| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1363636363636362, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011363636363636364, | |
| "eval_loss": 3.371885061264038, | |
| "eval_runtime": 16.7352, | |
| "eval_samples_per_second": 8.844, | |
| "eval_steps_per_second": 1.135, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03409090909090909, | |
| "grad_norm": 0.9526873826980591, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.4877, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.06818181818181818, | |
| "grad_norm": 0.9885739088058472, | |
| "learning_rate": 3e-05, | |
| "loss": 3.3668, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10227272727272728, | |
| "grad_norm": 1.137595772743225, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.5998, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10227272727272728, | |
| "eval_loss": 3.313127279281616, | |
| "eval_runtime": 17.0437, | |
| "eval_samples_per_second": 8.684, | |
| "eval_steps_per_second": 1.115, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13636363636363635, | |
| "grad_norm": 1.3410749435424805, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 3.3443, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.17045454545454544, | |
| "grad_norm": 1.3125852346420288, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 3.127, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.20454545454545456, | |
| "grad_norm": 1.3217312097549438, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 3.0135, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20454545454545456, | |
| "eval_loss": 2.8074119091033936, | |
| "eval_runtime": 17.0586, | |
| "eval_samples_per_second": 8.676, | |
| "eval_steps_per_second": 1.114, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.23863636363636365, | |
| "grad_norm": 1.6107302904129028, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 2.6472, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 1.6939775943756104, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.6063, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3068181818181818, | |
| "grad_norm": 1.569953203201294, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.5102, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3068181818181818, | |
| "eval_loss": 2.223207473754883, | |
| "eval_runtime": 17.0516, | |
| "eval_samples_per_second": 8.68, | |
| "eval_steps_per_second": 1.114, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3409090909090909, | |
| "grad_norm": 1.427184820175171, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 2.1669, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 1.5676186084747314, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.2602, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4090909090909091, | |
| "grad_norm": 1.4393879175186157, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 2.3522, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4090909090909091, | |
| "eval_loss": 2.0838232040405273, | |
| "eval_runtime": 17.0641, | |
| "eval_samples_per_second": 8.673, | |
| "eval_steps_per_second": 1.113, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4431818181818182, | |
| "grad_norm": 1.435742974281311, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.0674, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.4772727272727273, | |
| "grad_norm": 1.354801893234253, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.2818, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5113636363636364, | |
| "grad_norm": 1.3888118267059326, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.0912, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5113636363636364, | |
| "eval_loss": 2.0078811645507812, | |
| "eval_runtime": 17.0471, | |
| "eval_samples_per_second": 8.682, | |
| "eval_steps_per_second": 1.115, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 1.4527229070663452, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.8334, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5795454545454546, | |
| "grad_norm": 1.287550449371338, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.0519, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6136363636363636, | |
| "grad_norm": 1.421475887298584, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.0602, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6136363636363636, | |
| "eval_loss": 1.96761155128479, | |
| "eval_runtime": 17.0648, | |
| "eval_samples_per_second": 8.673, | |
| "eval_steps_per_second": 1.113, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6477272727272727, | |
| "grad_norm": 1.5127720832824707, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.1334, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6818181818181818, | |
| "grad_norm": 1.5170916318893433, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.007, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7159090909090909, | |
| "grad_norm": 1.511754035949707, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.039, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7159090909090909, | |
| "eval_loss": 1.945967435836792, | |
| "eval_runtime": 17.0402, | |
| "eval_samples_per_second": 8.685, | |
| "eval_steps_per_second": 1.115, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.5611883401870728, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.1365, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7840909090909091, | |
| "grad_norm": 1.4377385377883911, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.111, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "grad_norm": 1.4909229278564453, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.0401, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "eval_loss": 1.9314323663711548, | |
| "eval_runtime": 17.0455, | |
| "eval_samples_per_second": 8.683, | |
| "eval_steps_per_second": 1.115, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8522727272727273, | |
| "grad_norm": 1.535409688949585, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.0856, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8863636363636364, | |
| "grad_norm": 1.479998230934143, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.9215, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.9204545454545454, | |
| "grad_norm": 1.4978396892547607, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.0011, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9204545454545454, | |
| "eval_loss": 1.9259672164916992, | |
| "eval_runtime": 17.0378, | |
| "eval_samples_per_second": 8.687, | |
| "eval_steps_per_second": 1.115, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9545454545454546, | |
| "grad_norm": 1.4255573749542236, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.1283, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.9886363636363636, | |
| "grad_norm": 1.549778938293457, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.9712, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.0227272727272727, | |
| "grad_norm": 1.6622482538223267, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.9208, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0227272727272727, | |
| "eval_loss": 1.922031283378601, | |
| "eval_runtime": 17.0501, | |
| "eval_samples_per_second": 8.68, | |
| "eval_steps_per_second": 1.114, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0568181818181819, | |
| "grad_norm": 1.373210072517395, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.8497, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 1.4251350164413452, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.8534, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 1.5110843181610107, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.7715, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "eval_loss": 1.9205644130706787, | |
| "eval_runtime": 17.0633, | |
| "eval_samples_per_second": 8.674, | |
| "eval_steps_per_second": 1.114, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 2.379352952411259e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |