| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00898485210091112, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.98485210091112e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1398.131, | |
| "eval_samples_per_second": 26.814, | |
| "eval_steps_per_second": 13.407, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002695455630273336, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005390911260546673, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008086366890820008, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008086366890820008, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1400.6834, | |
| "eval_samples_per_second": 26.766, | |
| "eval_steps_per_second": 13.383, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0010781822521093345, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001347727815136668, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016172733781640017, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0016172733781640017, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1400.0163, | |
| "eval_samples_per_second": 26.778, | |
| "eval_steps_per_second": 13.389, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0018868189411913353, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002156364504218669, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0024259100672460025, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0024259100672460025, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.5792, | |
| "eval_samples_per_second": 26.825, | |
| "eval_steps_per_second": 13.412, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002695455630273336, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00296500119330067, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0032345467563280033, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0032345467563280033, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.6922, | |
| "eval_samples_per_second": 26.823, | |
| "eval_steps_per_second": 13.411, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0035040923193553368, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0037736378823826707, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004043183445410004, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004043183445410004, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1398.6647, | |
| "eval_samples_per_second": 26.804, | |
| "eval_steps_per_second": 13.402, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004312729008437338, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0045822745714646715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.004851820134492005, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.004851820134492005, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.5409, | |
| "eval_samples_per_second": 26.826, | |
| "eval_steps_per_second": 13.413, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005121365697519338, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005390911260546672, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005660456823574005, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005660456823574005, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.6931, | |
| "eval_samples_per_second": 26.823, | |
| "eval_steps_per_second": 13.411, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.00593000238660134, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.006199547949628673, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.006469093512656007, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006469093512656007, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.7111, | |
| "eval_samples_per_second": 26.822, | |
| "eval_steps_per_second": 13.411, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.00673863907568334, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0070081846387106735, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.007277730201738007, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007277730201738007, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1398.2001, | |
| "eval_samples_per_second": 26.813, | |
| "eval_steps_per_second": 13.407, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007547275764765341, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.007816821327792675, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.008086366890820007, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008086366890820007, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.6994, | |
| "eval_samples_per_second": 26.823, | |
| "eval_steps_per_second": 13.411, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008355912453847342, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.008625458016874676, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.008895003579902009, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.008895003579902009, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1397.3879, | |
| "eval_samples_per_second": 26.829, | |
| "eval_steps_per_second": 13.414, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0458975166267392e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |