| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1697072549851506, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016970725498515061, | |
| "eval_loss": 0.7674897909164429, | |
| "eval_runtime": 71.4945, | |
| "eval_samples_per_second": 13.889, | |
| "eval_steps_per_second": 1.748, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005091217649554518, | |
| "grad_norm": 0.7084022164344788, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.7782, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010182435299109036, | |
| "grad_norm": 0.799297034740448, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7424, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.015273652948663556, | |
| "grad_norm": 0.63190096616745, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6727, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.015273652948663556, | |
| "eval_loss": 0.6583356261253357, | |
| "eval_runtime": 72.6186, | |
| "eval_samples_per_second": 13.674, | |
| "eval_steps_per_second": 1.721, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.020364870598218072, | |
| "grad_norm": 0.6133125424385071, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.6068, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.025456088247772592, | |
| "grad_norm": 0.6916378736495972, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.5521, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.030547305897327112, | |
| "grad_norm": 0.5799879431724548, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.5013, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.030547305897327112, | |
| "eval_loss": 0.5111116170883179, | |
| "eval_runtime": 72.6202, | |
| "eval_samples_per_second": 13.674, | |
| "eval_steps_per_second": 1.721, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03563852354688163, | |
| "grad_norm": 0.5423109531402588, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.4864, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.040729741196436145, | |
| "grad_norm": 0.5172947645187378, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.4619, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04582095884599067, | |
| "grad_norm": 0.48413389921188354, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.4352, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04582095884599067, | |
| "eval_loss": 0.4471208155155182, | |
| "eval_runtime": 72.6272, | |
| "eval_samples_per_second": 13.673, | |
| "eval_steps_per_second": 1.721, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.050912176495545185, | |
| "grad_norm": 0.4537797272205353, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.4472, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0560033941450997, | |
| "grad_norm": 0.5130160450935364, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.432, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.061094611794654224, | |
| "grad_norm": 0.42606261372566223, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.4192, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.061094611794654224, | |
| "eval_loss": 0.4150956869125366, | |
| "eval_runtime": 72.6093, | |
| "eval_samples_per_second": 13.676, | |
| "eval_steps_per_second": 1.722, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06618582944420874, | |
| "grad_norm": 0.4711809754371643, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.4205, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07127704709376326, | |
| "grad_norm": 0.44911667704582214, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.3916, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07636826474331777, | |
| "grad_norm": 0.7759262323379517, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.3813, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07636826474331777, | |
| "eval_loss": 0.39792174100875854, | |
| "eval_runtime": 72.5601, | |
| "eval_samples_per_second": 13.685, | |
| "eval_steps_per_second": 1.723, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08145948239287229, | |
| "grad_norm": 0.4920569658279419, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.3736, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08655070004242682, | |
| "grad_norm": 0.4970743954181671, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.3831, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.09164191769198134, | |
| "grad_norm": 0.46638166904449463, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.3872, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.09164191769198134, | |
| "eval_loss": 0.3839256465435028, | |
| "eval_runtime": 72.6322, | |
| "eval_samples_per_second": 13.672, | |
| "eval_steps_per_second": 1.721, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.09673313534153585, | |
| "grad_norm": 0.5236009359359741, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.3653, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.10182435299109037, | |
| "grad_norm": 0.49318793416023254, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.3813, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.10691557064064489, | |
| "grad_norm": 0.542844831943512, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.3749, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.10691557064064489, | |
| "eval_loss": 0.3750540316104889, | |
| "eval_runtime": 72.6222, | |
| "eval_samples_per_second": 13.674, | |
| "eval_steps_per_second": 1.721, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.1120067882901994, | |
| "grad_norm": 0.5431241393089294, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.3732, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.11709800593975392, | |
| "grad_norm": 0.5867822766304016, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.3811, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.12218922358930845, | |
| "grad_norm": 0.5038670897483826, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.3653, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.12218922358930845, | |
| "eval_loss": 0.36931174993515015, | |
| "eval_runtime": 72.6161, | |
| "eval_samples_per_second": 13.675, | |
| "eval_steps_per_second": 1.721, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.12728044123886295, | |
| "grad_norm": 0.5324311852455139, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.3728, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.13237165888841748, | |
| "grad_norm": 0.47567200660705566, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.3449, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.137462876537972, | |
| "grad_norm": 0.46117860078811646, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.3702, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.137462876537972, | |
| "eval_loss": 0.3658164441585541, | |
| "eval_runtime": 72.5356, | |
| "eval_samples_per_second": 13.69, | |
| "eval_steps_per_second": 1.723, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.14255409418752651, | |
| "grad_norm": 0.4353705644607544, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.3578, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.14764531183708104, | |
| "grad_norm": 0.5827435851097107, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.3565, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.15273652948663555, | |
| "grad_norm": 0.45924094319343567, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.3456, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.15273652948663555, | |
| "eval_loss": 0.3645370602607727, | |
| "eval_runtime": 72.5734, | |
| "eval_samples_per_second": 13.683, | |
| "eval_steps_per_second": 1.722, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.15782774713619008, | |
| "grad_norm": 0.5294960141181946, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.3643, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.16291896478574458, | |
| "grad_norm": 0.4545383155345917, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.3463, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.1680101824352991, | |
| "grad_norm": 0.6580240726470947, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.3619, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.1680101824352991, | |
| "eval_loss": 0.364297091960907, | |
| "eval_runtime": 72.5705, | |
| "eval_samples_per_second": 13.683, | |
| "eval_steps_per_second": 1.722, | |
| "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": 1.4874840035386982e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |