| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.029448575425163808, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00029448575425163807, | |
| "eval_loss": 2.0751140117645264, | |
| "eval_runtime": 395.9969, | |
| "eval_samples_per_second": 14.445, | |
| "eval_steps_per_second": 1.806, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008834572627549142, | |
| "grad_norm": 45.21540451049805, | |
| "learning_rate": 1.5e-05, | |
| "loss": 8.1698, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0017669145255098284, | |
| "grad_norm": 29.80286407470703, | |
| "learning_rate": 3e-05, | |
| "loss": 7.8228, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002650371788264743, | |
| "grad_norm": 14.917160034179688, | |
| "learning_rate": 4.5e-05, | |
| "loss": 5.8253, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002650371788264743, | |
| "eval_loss": 1.2880661487579346, | |
| "eval_runtime": 398.9272, | |
| "eval_samples_per_second": 14.338, | |
| "eval_steps_per_second": 1.792, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003533829051019657, | |
| "grad_norm": 12.576233863830566, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 4.4652, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004417286313774571, | |
| "grad_norm": 12.087996482849121, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 4.1895, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005300743576529486, | |
| "grad_norm": 9.449481964111328, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 3.6443, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005300743576529486, | |
| "eval_loss": 0.9807960987091064, | |
| "eval_runtime": 398.492, | |
| "eval_samples_per_second": 14.354, | |
| "eval_steps_per_second": 1.794, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0061842008392844, | |
| "grad_norm": 8.682868003845215, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 3.8094, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.007067658102039314, | |
| "grad_norm": 7.913160800933838, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 3.8054, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007951115364794228, | |
| "grad_norm": 9.07978343963623, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.5453, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007951115364794228, | |
| "eval_loss": 0.9140304327011108, | |
| "eval_runtime": 398.4209, | |
| "eval_samples_per_second": 14.357, | |
| "eval_steps_per_second": 1.795, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008834572627549142, | |
| "grad_norm": 8.685811042785645, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 3.5354, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009718029890304056, | |
| "grad_norm": 7.87326192855835, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 3.519, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.010601487153058971, | |
| "grad_norm": 8.163304328918457, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 3.73, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010601487153058971, | |
| "eval_loss": 0.8961595892906189, | |
| "eval_runtime": 398.4206, | |
| "eval_samples_per_second": 14.357, | |
| "eval_steps_per_second": 1.795, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.011484944415813885, | |
| "grad_norm": 8.32168960571289, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 3.6521, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0123684016785688, | |
| "grad_norm": 5.999353885650635, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 3.7375, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.013251858941323713, | |
| "grad_norm": 6.515491485595703, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 3.1763, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013251858941323713, | |
| "eval_loss": 0.886064887046814, | |
| "eval_runtime": 398.3365, | |
| "eval_samples_per_second": 14.36, | |
| "eval_steps_per_second": 1.795, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.014135316204078627, | |
| "grad_norm": 7.323680400848389, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 3.6858, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.015018773466833541, | |
| "grad_norm": 8.582962036132812, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 3.484, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.015902230729588455, | |
| "grad_norm": 7.271780490875244, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 3.4188, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.015902230729588455, | |
| "eval_loss": 0.8784601092338562, | |
| "eval_runtime": 398.3464, | |
| "eval_samples_per_second": 14.359, | |
| "eval_steps_per_second": 1.795, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01678568799234337, | |
| "grad_norm": 7.015294075012207, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 3.6007, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.017669145255098283, | |
| "grad_norm": 8.451822280883789, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 3.2948, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0185526025178532, | |
| "grad_norm": 7.4288129806518555, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 3.4584, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0185526025178532, | |
| "eval_loss": 0.872480034828186, | |
| "eval_runtime": 398.1418, | |
| "eval_samples_per_second": 14.367, | |
| "eval_steps_per_second": 1.796, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01943605978060811, | |
| "grad_norm": 6.286691188812256, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 3.5261, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.020319517043363027, | |
| "grad_norm": 7.259737014770508, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 3.9765, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.021202974306117943, | |
| "grad_norm": 7.015435695648193, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 3.3159, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.021202974306117943, | |
| "eval_loss": 0.8712356090545654, | |
| "eval_runtime": 398.3235, | |
| "eval_samples_per_second": 14.36, | |
| "eval_steps_per_second": 1.795, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.022086431568872855, | |
| "grad_norm": 6.05751895904541, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 3.464, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02296988883162777, | |
| "grad_norm": 6.652122974395752, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 3.3039, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.023853346094382683, | |
| "grad_norm": 6.0785136222839355, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 3.3843, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.023853346094382683, | |
| "eval_loss": 0.8678173422813416, | |
| "eval_runtime": 398.2552, | |
| "eval_samples_per_second": 14.363, | |
| "eval_steps_per_second": 1.795, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0247368033571376, | |
| "grad_norm": 6.711858749389648, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 3.6375, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02562026061989251, | |
| "grad_norm": 6.864299297332764, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 3.1301, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.026503717882647427, | |
| "grad_norm": 7.26350736618042, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 3.1821, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.026503717882647427, | |
| "eval_loss": 0.8663612008094788, | |
| "eval_runtime": 398.3373, | |
| "eval_samples_per_second": 14.36, | |
| "eval_steps_per_second": 1.795, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.027387175145402343, | |
| "grad_norm": 6.898801326751709, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 3.1039, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.028270632408157255, | |
| "grad_norm": 6.446213722229004, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 3.1602, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.02915408967091217, | |
| "grad_norm": 6.621945381164551, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 3.3198, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02915408967091217, | |
| "eval_loss": 0.8659896850585938, | |
| "eval_runtime": 398.2351, | |
| "eval_samples_per_second": 14.363, | |
| "eval_steps_per_second": 1.795, | |
| "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": 1.406877472653312e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |