| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.019648938957296306, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00026198585276395077, | |
| "grad_norm": 8.06531810760498, | |
| "learning_rate": 1e-05, | |
| "loss": 5.4551, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00026198585276395077, | |
| "eval_loss": 5.552332878112793, | |
| "eval_runtime": 352.8496, | |
| "eval_samples_per_second": 9.112, | |
| "eval_steps_per_second": 1.139, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005239717055279015, | |
| "grad_norm": 7.758414268493652, | |
| "learning_rate": 2e-05, | |
| "loss": 5.0221, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0007859575582918522, | |
| "grad_norm": 9.16843318939209, | |
| "learning_rate": 3e-05, | |
| "loss": 5.2987, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001047943411055803, | |
| "grad_norm": 9.832208633422852, | |
| "learning_rate": 4e-05, | |
| "loss": 5.4247, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0013099292638197536, | |
| "grad_norm": 7.5622029304504395, | |
| "learning_rate": 5e-05, | |
| "loss": 4.8809, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0015719151165837044, | |
| "grad_norm": 6.857791423797607, | |
| "learning_rate": 6e-05, | |
| "loss": 4.3287, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0018339009693476552, | |
| "grad_norm": 9.882523536682129, | |
| "learning_rate": 7e-05, | |
| "loss": 4.8656, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.002095886822111606, | |
| "grad_norm": 9.96164608001709, | |
| "learning_rate": 8e-05, | |
| "loss": 4.2639, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.002357872674875557, | |
| "grad_norm": 9.73707103729248, | |
| "learning_rate": 9e-05, | |
| "loss": 2.8029, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002357872674875557, | |
| "eval_loss": 2.4069762229919434, | |
| "eval_runtime": 352.9261, | |
| "eval_samples_per_second": 9.11, | |
| "eval_steps_per_second": 1.139, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0026198585276395073, | |
| "grad_norm": 12.502920150756836, | |
| "learning_rate": 0.0001, | |
| "loss": 2.4574, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002881844380403458, | |
| "grad_norm": 10.564509391784668, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.0574, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.003143830233167409, | |
| "grad_norm": 11.528499603271484, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.5046, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0034058160859313596, | |
| "grad_norm": 5.828344345092773, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.8369, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0036678019386953103, | |
| "grad_norm": 3.885608196258545, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.7905, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0039297877914592615, | |
| "grad_norm": 4.5050787925720215, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.6641, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004191773644223212, | |
| "grad_norm": 2.9515435695648193, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.5273, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004453759496987163, | |
| "grad_norm": 3.4887335300445557, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.5964, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.004715745349751114, | |
| "grad_norm": 5.508411407470703, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.6012, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004715745349751114, | |
| "eval_loss": 0.5490872263908386, | |
| "eval_runtime": 352.8627, | |
| "eval_samples_per_second": 9.111, | |
| "eval_steps_per_second": 1.139, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.004977731202515065, | |
| "grad_norm": 4.160599231719971, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.5502, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0052397170552790145, | |
| "grad_norm": 3.6146748065948486, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.4695, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005501702908042965, | |
| "grad_norm": 4.098548412322998, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.6587, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.005763688760806916, | |
| "grad_norm": 5.742251396179199, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.7347, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.006025674613570867, | |
| "grad_norm": 3.5661041736602783, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.4111, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.006287660466334818, | |
| "grad_norm": 2.88602614402771, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.5704, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006549646319098768, | |
| "grad_norm": 2.903674602508545, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.5034, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.006811632171862719, | |
| "grad_norm": 1.9433157444000244, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.4729, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.00707361802462667, | |
| "grad_norm": 3.4378275871276855, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.6066, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00707361802462667, | |
| "eval_loss": 0.4995511770248413, | |
| "eval_runtime": 352.9658, | |
| "eval_samples_per_second": 9.109, | |
| "eval_steps_per_second": 1.139, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007335603877390621, | |
| "grad_norm": 2.4910888671875, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.5026, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.007597589730154571, | |
| "grad_norm": 2.5030782222747803, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.5058, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.007859575582918523, | |
| "grad_norm": 3.438540458679199, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.6847, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008121561435682474, | |
| "grad_norm": 3.054363489151001, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.4804, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.008383547288446425, | |
| "grad_norm": 2.9105348587036133, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.3873, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.008645533141210375, | |
| "grad_norm": 2.1625113487243652, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.4134, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.008907518993974326, | |
| "grad_norm": 2.532012939453125, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.5296, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.009169504846738277, | |
| "grad_norm": 3.0135233402252197, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.4023, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.009431490699502228, | |
| "grad_norm": 3.0963938236236572, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.3808, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.009431490699502228, | |
| "eval_loss": 0.48174405097961426, | |
| "eval_runtime": 351.9258, | |
| "eval_samples_per_second": 9.135, | |
| "eval_steps_per_second": 1.142, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.009693476552266178, | |
| "grad_norm": 3.2138330936431885, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.448, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.00995546240503013, | |
| "grad_norm": 2.795412302017212, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.4665, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.010217448257794078, | |
| "grad_norm": 2.621596336364746, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.5508, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.010479434110558029, | |
| "grad_norm": 1.9961961507797241, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.4201, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01074141996332198, | |
| "grad_norm": 5.01107120513916, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.6618, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.01100340581608593, | |
| "grad_norm": 2.090813159942627, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.516, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.011265391668849881, | |
| "grad_norm": 2.791059970855713, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.5553, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.011527377521613832, | |
| "grad_norm": 1.8673609495162964, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.4062, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.011789363374377783, | |
| "grad_norm": 2.381378412246704, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.3776, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.011789363374377783, | |
| "eval_loss": 0.46064993739128113, | |
| "eval_runtime": 351.8933, | |
| "eval_samples_per_second": 9.136, | |
| "eval_steps_per_second": 1.142, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.012051349227141734, | |
| "grad_norm": 2.2704169750213623, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.4949, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.012313335079905684, | |
| "grad_norm": 2.7573747634887695, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.5682, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.012575320932669635, | |
| "grad_norm": 1.6240943670272827, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.4033, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.012837306785433586, | |
| "grad_norm": 2.1533713340759277, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.5008, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.013099292638197537, | |
| "grad_norm": 1.7656461000442505, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.3638, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.013361278490961487, | |
| "grad_norm": 2.3180577754974365, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.4152, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.013623264343725438, | |
| "grad_norm": 2.2583389282226562, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.4253, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.013885250196489389, | |
| "grad_norm": 2.602883815765381, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.4055, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.01414723604925334, | |
| "grad_norm": 2.4522361755371094, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.4184, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01414723604925334, | |
| "eval_loss": 0.4348742663860321, | |
| "eval_runtime": 350.6903, | |
| "eval_samples_per_second": 9.168, | |
| "eval_steps_per_second": 1.146, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01440922190201729, | |
| "grad_norm": 2.1054177284240723, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4777, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.014671207754781241, | |
| "grad_norm": 1.9587393999099731, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.4286, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.014933193607545192, | |
| "grad_norm": 2.988269329071045, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.4282, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.015195179460309143, | |
| "grad_norm": 2.121634006500244, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.4101, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.015457165313073094, | |
| "grad_norm": 2.1134636402130127, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.3569, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.015719151165837046, | |
| "grad_norm": 2.068063735961914, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.3677, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.015981137018600995, | |
| "grad_norm": 2.390730142593384, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.3862, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.016243122871364948, | |
| "grad_norm": 2.6314713954925537, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.4921, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.016505108724128897, | |
| "grad_norm": 2.0429060459136963, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.3892, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.016505108724128897, | |
| "eval_loss": 0.4214228689670563, | |
| "eval_runtime": 350.9117, | |
| "eval_samples_per_second": 9.162, | |
| "eval_steps_per_second": 1.146, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01676709457689285, | |
| "grad_norm": 1.9817379713058472, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.3156, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.017029080429656798, | |
| "grad_norm": 2.7622270584106445, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.4318, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.01729106628242075, | |
| "grad_norm": 1.9306707382202148, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.3675, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0175530521351847, | |
| "grad_norm": 3.567344903945923, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.4472, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.017815037987948652, | |
| "grad_norm": 2.1826930046081543, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.3412, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.0180770238407126, | |
| "grad_norm": 2.021453857421875, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.367, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.018339009693476554, | |
| "grad_norm": 2.4057135581970215, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.3575, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.018600995546240503, | |
| "grad_norm": 2.1014580726623535, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.3046, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.018862981399004455, | |
| "grad_norm": 2.1180341243743896, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.3779, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.018862981399004455, | |
| "eval_loss": 0.3998485803604126, | |
| "eval_runtime": 350.8524, | |
| "eval_samples_per_second": 9.163, | |
| "eval_steps_per_second": 1.146, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.019124967251768404, | |
| "grad_norm": 2.127703905105591, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.2859, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.019386953104532357, | |
| "grad_norm": 1.8953347206115723, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.2619, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.019648938957296306, | |
| "grad_norm": 3.1371781826019287, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.4874, | |
| "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": 5.56415462670336e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |