| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03409187761015938, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0003409187761015938, | |
| "eval_loss": 11.089664459228516, | |
| "eval_runtime": 10.5672, | |
| "eval_samples_per_second": 467.578, | |
| "eval_steps_per_second": 58.483, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010227563283047815, | |
| "grad_norm": 0.45604798197746277, | |
| "learning_rate": 1.5e-05, | |
| "loss": 44.3568, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002045512656609563, | |
| "grad_norm": 0.46130335330963135, | |
| "learning_rate": 3e-05, | |
| "loss": 44.3545, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003068268984914344, | |
| "grad_norm": 0.44202330708503723, | |
| "learning_rate": 4.5e-05, | |
| "loss": 44.3516, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003068268984914344, | |
| "eval_loss": 11.088760375976562, | |
| "eval_runtime": 10.1289, | |
| "eval_samples_per_second": 487.81, | |
| "eval_steps_per_second": 61.013, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004091025313219126, | |
| "grad_norm": 0.4335339367389679, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 44.3549, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.005113781641523907, | |
| "grad_norm": 0.4942709803581238, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 44.3544, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006136537969828688, | |
| "grad_norm": 0.4659290015697479, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 44.3528, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006136537969828688, | |
| "eval_loss": 11.08657455444336, | |
| "eval_runtime": 10.2663, | |
| "eval_samples_per_second": 481.283, | |
| "eval_steps_per_second": 60.197, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.007159294298133469, | |
| "grad_norm": 0.4347454905509949, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 44.338, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.008182050626438252, | |
| "grad_norm": 0.4440023601055145, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 44.3387, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.009204806954743032, | |
| "grad_norm": 0.4840239882469177, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 44.341, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.009204806954743032, | |
| "eval_loss": 11.084238052368164, | |
| "eval_runtime": 10.1848, | |
| "eval_samples_per_second": 485.137, | |
| "eval_steps_per_second": 60.679, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.010227563283047815, | |
| "grad_norm": 0.49456870555877686, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 44.3389, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.011250319611352595, | |
| "grad_norm": 0.4676015377044678, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 44.3346, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.012273075939657376, | |
| "grad_norm": 0.4602993428707123, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 44.3378, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.012273075939657376, | |
| "eval_loss": 11.082551956176758, | |
| "eval_runtime": 10.2119, | |
| "eval_samples_per_second": 483.848, | |
| "eval_steps_per_second": 60.518, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.013295832267962158, | |
| "grad_norm": 0.49384036660194397, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 44.3236, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.014318588596266939, | |
| "grad_norm": 0.47107622027397156, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 44.3268, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.015341344924571721, | |
| "grad_norm": 0.49026861786842346, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 44.3271, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.015341344924571721, | |
| "eval_loss": 11.080031394958496, | |
| "eval_runtime": 10.1875, | |
| "eval_samples_per_second": 485.006, | |
| "eval_steps_per_second": 60.663, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.016364101252876503, | |
| "grad_norm": 0.4897162914276123, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 44.3183, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.017386857581181282, | |
| "grad_norm": 0.45754456520080566, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 44.3153, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.018409613909486065, | |
| "grad_norm": 0.5274067521095276, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 44.3143, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.018409613909486065, | |
| "eval_loss": 11.078282356262207, | |
| "eval_runtime": 10.1669, | |
| "eval_samples_per_second": 485.988, | |
| "eval_steps_per_second": 60.785, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.019432370237790847, | |
| "grad_norm": 0.4819900393486023, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 44.3073, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.02045512656609563, | |
| "grad_norm": 0.5234776735305786, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 44.3053, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.021477882894400408, | |
| "grad_norm": 0.5196526646614075, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 44.3021, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.021477882894400408, | |
| "eval_loss": 11.077031135559082, | |
| "eval_runtime": 10.1496, | |
| "eval_samples_per_second": 486.817, | |
| "eval_steps_per_second": 60.889, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.02250063922270519, | |
| "grad_norm": 0.5121042728424072, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 44.3014, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.023523395551009973, | |
| "grad_norm": 0.47857794165611267, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 44.3116, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.02454615187931475, | |
| "grad_norm": 0.5120978951454163, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 44.3001, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02454615187931475, | |
| "eval_loss": 11.07617473602295, | |
| "eval_runtime": 10.1688, | |
| "eval_samples_per_second": 485.897, | |
| "eval_steps_per_second": 60.774, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.025568908207619534, | |
| "grad_norm": 0.48292016983032227, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 44.2999, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.026591664535924316, | |
| "grad_norm": 0.46209338307380676, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 44.301, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0276144208642291, | |
| "grad_norm": 0.4896546006202698, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 44.3038, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0276144208642291, | |
| "eval_loss": 11.075565338134766, | |
| "eval_runtime": 10.139, | |
| "eval_samples_per_second": 487.324, | |
| "eval_steps_per_second": 60.952, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.028637177192533878, | |
| "grad_norm": 0.4694303274154663, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 44.3048, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02965993352083866, | |
| "grad_norm": 0.48952430486679077, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 44.3033, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.030682689849143442, | |
| "grad_norm": 0.46993014216423035, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 44.3012, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.030682689849143442, | |
| "eval_loss": 11.075267791748047, | |
| "eval_runtime": 10.2328, | |
| "eval_samples_per_second": 482.858, | |
| "eval_steps_per_second": 60.394, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03170544617744822, | |
| "grad_norm": 0.49486178159713745, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 44.3013, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.03272820250575301, | |
| "grad_norm": 0.47963476181030273, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 44.3049, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.033750958834057786, | |
| "grad_norm": 0.4520646333694458, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 44.3065, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.033750958834057786, | |
| "eval_loss": 11.075203895568848, | |
| "eval_runtime": 10.2212, | |
| "eval_samples_per_second": 483.409, | |
| "eval_steps_per_second": 60.463, | |
| "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": 748683264000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |