| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05076679005817028, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016922263352723427, | |
| "grad_norm": 2.0210459232330322, | |
| "learning_rate": 1e-05, | |
| "loss": 2.3704, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016922263352723427, | |
| "eval_loss": 2.3479883670806885, | |
| "eval_runtime": 96.0898, | |
| "eval_samples_per_second": 10.365, | |
| "eval_steps_per_second": 5.183, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0033844526705446853, | |
| "grad_norm": 2.177330732345581, | |
| "learning_rate": 2e-05, | |
| "loss": 2.6717, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005076679005817028, | |
| "grad_norm": 1.6872704029083252, | |
| "learning_rate": 3e-05, | |
| "loss": 2.142, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006768905341089371, | |
| "grad_norm": 2.007863759994507, | |
| "learning_rate": 4e-05, | |
| "loss": 2.4436, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.008461131676361713, | |
| "grad_norm": 1.640100121498108, | |
| "learning_rate": 5e-05, | |
| "loss": 2.293, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010153358011634056, | |
| "grad_norm": 1.901017427444458, | |
| "learning_rate": 6e-05, | |
| "loss": 2.3465, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011845584346906399, | |
| "grad_norm": 1.1159406900405884, | |
| "learning_rate": 7e-05, | |
| "loss": 2.083, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.013537810682178741, | |
| "grad_norm": 0.87894606590271, | |
| "learning_rate": 8e-05, | |
| "loss": 2.0493, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013537810682178741, | |
| "eval_loss": 2.071341037750244, | |
| "eval_runtime": 71.6484, | |
| "eval_samples_per_second": 13.901, | |
| "eval_steps_per_second": 6.951, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.015230037017451084, | |
| "grad_norm": 0.5529594421386719, | |
| "learning_rate": 9e-05, | |
| "loss": 2.2644, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.016922263352723427, | |
| "grad_norm": 0.5951188802719116, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0647, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01861448968799577, | |
| "grad_norm": 0.522255003452301, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.9332, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.020306716023268112, | |
| "grad_norm": 0.6271169781684875, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.9483, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.021998942358540455, | |
| "grad_norm": 0.6509926915168762, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 2.0758, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.023691168693812797, | |
| "grad_norm": 0.5996648073196411, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.0503, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02538339502908514, | |
| "grad_norm": 0.5467623472213745, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.037, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.027075621364357483, | |
| "grad_norm": 0.5379562377929688, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.1476, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.027075621364357483, | |
| "eval_loss": 2.0144705772399902, | |
| "eval_runtime": 71.6861, | |
| "eval_samples_per_second": 13.894, | |
| "eval_steps_per_second": 6.947, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.028767847699629825, | |
| "grad_norm": 0.4585031569004059, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.9176, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.030460074034902168, | |
| "grad_norm": 0.378377228975296, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.7755, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03215230037017451, | |
| "grad_norm": 0.44114381074905396, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.9266, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.033844526705446853, | |
| "grad_norm": 0.4758003354072571, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9804, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.035536753040719196, | |
| "grad_norm": 0.3892809748649597, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 2.1765, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03722897937599154, | |
| "grad_norm": 0.4293234050273895, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.9866, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03892120571126388, | |
| "grad_norm": 0.44209152460098267, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.9257, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.040613432046536224, | |
| "grad_norm": 0.4333565831184387, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.7604, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.040613432046536224, | |
| "eval_loss": 1.993262529373169, | |
| "eval_runtime": 71.5545, | |
| "eval_samples_per_second": 13.919, | |
| "eval_steps_per_second": 6.96, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04230565838180857, | |
| "grad_norm": 0.36267876625061035, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.9153, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04399788471708091, | |
| "grad_norm": 0.40100884437561035, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.2008, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04569011105235325, | |
| "grad_norm": 0.3713359534740448, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.0199, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.047382337387625595, | |
| "grad_norm": 0.38803306221961975, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 2.1263, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04907456372289794, | |
| "grad_norm": 0.38876503705978394, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.9215, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.05076679005817028, | |
| "grad_norm": 0.39528387784957886, | |
| "learning_rate": 0.0, | |
| "loss": 2.0199, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.865643726136934e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |