| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0021253985122210413, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.084661707403471e-05, | |
| "grad_norm": 11.687026023864746, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 8.409, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 7.084661707403471e-05, | |
| "eval_loss": 7.959962844848633, | |
| "eval_runtime": 853.269, | |
| "eval_samples_per_second": 6.966, | |
| "eval_steps_per_second": 1.742, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00014169323414806942, | |
| "grad_norm": 14.769333839416504, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 7.4806, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00021253985122210415, | |
| "grad_norm": 16.0882625579834, | |
| "learning_rate": 6e-06, | |
| "loss": 8.2333, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00028338646829613885, | |
| "grad_norm": 11.657814979553223, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 7.5361, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00028338646829613885, | |
| "eval_loss": 7.9468793869018555, | |
| "eval_runtime": 854.4406, | |
| "eval_samples_per_second": 6.957, | |
| "eval_steps_per_second": 1.739, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00035423308537017357, | |
| "grad_norm": 10.549196243286133, | |
| "learning_rate": 1e-05, | |
| "loss": 8.4016, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0004250797024442083, | |
| "grad_norm": 10.008234024047852, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 7.4369, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.000495926319518243, | |
| "grad_norm": 10.50498104095459, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 6.433, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0005667729365922777, | |
| "grad_norm": 10.711443901062012, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 7.2974, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0005667729365922777, | |
| "eval_loss": 7.829478740692139, | |
| "eval_runtime": 854.4135, | |
| "eval_samples_per_second": 6.957, | |
| "eval_steps_per_second": 1.739, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0006376195536663124, | |
| "grad_norm": 17.665964126586914, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 7.4679, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0007084661707403471, | |
| "grad_norm": 17.61928939819336, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 7.2824, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0007793127878143819, | |
| "grad_norm": 15.639145851135254, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 8.937, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0008501594048884166, | |
| "grad_norm": 21.38347625732422, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 7.4862, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008501594048884166, | |
| "eval_loss": 7.602104187011719, | |
| "eval_runtime": 854.3957, | |
| "eval_samples_per_second": 6.957, | |
| "eval_steps_per_second": 1.739, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0009210060219624513, | |
| "grad_norm": 17.321067810058594, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 5.8041, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.000991852639036486, | |
| "grad_norm": 13.502540588378906, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 7.3902, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0010626992561105207, | |
| "grad_norm": 12.128912925720215, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 7.7259, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0011335458731845554, | |
| "grad_norm": 10.935140609741211, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 7.9965, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0011335458731845554, | |
| "eval_loss": 7.300816535949707, | |
| "eval_runtime": 854.0987, | |
| "eval_samples_per_second": 6.959, | |
| "eval_steps_per_second": 1.74, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0012043924902585901, | |
| "grad_norm": 16.006664276123047, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 6.2996, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0012752391073326248, | |
| "grad_norm": 14.367422103881836, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 7.2675, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0013460857244066596, | |
| "grad_norm": 12.350759506225586, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 6.3202, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0014169323414806943, | |
| "grad_norm": 14.988066673278809, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 7.5826, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0014169323414806943, | |
| "eval_loss": 7.063891887664795, | |
| "eval_runtime": 853.8676, | |
| "eval_samples_per_second": 6.961, | |
| "eval_steps_per_second": 1.74, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.001487778958554729, | |
| "grad_norm": 17.12288475036621, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 6.8697, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0015586255756287637, | |
| "grad_norm": 12.52187442779541, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 5.9574, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0016294721927027985, | |
| "grad_norm": 14.923164367675781, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 8.2413, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0017003188097768332, | |
| "grad_norm": 13.699613571166992, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 7.6529, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0017003188097768332, | |
| "eval_loss": 6.94155216217041, | |
| "eval_runtime": 854.7337, | |
| "eval_samples_per_second": 6.954, | |
| "eval_steps_per_second": 1.739, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001771165426850868, | |
| "grad_norm": 15.484549522399902, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 5.9348, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0018420120439249026, | |
| "grad_norm": 15.122868537902832, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 6.1889, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0019128586609989374, | |
| "grad_norm": 17.317827224731445, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 7.4455, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.001983705278072972, | |
| "grad_norm": 14.604246139526367, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 5.7908, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.001983705278072972, | |
| "eval_loss": 6.888062000274658, | |
| "eval_runtime": 853.8107, | |
| "eval_samples_per_second": 6.962, | |
| "eval_steps_per_second": 1.74, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.002054551895147007, | |
| "grad_norm": 14.61190414428711, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 6.7001, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0021253985122210413, | |
| "grad_norm": 12.80364990234375, | |
| "learning_rate": 0.0, | |
| "loss": 6.8942, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4466802029428736e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |