| { | |
| "best_metric": 0.5069935321807861, | |
| "best_model_checkpoint": "saves/Llama2-7B/lora/train_1/checkpoint-110", | |
| "epoch": 0.22969647251845776, | |
| "eval_steps": 10, | |
| "global_step": 210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010937927262783703, | |
| "grad_norm": 0.5937469005584717, | |
| "learning_rate": 0.00029999015487222375, | |
| "loss": 1.2211, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.010937927262783703, | |
| "eval_loss": 0.550807535648346, | |
| "eval_runtime": 234.7461, | |
| "eval_samples_per_second": 20.073, | |
| "eval_steps_per_second": 0.63, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "grad_norm": 0.4648227393627167, | |
| "learning_rate": 0.00029996062078124905, | |
| "loss": 0.446, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "eval_loss": 0.5179979205131531, | |
| "eval_runtime": 234.6313, | |
| "eval_samples_per_second": 20.083, | |
| "eval_steps_per_second": 0.631, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "grad_norm": 0.42532142996788025, | |
| "learning_rate": 0.0002999114016039678, | |
| "loss": 0.3889, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "eval_loss": 0.5139310956001282, | |
| "eval_runtime": 234.5844, | |
| "eval_samples_per_second": 20.087, | |
| "eval_steps_per_second": 0.631, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "grad_norm": 0.47056564688682556, | |
| "learning_rate": 0.00029984250380130117, | |
| "loss": 0.3757, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "eval_loss": 0.575846791267395, | |
| "eval_runtime": 234.7308, | |
| "eval_samples_per_second": 20.074, | |
| "eval_steps_per_second": 0.631, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "grad_norm": 0.2359870821237564, | |
| "learning_rate": 0.0002997539364173515, | |
| "loss": 0.3551, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "eval_loss": 0.5461050271987915, | |
| "eval_runtime": 234.7378, | |
| "eval_samples_per_second": 20.073, | |
| "eval_steps_per_second": 0.63, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "grad_norm": 0.19186203181743622, | |
| "learning_rate": 0.00029964571107821494, | |
| "loss": 0.3351, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "eval_loss": 0.5407155752182007, | |
| "eval_runtime": 234.7344, | |
| "eval_samples_per_second": 20.074, | |
| "eval_steps_per_second": 0.63, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "grad_norm": 0.25036656856536865, | |
| "learning_rate": 0.00029951784199045534, | |
| "loss": 0.3269, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "eval_loss": 0.5335067510604858, | |
| "eval_runtime": 234.559, | |
| "eval_samples_per_second": 20.089, | |
| "eval_steps_per_second": 0.631, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "grad_norm": 0.3150342106819153, | |
| "learning_rate": 0.0002993703459392396, | |
| "loss": 0.3231, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "eval_loss": 0.5120783448219299, | |
| "eval_runtime": 234.641, | |
| "eval_samples_per_second": 20.082, | |
| "eval_steps_per_second": 0.631, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "grad_norm": 0.3419171869754791, | |
| "learning_rate": 0.00029920324228613376, | |
| "loss": 0.3324, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "eval_loss": 0.5196456909179688, | |
| "eval_runtime": 234.6621, | |
| "eval_samples_per_second": 20.08, | |
| "eval_steps_per_second": 0.631, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "grad_norm": 0.2197423279285431, | |
| "learning_rate": 0.0002990165529665622, | |
| "loss": 0.3039, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "eval_loss": 0.5251961350440979, | |
| "eval_runtime": 234.8661, | |
| "eval_samples_per_second": 20.062, | |
| "eval_steps_per_second": 0.63, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "grad_norm": 0.16334187984466553, | |
| "learning_rate": 0.0002988103024869277, | |
| "loss": 0.3083, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "eval_loss": 0.5069935321807861, | |
| "eval_runtime": 234.5526, | |
| "eval_samples_per_second": 20.089, | |
| "eval_steps_per_second": 0.631, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "grad_norm": 0.2504599392414093, | |
| "learning_rate": 0.00029858451792139453, | |
| "loss": 0.3034, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "eval_loss": 0.5375232100486755, | |
| "eval_runtime": 234.9066, | |
| "eval_samples_per_second": 20.059, | |
| "eval_steps_per_second": 0.63, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "grad_norm": 0.20455080270767212, | |
| "learning_rate": 0.0002983392289083346, | |
| "loss": 0.3061, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "eval_loss": 0.5345537066459656, | |
| "eval_runtime": 234.955, | |
| "eval_samples_per_second": 20.055, | |
| "eval_steps_per_second": 0.63, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "grad_norm": 0.20371787250041962, | |
| "learning_rate": 0.0002980744676464371, | |
| "loss": 0.3021, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "eval_loss": 0.5111611485481262, | |
| "eval_runtime": 234.7152, | |
| "eval_samples_per_second": 20.075, | |
| "eval_steps_per_second": 0.631, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "grad_norm": 0.14063598215579987, | |
| "learning_rate": 0.0002977902688904813, | |
| "loss": 0.3067, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "eval_loss": 0.5191352367401123, | |
| "eval_runtime": 234.5955, | |
| "eval_samples_per_second": 20.086, | |
| "eval_steps_per_second": 0.631, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "grad_norm": 0.20113405585289001, | |
| "learning_rate": 0.00029748666994677467, | |
| "loss": 0.2958, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "eval_loss": 0.5278245210647583, | |
| "eval_runtime": 234.5747, | |
| "eval_samples_per_second": 20.087, | |
| "eval_steps_per_second": 0.631, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18594476346732294, | |
| "grad_norm": 0.20178332924842834, | |
| "learning_rate": 0.00029716371066825593, | |
| "loss": 0.3002, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18594476346732294, | |
| "eval_loss": 0.5170465111732483, | |
| "eval_runtime": 234.6689, | |
| "eval_samples_per_second": 20.079, | |
| "eval_steps_per_second": 0.631, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.19688269073010664, | |
| "grad_norm": 0.19617249071598053, | |
| "learning_rate": 0.0002968214334492632, | |
| "loss": 0.2896, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19688269073010664, | |
| "eval_loss": 0.5085877180099487, | |
| "eval_runtime": 234.6318, | |
| "eval_samples_per_second": 20.083, | |
| "eval_steps_per_second": 0.631, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "grad_norm": 0.2610602080821991, | |
| "learning_rate": 0.00029645988321996917, | |
| "loss": 0.2989, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "eval_loss": 0.530036985874176, | |
| "eval_runtime": 234.599, | |
| "eval_samples_per_second": 20.085, | |
| "eval_steps_per_second": 0.631, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21875854525567404, | |
| "grad_norm": 0.1975240856409073, | |
| "learning_rate": 0.00029607910744048336, | |
| "loss": 0.3012, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21875854525567404, | |
| "eval_loss": 0.5137269496917725, | |
| "eval_runtime": 234.4763, | |
| "eval_samples_per_second": 20.096, | |
| "eval_steps_per_second": 0.631, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22969647251845776, | |
| "grad_norm": 0.19913341104984283, | |
| "learning_rate": 0.00029567915609462174, | |
| "loss": 0.2802, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22969647251845776, | |
| "eval_loss": 0.5159074068069458, | |
| "eval_runtime": 234.549, | |
| "eval_samples_per_second": 20.09, | |
| "eval_steps_per_second": 0.631, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22969647251845776, | |
| "step": 210, | |
| "total_flos": 3.9689172647569e+17, | |
| "train_loss": 0.36669377031780426, | |
| "train_runtime": 15000.4627, | |
| "train_samples_per_second": 46.807, | |
| "train_steps_per_second": 0.183 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2742, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 10, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.9689172647569e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |