |