| { | |
| "best_metric": 0.8665329813957214, | |
| "best_model_checkpoint": "./lora-alpaca/checkpoint-200", | |
| "epoch": 0.5119590432765379, | |
| "eval_steps": 200, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025597952163826893, | |
| "grad_norm": 0.1311839073896408, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2034, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.051195904327653786, | |
| "grad_norm": 0.22223667800426483, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4009, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07679385649148068, | |
| "grad_norm": 0.47330865263938904, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6715, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10239180865530757, | |
| "grad_norm": 0.9922175407409668, | |
| "learning_rate": 4e-05, | |
| "loss": 1.9561, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12798976081913446, | |
| "grad_norm": 2.9832212924957275, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8355, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15358771298296137, | |
| "grad_norm": 0.2527308464050293, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9926, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.17918566514678827, | |
| "grad_norm": 0.37248659133911133, | |
| "learning_rate": 7e-05, | |
| "loss": 1.0282, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.20478361731061515, | |
| "grad_norm": 0.29374104738235474, | |
| "learning_rate": 8e-05, | |
| "loss": 0.983, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23038156947444205, | |
| "grad_norm": 0.2772407829761505, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8514, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2559795216382689, | |
| "grad_norm": 0.44678089022636414, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6994, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28157747380209586, | |
| "grad_norm": 0.1145705059170723, | |
| "learning_rate": 9.906542056074767e-05, | |
| "loss": 0.8862, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.30717542596592273, | |
| "grad_norm": 0.1596599817276001, | |
| "learning_rate": 9.813084112149533e-05, | |
| "loss": 0.9123, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3327733781297496, | |
| "grad_norm": 0.15576007962226868, | |
| "learning_rate": 9.7196261682243e-05, | |
| "loss": 0.8971, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.35837133029357654, | |
| "grad_norm": 0.17032374441623688, | |
| "learning_rate": 9.626168224299066e-05, | |
| "loss": 0.7929, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3839692824574034, | |
| "grad_norm": 0.505789577960968, | |
| "learning_rate": 9.532710280373832e-05, | |
| "loss": 0.6719, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4095672346212303, | |
| "grad_norm": 0.11598572134971619, | |
| "learning_rate": 9.439252336448599e-05, | |
| "loss": 0.8772, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4351651867850572, | |
| "grad_norm": 0.17679552733898163, | |
| "learning_rate": 9.345794392523365e-05, | |
| "loss": 0.8909, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.4607631389488841, | |
| "grad_norm": 0.177357017993927, | |
| "learning_rate": 9.252336448598131e-05, | |
| "loss": 0.8708, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.486361091112711, | |
| "grad_norm": 0.14735093712806702, | |
| "learning_rate": 9.158878504672898e-05, | |
| "loss": 0.7538, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5119590432765379, | |
| "grad_norm": 0.36298009753227234, | |
| "learning_rate": 9.065420560747664e-05, | |
| "loss": 0.6309, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5119590432765379, | |
| "eval_loss": 0.8665329813957214, | |
| "eval_runtime": 147.8844, | |
| "eval_samples_per_second": 13.524, | |
| "eval_steps_per_second": 1.691, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1170, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.2872832292683776e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |