| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.017681902572716825, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00017681902572716824, | |
| "eval_loss": 1.5049527883529663, | |
| "eval_runtime": 671.7063, | |
| "eval_samples_per_second": 14.18, | |
| "eval_steps_per_second": 1.773, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005304570771815047, | |
| "grad_norm": 17.419424057006836, | |
| "learning_rate": 3e-05, | |
| "loss": 6.0015, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010609141543630094, | |
| "grad_norm": 10.581345558166504, | |
| "learning_rate": 6e-05, | |
| "loss": 5.1541, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001591371231544514, | |
| "grad_norm": 7.764749526977539, | |
| "learning_rate": 9e-05, | |
| "loss": 4.2045, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001591371231544514, | |
| "eval_loss": 0.9556123614311218, | |
| "eval_runtime": 675.7088, | |
| "eval_samples_per_second": 14.096, | |
| "eval_steps_per_second": 1.763, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002121828308726019, | |
| "grad_norm": 7.4701151847839355, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.6782, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0026522853859075237, | |
| "grad_norm": 5.40271520614624, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.4578, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.003182742463089028, | |
| "grad_norm": 5.472065448760986, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.1924, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.003182742463089028, | |
| "eval_loss": 0.811770498752594, | |
| "eval_runtime": 675.7606, | |
| "eval_samples_per_second": 14.095, | |
| "eval_steps_per_second": 1.762, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.003713199540270533, | |
| "grad_norm": 4.88375186920166, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.1914, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004243656617452038, | |
| "grad_norm": 5.257078170776367, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.2435, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0047741136946335425, | |
| "grad_norm": 4.208374977111816, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.0918, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0047741136946335425, | |
| "eval_loss": 0.7676888108253479, | |
| "eval_runtime": 675.6513, | |
| "eval_samples_per_second": 14.098, | |
| "eval_steps_per_second": 1.763, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.005304570771815047, | |
| "grad_norm": 4.615727424621582, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.0164, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005835027848996552, | |
| "grad_norm": 4.36652946472168, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.023, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.006365484926178056, | |
| "grad_norm": 4.23211145401001, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.0718, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006365484926178056, | |
| "eval_loss": 0.7438393235206604, | |
| "eval_runtime": 675.8638, | |
| "eval_samples_per_second": 14.093, | |
| "eval_steps_per_second": 1.762, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006895942003359561, | |
| "grad_norm": 4.29207181930542, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.996, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.007426399080541066, | |
| "grad_norm": 4.204301834106445, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.9596, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.007956856157722571, | |
| "grad_norm": 3.9925191402435303, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.8552, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.007956856157722571, | |
| "eval_loss": 0.7274059057235718, | |
| "eval_runtime": 675.4704, | |
| "eval_samples_per_second": 14.101, | |
| "eval_steps_per_second": 1.763, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.008487313234904075, | |
| "grad_norm": 4.028677940368652, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.8459, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.009017770312085581, | |
| "grad_norm": 4.294224739074707, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.8475, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.009548227389267085, | |
| "grad_norm": 3.748793840408325, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.9537, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.009548227389267085, | |
| "eval_loss": 0.7180169224739075, | |
| "eval_runtime": 675.8048, | |
| "eval_samples_per_second": 14.094, | |
| "eval_steps_per_second": 1.762, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.010078684466448589, | |
| "grad_norm": 3.9666662216186523, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.7154, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.010609141543630095, | |
| "grad_norm": 4.252094268798828, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.8792, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.011139598620811599, | |
| "grad_norm": 3.9958548545837402, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.8798, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.011139598620811599, | |
| "eval_loss": 0.7039853930473328, | |
| "eval_runtime": 675.7018, | |
| "eval_samples_per_second": 14.096, | |
| "eval_steps_per_second": 1.763, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.011670055697993105, | |
| "grad_norm": 4.007390022277832, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.8367, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.012200512775174609, | |
| "grad_norm": 3.7783989906311035, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.7552, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.012730969852356113, | |
| "grad_norm": 4.115812301635742, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.573, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.012730969852356113, | |
| "eval_loss": 0.6983979940414429, | |
| "eval_runtime": 674.9598, | |
| "eval_samples_per_second": 14.112, | |
| "eval_steps_per_second": 1.765, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.013261426929537619, | |
| "grad_norm": 3.7902400493621826, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.8309, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.013791884006719123, | |
| "grad_norm": 3.6336324214935303, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 2.6799, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.014322341083900628, | |
| "grad_norm": 3.9960505962371826, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 2.7011, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.014322341083900628, | |
| "eval_loss": 0.6926208734512329, | |
| "eval_runtime": 677.1462, | |
| "eval_samples_per_second": 14.066, | |
| "eval_steps_per_second": 1.759, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.014852798161082132, | |
| "grad_norm": 3.490804672241211, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.6716, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.015383255238263636, | |
| "grad_norm": 3.8302175998687744, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 2.7132, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.015913712315445142, | |
| "grad_norm": 4.030756950378418, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 2.7684, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.015913712315445142, | |
| "eval_loss": 0.6902585625648499, | |
| "eval_runtime": 677.2291, | |
| "eval_samples_per_second": 14.065, | |
| "eval_steps_per_second": 1.759, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.016444169392626648, | |
| "grad_norm": 3.537766218185425, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.7386, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.01697462646980815, | |
| "grad_norm": 3.5246758460998535, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 2.8188, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.017505083546989656, | |
| "grad_norm": 3.6730434894561768, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.7731, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.017505083546989656, | |
| "eval_loss": 0.6894702911376953, | |
| "eval_runtime": 676.8252, | |
| "eval_samples_per_second": 14.073, | |
| "eval_steps_per_second": 1.76, | |
| "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": 1.406260607975424e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |