| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06216006216006216, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006216006216006216, | |
| "eval_loss": 5.483861446380615, | |
| "eval_runtime": 267.3642, | |
| "eval_samples_per_second": 10.136, | |
| "eval_steps_per_second": 1.268, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0018648018648018648, | |
| "grad_norm": 2.267636775970459, | |
| "learning_rate": 1.5e-05, | |
| "loss": 5.2292, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0037296037296037296, | |
| "grad_norm": 2.6419992446899414, | |
| "learning_rate": 3e-05, | |
| "loss": 5.535, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005594405594405594, | |
| "grad_norm": 2.7405083179473877, | |
| "learning_rate": 4.5e-05, | |
| "loss": 5.2825, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005594405594405594, | |
| "eval_loss": 5.21874475479126, | |
| "eval_runtime": 269.8177, | |
| "eval_samples_per_second": 10.044, | |
| "eval_steps_per_second": 1.256, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007459207459207459, | |
| "grad_norm": 2.609468698501587, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 4.8802, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.009324009324009324, | |
| "grad_norm": 2.8223719596862793, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 4.5065, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011188811188811189, | |
| "grad_norm": 2.5072696208953857, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 3.8598, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011188811188811189, | |
| "eval_loss": 3.936025381088257, | |
| "eval_runtime": 269.871, | |
| "eval_samples_per_second": 10.042, | |
| "eval_steps_per_second": 1.256, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.013053613053613054, | |
| "grad_norm": 2.995809555053711, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 3.7813, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.014918414918414918, | |
| "grad_norm": 3.6315479278564453, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 3.5566, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.016783216783216783, | |
| "grad_norm": 3.4310660362243652, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.4073, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.016783216783216783, | |
| "eval_loss": 3.389742851257324, | |
| "eval_runtime": 269.8486, | |
| "eval_samples_per_second": 10.043, | |
| "eval_steps_per_second": 1.256, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.018648018648018648, | |
| "grad_norm": 2.20501971244812, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 3.2354, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.020512820512820513, | |
| "grad_norm": 2.6462857723236084, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 3.3163, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.022377622377622378, | |
| "grad_norm": 2.307749032974243, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 3.1726, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.022377622377622378, | |
| "eval_loss": 3.2075862884521484, | |
| "eval_runtime": 269.8635, | |
| "eval_samples_per_second": 10.042, | |
| "eval_steps_per_second": 1.256, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.024242424242424242, | |
| "grad_norm": 2.687164545059204, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 3.3259, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.026107226107226107, | |
| "grad_norm": 2.353116750717163, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 3.1402, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.027972027972027972, | |
| "grad_norm": 2.298196315765381, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.9699, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.027972027972027972, | |
| "eval_loss": 3.1050002574920654, | |
| "eval_runtime": 269.7032, | |
| "eval_samples_per_second": 10.048, | |
| "eval_steps_per_second": 1.257, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.029836829836829837, | |
| "grad_norm": 2.5003011226654053, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 3.0484, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.031701631701631705, | |
| "grad_norm": 2.3574583530426025, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 3.1361, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.033566433566433566, | |
| "grad_norm": 2.1045308113098145, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.9611, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.033566433566433566, | |
| "eval_loss": 3.046522617340088, | |
| "eval_runtime": 269.8245, | |
| "eval_samples_per_second": 10.044, | |
| "eval_steps_per_second": 1.256, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.035431235431235435, | |
| "grad_norm": 2.1651206016540527, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 3.0361, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.037296037296037296, | |
| "grad_norm": 2.734593629837036, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 3.0934, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.039160839160839164, | |
| "grad_norm": 2.2802178859710693, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 3.0565, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.039160839160839164, | |
| "eval_loss": 3.0123836994171143, | |
| "eval_runtime": 269.8579, | |
| "eval_samples_per_second": 10.042, | |
| "eval_steps_per_second": 1.256, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.041025641025641026, | |
| "grad_norm": 2.0753777027130127, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.8726, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.042890442890442894, | |
| "grad_norm": 2.542839527130127, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 3.093, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.044755244755244755, | |
| "grad_norm": 2.0449094772338867, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.9422, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.044755244755244755, | |
| "eval_loss": 2.9884579181671143, | |
| "eval_runtime": 269.9643, | |
| "eval_samples_per_second": 10.038, | |
| "eval_steps_per_second": 1.256, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.046620046620046623, | |
| "grad_norm": 2.5215494632720947, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 3.1116, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.048484848484848485, | |
| "grad_norm": 2.085618019104004, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.916, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.05034965034965035, | |
| "grad_norm": 2.152615547180176, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.9658, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.05034965034965035, | |
| "eval_loss": 2.9746768474578857, | |
| "eval_runtime": 269.9546, | |
| "eval_samples_per_second": 10.039, | |
| "eval_steps_per_second": 1.256, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.052214452214452214, | |
| "grad_norm": 2.266479015350342, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.9542, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.05407925407925408, | |
| "grad_norm": 2.6077001094818115, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.9807, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.055944055944055944, | |
| "grad_norm": 1.9626648426055908, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.956, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.055944055944055944, | |
| "eval_loss": 2.9686450958251953, | |
| "eval_runtime": 269.9369, | |
| "eval_samples_per_second": 10.039, | |
| "eval_steps_per_second": 1.256, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05780885780885781, | |
| "grad_norm": 2.821603298187256, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 3.0367, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.059673659673659674, | |
| "grad_norm": 2.2452235221862793, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.7214, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "grad_norm": 2.7175397872924805, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.8909, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "eval_loss": 2.9674060344696045, | |
| "eval_runtime": 269.5194, | |
| "eval_samples_per_second": 10.055, | |
| "eval_steps_per_second": 1.258, | |
| "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": 2.073574957551452e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |