| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.3264033264033266, | |
| "eval_steps": 2000, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2079002079002079, | |
| "grad_norm": 45.465179443359375, | |
| "learning_rate": 1.8960498960498964e-05, | |
| "loss": 3.2736, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4158004158004158, | |
| "grad_norm": 36.97869110107422, | |
| "learning_rate": 1.7920997920997923e-05, | |
| "loss": 1.9237, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6237006237006237, | |
| "grad_norm": 26.419578552246094, | |
| "learning_rate": 1.6881496881496885e-05, | |
| "loss": 1.6396, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8316008316008316, | |
| "grad_norm": 46.70088195800781, | |
| "learning_rate": 1.5841995841995844e-05, | |
| "loss": 1.5594, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8316008316008316, | |
| "eval_loss": 1.0434484481811523, | |
| "eval_runtime": 75.5621, | |
| "eval_samples_per_second": 18.832, | |
| "eval_steps_per_second": 4.711, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0395010395010396, | |
| "grad_norm": 36.45808792114258, | |
| "learning_rate": 1.4802494802494804e-05, | |
| "loss": 1.3482, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2474012474012475, | |
| "grad_norm": 52.52505111694336, | |
| "learning_rate": 1.3762993762993763e-05, | |
| "loss": 1.0797, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4553014553014554, | |
| "grad_norm": 35.81852722167969, | |
| "learning_rate": 1.2723492723492725e-05, | |
| "loss": 1.0232, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6632016632016633, | |
| "grad_norm": 22.060670852661133, | |
| "learning_rate": 1.1683991683991684e-05, | |
| "loss": 1.0474, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6632016632016633, | |
| "eval_loss": 1.0084456205368042, | |
| "eval_runtime": 75.5278, | |
| "eval_samples_per_second": 18.841, | |
| "eval_steps_per_second": 4.713, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.871101871101871, | |
| "grad_norm": 28.490312576293945, | |
| "learning_rate": 1.0644490644490646e-05, | |
| "loss": 1.0069, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.079002079002079, | |
| "grad_norm": 31.279296875, | |
| "learning_rate": 9.604989604989607e-06, | |
| "loss": 0.8868, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.286902286902287, | |
| "grad_norm": 36.98978805541992, | |
| "learning_rate": 8.565488565488567e-06, | |
| "loss": 0.6839, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.494802494802495, | |
| "grad_norm": 38.62571716308594, | |
| "learning_rate": 7.525987525987527e-06, | |
| "loss": 0.6722, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.494802494802495, | |
| "eval_loss": 1.0827797651290894, | |
| "eval_runtime": 75.4831, | |
| "eval_samples_per_second": 18.852, | |
| "eval_steps_per_second": 4.716, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 15.777339935302734, | |
| "learning_rate": 6.486486486486487e-06, | |
| "loss": 0.6806, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.9106029106029108, | |
| "grad_norm": 21.232311248779297, | |
| "learning_rate": 5.446985446985448e-06, | |
| "loss": 0.6574, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.1185031185031185, | |
| "grad_norm": 31.68172264099121, | |
| "learning_rate": 4.4074844074844084e-06, | |
| "loss": 0.5586, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.3264033264033266, | |
| "grad_norm": 30.642192840576172, | |
| "learning_rate": 3.3679833679833685e-06, | |
| "loss": 0.4473, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3264033264033266, | |
| "eval_loss": 1.2475123405456543, | |
| "eval_runtime": 75.4548, | |
| "eval_samples_per_second": 18.859, | |
| "eval_steps_per_second": 4.718, | |
| "step": 8000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9620, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.793990720650445e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |