| { | |
| "best_metric": 0.8177282214164734, | |
| "best_model_checkpoint": "saves/BLOOM-7B/lora/train_1/checkpoint-140", | |
| "epoch": 0.20782061799289034, | |
| "eval_steps": 10, | |
| "global_step": 190, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010937927262783703, | |
| "grad_norm": 0.988913357257843, | |
| "learning_rate": 0.00029999015487222375, | |
| "loss": 1.7338, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.010937927262783703, | |
| "eval_loss": 0.9125174283981323, | |
| "eval_runtime": 2020.5977, | |
| "eval_samples_per_second": 23.32, | |
| "eval_steps_per_second": 0.729, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "grad_norm": 0.40076708793640137, | |
| "learning_rate": 0.00029996062078124905, | |
| "loss": 0.6341, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "eval_loss": 0.8568164706230164, | |
| "eval_runtime": 2020.8868, | |
| "eval_samples_per_second": 23.316, | |
| "eval_steps_per_second": 0.729, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "grad_norm": 0.2821263074874878, | |
| "learning_rate": 0.0002999114016039678, | |
| "loss": 0.5526, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "eval_loss": 0.899090051651001, | |
| "eval_runtime": 2020.647, | |
| "eval_samples_per_second": 23.319, | |
| "eval_steps_per_second": 0.729, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "grad_norm": 0.34677910804748535, | |
| "learning_rate": 0.00029984250380130117, | |
| "loss": 0.5651, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "eval_loss": 0.894440770149231, | |
| "eval_runtime": 2020.4948, | |
| "eval_samples_per_second": 23.321, | |
| "eval_steps_per_second": 0.729, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "grad_norm": 0.36296069622039795, | |
| "learning_rate": 0.0002997539364173515, | |
| "loss": 0.5392, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "eval_loss": 0.8795871138572693, | |
| "eval_runtime": 2020.3744, | |
| "eval_samples_per_second": 23.322, | |
| "eval_steps_per_second": 0.729, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "grad_norm": 0.3470546305179596, | |
| "learning_rate": 0.00029964571107821494, | |
| "loss": 0.5038, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "eval_loss": 0.8612168431282043, | |
| "eval_runtime": 2020.1194, | |
| "eval_samples_per_second": 23.325, | |
| "eval_steps_per_second": 0.729, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "grad_norm": 0.4996536374092102, | |
| "learning_rate": 0.00029951784199045534, | |
| "loss": 0.4904, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "eval_loss": 0.8335446715354919, | |
| "eval_runtime": 2020.0047, | |
| "eval_samples_per_second": 23.327, | |
| "eval_steps_per_second": 0.729, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "grad_norm": 0.4522307515144348, | |
| "learning_rate": 0.0002993703459392396, | |
| "loss": 0.476, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "eval_loss": 0.8787203431129456, | |
| "eval_runtime": 2021.176, | |
| "eval_samples_per_second": 23.313, | |
| "eval_steps_per_second": 0.729, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "grad_norm": 0.4299957752227783, | |
| "learning_rate": 0.00029920324228613376, | |
| "loss": 0.4819, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "eval_loss": 0.8441948294639587, | |
| "eval_runtime": 2020.7328, | |
| "eval_samples_per_second": 23.318, | |
| "eval_steps_per_second": 0.729, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "grad_norm": 0.49023598432540894, | |
| "learning_rate": 0.0002990165529665622, | |
| "loss": 0.4376, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "eval_loss": 0.8534411787986755, | |
| "eval_runtime": 2019.4689, | |
| "eval_samples_per_second": 23.333, | |
| "eval_steps_per_second": 0.729, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "grad_norm": 0.4622706174850464, | |
| "learning_rate": 0.0002988103024869277, | |
| "loss": 0.4443, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "eval_loss": 0.8330591917037964, | |
| "eval_runtime": 2020.301, | |
| "eval_samples_per_second": 23.323, | |
| "eval_steps_per_second": 0.729, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "grad_norm": 0.4706326425075531, | |
| "learning_rate": 0.00029858451792139453, | |
| "loss": 0.44, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "eval_loss": 0.8530427813529968, | |
| "eval_runtime": 2020.3749, | |
| "eval_samples_per_second": 23.322, | |
| "eval_steps_per_second": 0.729, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "grad_norm": 0.3258361220359802, | |
| "learning_rate": 0.0002983392289083346, | |
| "loss": 0.4362, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "eval_loss": 0.8594123125076294, | |
| "eval_runtime": 2019.0126, | |
| "eval_samples_per_second": 23.338, | |
| "eval_steps_per_second": 0.73, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "grad_norm": 0.2880217134952545, | |
| "learning_rate": 0.0002980744676464371, | |
| "loss": 0.4273, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "eval_loss": 0.8177282214164734, | |
| "eval_runtime": 2019.0397, | |
| "eval_samples_per_second": 23.338, | |
| "eval_steps_per_second": 0.73, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "grad_norm": 0.357930451631546, | |
| "learning_rate": 0.0002977902688904813, | |
| "loss": 0.438, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "eval_loss": 0.8449589610099792, | |
| "eval_runtime": 2020.3435, | |
| "eval_samples_per_second": 23.323, | |
| "eval_steps_per_second": 0.729, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "grad_norm": 0.3165152072906494, | |
| "learning_rate": 0.00029748666994677467, | |
| "loss": 0.4234, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "eval_loss": 0.8483649492263794, | |
| "eval_runtime": 2020.2184, | |
| "eval_samples_per_second": 23.324, | |
| "eval_steps_per_second": 0.729, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18594476346732294, | |
| "grad_norm": 0.31138718128204346, | |
| "learning_rate": 0.00029716371066825593, | |
| "loss": 0.4254, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18594476346732294, | |
| "eval_loss": 0.826312243938446, | |
| "eval_runtime": 2020.0047, | |
| "eval_samples_per_second": 23.327, | |
| "eval_steps_per_second": 0.729, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.19688269073010664, | |
| "grad_norm": 0.36661282181739807, | |
| "learning_rate": 0.0002968214334492632, | |
| "loss": 0.4074, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19688269073010664, | |
| "eval_loss": 0.8609428405761719, | |
| "eval_runtime": 2020.6157, | |
| "eval_samples_per_second": 23.32, | |
| "eval_steps_per_second": 0.729, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "grad_norm": 0.44822070002555847, | |
| "learning_rate": 0.00029645988321996917, | |
| "loss": 0.4209, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "eval_loss": 0.8370873332023621, | |
| "eval_runtime": 2020.0421, | |
| "eval_samples_per_second": 23.326, | |
| "eval_steps_per_second": 0.729, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "step": 190, | |
| "total_flos": 2.7046982162212454e+17, | |
| "train_loss": 0.5409229328757839, | |
| "train_runtime": 46907.2892, | |
| "train_samples_per_second": 14.968, | |
| "train_steps_per_second": 0.058 | |
| } | |
| ], | |
| "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": 5, | |
| "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": 2.7046982162212454e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |