| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1314060446780552, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002628120893561104, | |
| "grad_norm": 5.410415172576904, | |
| "learning_rate": 1e-05, | |
| "loss": 7.6058, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002628120893561104, | |
| "eval_loss": 7.706796169281006, | |
| "eval_runtime": 7.8198, | |
| "eval_samples_per_second": 41.05, | |
| "eval_steps_per_second": 5.243, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005256241787122208, | |
| "grad_norm": 4.60823392868042, | |
| "learning_rate": 2e-05, | |
| "loss": 7.7311, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00788436268068331, | |
| "grad_norm": 4.575686931610107, | |
| "learning_rate": 3e-05, | |
| "loss": 7.018, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010512483574244415, | |
| "grad_norm": 5.470172882080078, | |
| "learning_rate": 4e-05, | |
| "loss": 8.0201, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.013140604467805518, | |
| "grad_norm": 4.377795696258545, | |
| "learning_rate": 5e-05, | |
| "loss": 7.1738, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01576872536136662, | |
| "grad_norm": 5.3419294357299805, | |
| "learning_rate": 6e-05, | |
| "loss": 7.993, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018396846254927726, | |
| "grad_norm": 4.29301643371582, | |
| "learning_rate": 7e-05, | |
| "loss": 7.4565, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02102496714848883, | |
| "grad_norm": 3.220489978790283, | |
| "learning_rate": 8e-05, | |
| "loss": 7.9336, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.023653088042049936, | |
| "grad_norm": 5.817854881286621, | |
| "learning_rate": 9e-05, | |
| "loss": 7.7279, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.023653088042049936, | |
| "eval_loss": 7.636051654815674, | |
| "eval_runtime": 7.8181, | |
| "eval_samples_per_second": 41.059, | |
| "eval_steps_per_second": 5.244, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.026281208935611037, | |
| "grad_norm": 3.344338893890381, | |
| "learning_rate": 0.0001, | |
| "loss": 7.4586, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02890932982917214, | |
| "grad_norm": 2.984314203262329, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 7.282, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03153745072273324, | |
| "grad_norm": 5.008783340454102, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 7.2996, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03416557161629435, | |
| "grad_norm": 5.446415901184082, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 7.7632, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03679369250985545, | |
| "grad_norm": 2.8699066638946533, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 7.6905, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03942181340341656, | |
| "grad_norm": 4.882473945617676, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 7.9727, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04204993429697766, | |
| "grad_norm": 3.1733548641204834, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 6.9207, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04467805519053877, | |
| "grad_norm": 4.807592868804932, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 7.7807, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04730617608409987, | |
| "grad_norm": 4.676242351531982, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 7.8454, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04730617608409987, | |
| "eval_loss": 7.441324234008789, | |
| "eval_runtime": 7.8231, | |
| "eval_samples_per_second": 41.032, | |
| "eval_steps_per_second": 5.241, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04993429697766097, | |
| "grad_norm": 4.598595142364502, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 7.7069, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.052562417871222074, | |
| "grad_norm": 4.850493907928467, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 7.5851, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05519053876478318, | |
| "grad_norm": 3.774733781814575, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 7.1, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05781865965834428, | |
| "grad_norm": 3.955918550491333, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 7.3246, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.06044678055190539, | |
| "grad_norm": 3.4571104049682617, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 6.9312, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06307490144546649, | |
| "grad_norm": 3.994743585586548, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 7.5883, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0657030223390276, | |
| "grad_norm": 3.493082284927368, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 7.5821, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0683311432325887, | |
| "grad_norm": 4.332574844360352, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 7.2888, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07095926412614981, | |
| "grad_norm": 4.837662696838379, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 6.845, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07095926412614981, | |
| "eval_loss": 7.245150089263916, | |
| "eval_runtime": 7.8267, | |
| "eval_samples_per_second": 41.013, | |
| "eval_steps_per_second": 5.238, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0735873850197109, | |
| "grad_norm": 4.511839866638184, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 7.7317, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07621550591327202, | |
| "grad_norm": 3.5038013458251953, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 7.1144, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07884362680683311, | |
| "grad_norm": 3.6477413177490234, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 6.769, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08147174770039421, | |
| "grad_norm": 3.408693790435791, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 6.6021, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08409986859395532, | |
| "grad_norm": 4.711971282958984, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 7.4735, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08672798948751642, | |
| "grad_norm": 5.069206237792969, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 7.4584, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08935611038107753, | |
| "grad_norm": 4.098939418792725, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 7.27, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09198423127463863, | |
| "grad_norm": 3.993582010269165, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 6.6474, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09461235216819974, | |
| "grad_norm": 4.524712562561035, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 7.0525, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09461235216819974, | |
| "eval_loss": 7.079109191894531, | |
| "eval_runtime": 7.8096, | |
| "eval_samples_per_second": 41.103, | |
| "eval_steps_per_second": 5.25, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09724047306176084, | |
| "grad_norm": 3.9363136291503906, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 7.2766, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09986859395532194, | |
| "grad_norm": 3.650977849960327, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 6.9182, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.10249671484888305, | |
| "grad_norm": 3.0577781200408936, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 6.5284, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10512483574244415, | |
| "grad_norm": 3.7681539058685303, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 6.9426, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10775295663600526, | |
| "grad_norm": 3.7199325561523438, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 7.1631, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.11038107752956636, | |
| "grad_norm": 4.243125915527344, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 7.092, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11300919842312747, | |
| "grad_norm": 4.146131992340088, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 7.5549, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11563731931668857, | |
| "grad_norm": 3.312317132949829, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 7.0711, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.11826544021024968, | |
| "grad_norm": 3.4535398483276367, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 7.2084, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11826544021024968, | |
| "eval_loss": 6.981412410736084, | |
| "eval_runtime": 7.8222, | |
| "eval_samples_per_second": 41.037, | |
| "eval_steps_per_second": 5.241, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.12089356110381078, | |
| "grad_norm": 3.1890342235565186, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 6.7782, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.12352168199737187, | |
| "grad_norm": 3.076508045196533, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 6.8948, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12614980289093297, | |
| "grad_norm": 4.208280563354492, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 7.0109, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1287779237844941, | |
| "grad_norm": 2.896726369857788, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 6.866, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.1314060446780552, | |
| "grad_norm": 5.758768558502197, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 6.5004, | |
| "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": 24163072868352.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |