| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5681818181818182, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007575757575757576, | |
| "grad_norm": 0.7533223628997803, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6509, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007575757575757576, | |
| "eval_loss": 1.8315911293029785, | |
| "eval_runtime": 20.8027, | |
| "eval_samples_per_second": 5.336, | |
| "eval_steps_per_second": 0.673, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.015151515151515152, | |
| "grad_norm": 0.9126298427581787, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9551, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.022727272727272728, | |
| "grad_norm": 1.181591272354126, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9402, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.030303030303030304, | |
| "grad_norm": 1.1208268404006958, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3827, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03787878787878788, | |
| "grad_norm": 1.2570971250534058, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5976, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.045454545454545456, | |
| "grad_norm": 1.388541340827942, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0091, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05303030303030303, | |
| "grad_norm": 1.4139080047607422, | |
| "learning_rate": 7e-05, | |
| "loss": 1.817, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.06060606060606061, | |
| "grad_norm": 0.9321380257606506, | |
| "learning_rate": 8e-05, | |
| "loss": 1.7755, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.06818181818181818, | |
| "grad_norm": 0.840682864189148, | |
| "learning_rate": 9e-05, | |
| "loss": 1.0306, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06818181818181818, | |
| "eval_loss": 1.762389898300171, | |
| "eval_runtime": 19.337, | |
| "eval_samples_per_second": 5.74, | |
| "eval_steps_per_second": 0.724, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "grad_norm": 1.8093039989471436, | |
| "learning_rate": 0.0001, | |
| "loss": 2.7722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "grad_norm": 1.243077278137207, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 1.0837, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 0.7991745471954346, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.6041, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09848484848484848, | |
| "grad_norm": 0.7331045866012573, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 1.2589, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.10606060606060606, | |
| "grad_norm": 0.8287887573242188, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 1.1675, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.11363636363636363, | |
| "grad_norm": 1.3653068542480469, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.0063, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12121212121212122, | |
| "grad_norm": 0.5731700658798218, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.5773, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.12878787878787878, | |
| "grad_norm": 0.9433817863464355, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 1.3717, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.13636363636363635, | |
| "grad_norm": 1.1009303331375122, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.3868, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.13636363636363635, | |
| "eval_loss": 1.5803662538528442, | |
| "eval_runtime": 19.3482, | |
| "eval_samples_per_second": 5.737, | |
| "eval_steps_per_second": 0.724, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.14393939393939395, | |
| "grad_norm": 0.6402528882026672, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.1916, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 0.6150195598602295, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.5243, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1590909090909091, | |
| "grad_norm": 0.5555574297904968, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.9703, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 0.5409879684448242, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 1.2571, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.17424242424242425, | |
| "grad_norm": 0.6518788933753967, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.3489, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.7059927582740784, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.6073, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1893939393939394, | |
| "grad_norm": 0.62007737159729, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.4875, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.19696969696969696, | |
| "grad_norm": 0.7638795971870422, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.4672, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.20454545454545456, | |
| "grad_norm": 1.4796452522277832, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.3313, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.20454545454545456, | |
| "eval_loss": 1.4663845300674438, | |
| "eval_runtime": 19.3368, | |
| "eval_samples_per_second": 5.74, | |
| "eval_steps_per_second": 0.724, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.21212121212121213, | |
| "grad_norm": 1.1566555500030518, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.5992, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.2196969696969697, | |
| "grad_norm": 1.1568214893341064, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 1.7715, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 1.0576083660125732, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.2864, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23484848484848486, | |
| "grad_norm": 2.068882703781128, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 1.8751, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "grad_norm": 0.8438417315483093, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.7766, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.0660183429718018, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.299, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.25757575757575757, | |
| "grad_norm": 0.6727858781814575, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 1.2796, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.26515151515151514, | |
| "grad_norm": 11.034136772155762, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.7611, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 1.0035715103149414, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.2469, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "eval_loss": 1.4248439073562622, | |
| "eval_runtime": 19.3143, | |
| "eval_samples_per_second": 5.747, | |
| "eval_steps_per_second": 0.725, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2803030303030303, | |
| "grad_norm": 0.628416121006012, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.4924, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.2878787878787879, | |
| "grad_norm": 1.3458894491195679, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 1.428, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.29545454545454547, | |
| "grad_norm": 0.8480287194252014, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.2251, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.30303030303030304, | |
| "grad_norm": 0.6633424758911133, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.576, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3106060606060606, | |
| "grad_norm": 0.6817510724067688, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 1.4075, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.3181818181818182, | |
| "grad_norm": 1.002746820449829, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.1732, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.32575757575757575, | |
| "grad_norm": 0.9981942176818848, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.8795, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.8124499917030334, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.1241, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.3409090909090909, | |
| "grad_norm": 0.6578967571258545, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.581, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3409090909090909, | |
| "eval_loss": 1.404599905014038, | |
| "eval_runtime": 19.3122, | |
| "eval_samples_per_second": 5.748, | |
| "eval_steps_per_second": 0.725, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3484848484848485, | |
| "grad_norm": 0.7794574499130249, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.1908, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.3560606060606061, | |
| "grad_norm": 0.9210363626480103, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 1.1709, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.9551446437835693, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.9377, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3712121212121212, | |
| "grad_norm": 1.0424453020095825, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 1.2212, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.3787878787878788, | |
| "grad_norm": 0.5941653251647949, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.9869, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.38636363636363635, | |
| "grad_norm": 0.5107312798500061, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.1914, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3939393939393939, | |
| "grad_norm": 0.9764262437820435, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 1.3385, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.4015151515151515, | |
| "grad_norm": 0.8512474894523621, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.8658, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.4090909090909091, | |
| "grad_norm": 0.7570309638977051, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.8693, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.4090909090909091, | |
| "eval_loss": 1.3887274265289307, | |
| "eval_runtime": 19.3267, | |
| "eval_samples_per_second": 5.743, | |
| "eval_steps_per_second": 0.724, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 0.48338302969932556, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1701, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.42424242424242425, | |
| "grad_norm": 4.494511127471924, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 1.3139, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.4318181818181818, | |
| "grad_norm": 0.8095260858535767, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.2047, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.4393939393939394, | |
| "grad_norm": 1.4627423286437988, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 1.2054, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.44696969696969696, | |
| "grad_norm": 1.4437791109085083, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 1.3913, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 0.6056512594223022, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.1049, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4621212121212121, | |
| "grad_norm": 1.0000839233398438, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.7936, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.4696969696969697, | |
| "grad_norm": 0.6748735308647156, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 1.2368, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.4772727272727273, | |
| "grad_norm": 0.5631169080734253, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.996, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.4772727272727273, | |
| "eval_loss": 1.3775228261947632, | |
| "eval_runtime": 19.3436, | |
| "eval_samples_per_second": 5.738, | |
| "eval_steps_per_second": 0.724, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "grad_norm": 0.7982277274131775, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.9235, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.49242424242424243, | |
| "grad_norm": 0.6029773354530334, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 1.176, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.6423137784004211, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.2562, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.5075757575757576, | |
| "grad_norm": 0.9672674536705017, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 2.1069, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.5151515151515151, | |
| "grad_norm": 0.7522253394126892, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 1.7731, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.5227272727272727, | |
| "grad_norm": 0.8013143539428711, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.9471, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.5303030303030303, | |
| "grad_norm": 1.507096767425537, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 1.9572, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5378787878787878, | |
| "grad_norm": 1.2770590782165527, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 1.3016, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.7512741684913635, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.9932, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "eval_loss": 1.3720847368240356, | |
| "eval_runtime": 19.3385, | |
| "eval_samples_per_second": 5.74, | |
| "eval_steps_per_second": 0.724, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.553030303030303, | |
| "grad_norm": 1.2508105039596558, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.5886, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.5606060606060606, | |
| "grad_norm": 1.4357924461364746, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 1.2307, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.5681818181818182, | |
| "grad_norm": 1.215354084968567, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.2534, | |
| "step": 75 | |
| } | |
| ], | |
| "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": 9.90419523553198e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |