| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02313743637204998, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00046274872744099955, | |
| "grad_norm": 12.331781387329102, | |
| "learning_rate": 1e-05, | |
| "loss": 5.9018, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00046274872744099955, | |
| "eval_loss": 6.24040412902832, | |
| "eval_runtime": 110.0457, | |
| "eval_samples_per_second": 16.539, | |
| "eval_steps_per_second": 2.072, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0009254974548819991, | |
| "grad_norm": 11.983473777770996, | |
| "learning_rate": 2e-05, | |
| "loss": 6.1357, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0013882461823229986, | |
| "grad_norm": 11.926801681518555, | |
| "learning_rate": 3e-05, | |
| "loss": 6.3402, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0018509949097639982, | |
| "grad_norm": 12.546172142028809, | |
| "learning_rate": 4e-05, | |
| "loss": 6.0555, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0023137436372049976, | |
| "grad_norm": 13.246417045593262, | |
| "learning_rate": 5e-05, | |
| "loss": 6.1008, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002776492364645997, | |
| "grad_norm": 11.150693893432617, | |
| "learning_rate": 6e-05, | |
| "loss": 6.134, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003239241092086997, | |
| "grad_norm": 12.205668449401855, | |
| "learning_rate": 7e-05, | |
| "loss": 6.3615, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0037019898195279964, | |
| "grad_norm": 11.150944709777832, | |
| "learning_rate": 8e-05, | |
| "loss": 4.6314, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0041647385469689956, | |
| "grad_norm": 11.63155460357666, | |
| "learning_rate": 9e-05, | |
| "loss": 3.4176, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0041647385469689956, | |
| "eval_loss": 3.59645414352417, | |
| "eval_runtime": 108.8575, | |
| "eval_samples_per_second": 16.719, | |
| "eval_steps_per_second": 2.094, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004627487274409995, | |
| "grad_norm": 8.951762199401855, | |
| "learning_rate": 0.0001, | |
| "loss": 3.0291, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005090236001850995, | |
| "grad_norm": 8.684383392333984, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.5782, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.005552984729291994, | |
| "grad_norm": 8.277400970458984, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.9991, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.006015733456732994, | |
| "grad_norm": 16.194562911987305, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.2676, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.006478482184173994, | |
| "grad_norm": 49.7684440612793, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 1.3732, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.006941230911614993, | |
| "grad_norm": 8.08752155303955, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.1734, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.007403979639055993, | |
| "grad_norm": 5.786002159118652, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.1763, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.007866728366496992, | |
| "grad_norm": 4.976882457733154, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.8316, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.008329477093937991, | |
| "grad_norm": 5.076099395751953, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.658, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008329477093937991, | |
| "eval_loss": 0.7718914747238159, | |
| "eval_runtime": 108.8254, | |
| "eval_samples_per_second": 16.724, | |
| "eval_steps_per_second": 2.095, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008792225821378992, | |
| "grad_norm": 4.327938556671143, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.7712, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.00925497454881999, | |
| "grad_norm": 4.913948059082031, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.6246, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00971772327626099, | |
| "grad_norm": 3.194206714630127, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.4449, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01018047200370199, | |
| "grad_norm": 2.053872585296631, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.2754, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.01064322073114299, | |
| "grad_norm": 5.223790168762207, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.6558, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.011105969458583989, | |
| "grad_norm": 9.427882194519043, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8059, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01156871818602499, | |
| "grad_norm": 3.001323938369751, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.1372, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.012031466913465988, | |
| "grad_norm": 5.2901458740234375, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.4054, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.012494215640906987, | |
| "grad_norm": 16.79204750061035, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.3945, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.012494215640906987, | |
| "eval_loss": 0.3398672342300415, | |
| "eval_runtime": 108.8534, | |
| "eval_samples_per_second": 16.72, | |
| "eval_steps_per_second": 2.095, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.012956964368347987, | |
| "grad_norm": 12.270838737487793, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.9365, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.013419713095788986, | |
| "grad_norm": 9.225953102111816, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.3784, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.013882461823229986, | |
| "grad_norm": 4.924163341522217, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.401, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.014345210550670985, | |
| "grad_norm": 2.290313720703125, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.2295, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.014807959278111986, | |
| "grad_norm": 2.4892070293426514, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.1456, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.015270708005552984, | |
| "grad_norm": 1.4364653825759888, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.1347, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.015733456732993985, | |
| "grad_norm": 3.6731722354888916, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.1451, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.016196205460434984, | |
| "grad_norm": 3.2023043632507324, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.1694, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.016658954187875982, | |
| "grad_norm": 3.8125226497650146, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.3038, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.016658954187875982, | |
| "eval_loss": 0.21688657999038696, | |
| "eval_runtime": 108.8202, | |
| "eval_samples_per_second": 16.725, | |
| "eval_steps_per_second": 2.095, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.017121702915316984, | |
| "grad_norm": 1.7498458623886108, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0593, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.017584451642757983, | |
| "grad_norm": 2.2368359565734863, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.1137, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.018047200370198982, | |
| "grad_norm": 5.773807048797607, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.3029, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01850994909763998, | |
| "grad_norm": 1.4326202869415283, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0836, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01897269782508098, | |
| "grad_norm": 1.1739600896835327, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.1741, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.01943544655252198, | |
| "grad_norm": 1.4607264995574951, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.1426, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01989819527996298, | |
| "grad_norm": 4.282503128051758, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.3293, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.02036094400740398, | |
| "grad_norm": 3.917597770690918, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.1327, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.020823692734844978, | |
| "grad_norm": 3.9601786136627197, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.2175, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.020823692734844978, | |
| "eval_loss": 0.21302354335784912, | |
| "eval_runtime": 108.8413, | |
| "eval_samples_per_second": 16.722, | |
| "eval_steps_per_second": 2.095, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.02128644146228598, | |
| "grad_norm": 2.9856042861938477, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.3059, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.02174919018972698, | |
| "grad_norm": 1.7201728820800781, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.1166, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.022211938917167977, | |
| "grad_norm": 2.847334623336792, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.1798, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.022674687644608976, | |
| "grad_norm": 2.4103736877441406, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.2387, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.02313743637204998, | |
| "grad_norm": 1.2504336833953857, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0647, | |
| "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": 9779003334328320.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |