| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002591680704937152, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.183361409874304e-05, | |
| "grad_norm": 2.3113319873809814, | |
| "learning_rate": 1e-05, | |
| "loss": 3.5351, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.183361409874304e-05, | |
| "eval_loss": 3.5113468170166016, | |
| "eval_runtime": 1769.9186, | |
| "eval_samples_per_second": 9.18, | |
| "eval_steps_per_second": 1.148, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010366722819748608, | |
| "grad_norm": 2.229635238647461, | |
| "learning_rate": 2e-05, | |
| "loss": 3.4005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00015550084229622911, | |
| "grad_norm": 2.0467476844787598, | |
| "learning_rate": 3e-05, | |
| "loss": 3.5197, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00020733445639497215, | |
| "grad_norm": 2.512348175048828, | |
| "learning_rate": 4e-05, | |
| "loss": 3.621, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0002591680704937152, | |
| "grad_norm": 2.532371997833252, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5631, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00031100168459245823, | |
| "grad_norm": 2.249849319458008, | |
| "learning_rate": 6e-05, | |
| "loss": 3.404, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00036283529869120127, | |
| "grad_norm": 2.5401272773742676, | |
| "learning_rate": 7e-05, | |
| "loss": 3.4107, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0004146689127899443, | |
| "grad_norm": 2.23197603225708, | |
| "learning_rate": 8e-05, | |
| "loss": 3.0562, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0004665025268886873, | |
| "grad_norm": 2.439784526824951, | |
| "learning_rate": 9e-05, | |
| "loss": 3.0909, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0004665025268886873, | |
| "eval_loss": 2.900395393371582, | |
| "eval_runtime": 1774.4876, | |
| "eval_samples_per_second": 9.156, | |
| "eval_steps_per_second": 1.145, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005183361409874304, | |
| "grad_norm": 2.121995687484741, | |
| "learning_rate": 0.0001, | |
| "loss": 2.7377, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005701697550861734, | |
| "grad_norm": 2.543422222137451, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.8312, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0006220033691849165, | |
| "grad_norm": 3.130173444747925, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.8057, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006738369832836595, | |
| "grad_norm": 2.596888542175293, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.6877, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0007256705973824025, | |
| "grad_norm": 2.7695186138153076, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.8845, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007775042114811456, | |
| "grad_norm": 2.208770751953125, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.4707, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0008293378255798886, | |
| "grad_norm": 2.075831174850464, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.6753, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0008811714396786315, | |
| "grad_norm": 3.6710920333862305, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.5295, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0009330050537773746, | |
| "grad_norm": 1.9427752494812012, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.5586, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009330050537773746, | |
| "eval_loss": 2.5122954845428467, | |
| "eval_runtime": 1772.457, | |
| "eval_samples_per_second": 9.166, | |
| "eval_steps_per_second": 1.146, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009848386678761177, | |
| "grad_norm": 2.3610899448394775, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.5588, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0010366722819748608, | |
| "grad_norm": 1.9613486528396606, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.5903, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010885058960736038, | |
| "grad_norm": 1.8334481716156006, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.4418, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0011403395101723468, | |
| "grad_norm": 1.7611281871795654, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.5619, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0011921731242710899, | |
| "grad_norm": 2.285313606262207, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.4165, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.001244006738369833, | |
| "grad_norm": 1.6585972309112549, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.4672, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001295840352468576, | |
| "grad_norm": 1.6722080707550049, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.3507, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.001347673966567319, | |
| "grad_norm": 2.0271666049957275, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 2.3528, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.001399507580666062, | |
| "grad_norm": 1.7645819187164307, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.4185, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.001399507580666062, | |
| "eval_loss": 2.4034855365753174, | |
| "eval_runtime": 1773.3095, | |
| "eval_samples_per_second": 9.162, | |
| "eval_steps_per_second": 1.145, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.001451341194764805, | |
| "grad_norm": 1.8423093557357788, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.4298, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.001503174808863548, | |
| "grad_norm": 2.868162155151367, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.3648, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0015550084229622911, | |
| "grad_norm": 2.26560378074646, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.3836, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0016068420370610342, | |
| "grad_norm": 1.755423665046692, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.3411, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0016586756511597772, | |
| "grad_norm": 1.673230528831482, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 2.5404, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0017105092652585203, | |
| "grad_norm": 1.7275108098983765, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.492, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.001762342879357263, | |
| "grad_norm": 1.8824987411499023, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.5956, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0018141764934560061, | |
| "grad_norm": 1.8801908493041992, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.427, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0018660101075547492, | |
| "grad_norm": 2.096254587173462, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.4255, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0018660101075547492, | |
| "eval_loss": 2.351825475692749, | |
| "eval_runtime": 1774.3205, | |
| "eval_samples_per_second": 9.157, | |
| "eval_steps_per_second": 1.145, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0019178437216534922, | |
| "grad_norm": 2.0004048347473145, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.0979, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0019696773357522354, | |
| "grad_norm": 2.086202621459961, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 2.2972, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0020215109498509785, | |
| "grad_norm": 1.5125240087509155, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.3179, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0020733445639497215, | |
| "grad_norm": 1.5626386404037476, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.2221, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0021251781780484646, | |
| "grad_norm": 1.8468570709228516, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 2.436, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0021770117921472076, | |
| "grad_norm": 1.7203456163406372, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.3326, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0022288454062459506, | |
| "grad_norm": 1.7609412670135498, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.4079, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0022806790203446937, | |
| "grad_norm": 1.9543553590774536, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 2.3076, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0023325126344434367, | |
| "grad_norm": 1.9948867559432983, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.3295, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0023325126344434367, | |
| "eval_loss": 2.3102049827575684, | |
| "eval_runtime": 1772.0158, | |
| "eval_samples_per_second": 9.169, | |
| "eval_steps_per_second": 1.146, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0023843462485421798, | |
| "grad_norm": 1.8627070188522339, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.2056, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.002436179862640923, | |
| "grad_norm": 1.692333459854126, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 2.2567, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.002488013476739666, | |
| "grad_norm": 1.8308770656585693, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.3291, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.002539847090838409, | |
| "grad_norm": 1.635390281677246, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.4812, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.002591680704937152, | |
| "grad_norm": 1.5308008193969727, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.2056, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.70943641780224e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |