| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03125, | |
| "grad_norm": 2.2531771659851074, | |
| "learning_rate": 6.25e-06, | |
| "loss": 2.2918, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 2.1257269382476807, | |
| "learning_rate": 3.125e-05, | |
| "loss": 2.284, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 1.5896614789962769, | |
| "learning_rate": 6.25e-05, | |
| "loss": 2.2225, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 0.7286556363105774, | |
| "learning_rate": 9.375e-05, | |
| "loss": 2.1021, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.6398385763168335, | |
| "learning_rate": 0.000125, | |
| "loss": 1.9996, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 0.4788893461227417, | |
| "learning_rate": 0.00015625, | |
| "loss": 1.9178, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.3934371769428253, | |
| "learning_rate": 0.0001875, | |
| "loss": 1.845, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.684882402420044, | |
| "eval_runtime": 0.8301, | |
| "eval_samples_per_second": 7.228, | |
| "eval_steps_per_second": 1.205, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.09375, | |
| "grad_norm": 0.28933560848236084, | |
| "learning_rate": 0.00019994645874763658, | |
| "loss": 1.786, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.20734435319900513, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 1.7402, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.40625, | |
| "grad_norm": 0.16501550376415253, | |
| "learning_rate": 0.00019899620837148077, | |
| "loss": 1.7072, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.13184033334255219, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 1.6846, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 0.11590086668729782, | |
| "learning_rate": 0.00019686915803565934, | |
| "loss": 1.665, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.11442063748836517, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 1.6455, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.612865686416626, | |
| "eval_runtime": 0.83, | |
| "eval_samples_per_second": 7.229, | |
| "eval_steps_per_second": 1.205, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 0.09601253271102905, | |
| "learning_rate": 0.0001935905926757326, | |
| "loss": 1.6364, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "grad_norm": 0.10327372699975967, | |
| "learning_rate": 0.00019153114791194473, | |
| "loss": 1.6227, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.34375, | |
| "grad_norm": 0.09720147401094437, | |
| "learning_rate": 0.00018919948565893142, | |
| "loss": 1.611, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.09051357209682465, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.6067, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.65625, | |
| "grad_norm": 0.10204631090164185, | |
| "learning_rate": 0.0001837480354951308, | |
| "loss": 1.6009, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "grad_norm": 0.11083294451236725, | |
| "learning_rate": 0.00018064446042674828, | |
| "loss": 1.5825, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 0.1064315065741539, | |
| "learning_rate": 0.0001773010453362737, | |
| "loss": 1.5798, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.6013729572296143, | |
| "eval_runtime": 0.8292, | |
| "eval_samples_per_second": 7.236, | |
| "eval_steps_per_second": 1.206, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 0.0996951013803482, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 1.5716, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.28125, | |
| "grad_norm": 0.10399119555950165, | |
| "learning_rate": 0.00016993515264033672, | |
| "loss": 1.5691, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.4375, | |
| "grad_norm": 0.0989290326833725, | |
| "learning_rate": 0.00016593458151000688, | |
| "loss": 1.5641, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.59375, | |
| "grad_norm": 0.108305424451828, | |
| "learning_rate": 0.00016173791815707051, | |
| "loss": 1.5539, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.11103138327598572, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 1.5511, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.90625, | |
| "grad_norm": 0.11553769558668137, | |
| "learning_rate": 0.0001528067850650368, | |
| "loss": 1.546, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.597513437271118, | |
| "eval_runtime": 0.8293, | |
| "eval_samples_per_second": 7.235, | |
| "eval_steps_per_second": 1.206, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 0.11389793455600739, | |
| "learning_rate": 0.00014809887689193877, | |
| "loss": 1.5361, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.21875, | |
| "grad_norm": 0.1214594691991806, | |
| "learning_rate": 0.00014324792063301662, | |
| "loss": 1.5346, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 0.11472882330417633, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.5339, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.53125, | |
| "grad_norm": 0.10563024878501892, | |
| "learning_rate": 0.00013317495417533524, | |
| "loss": 1.5314, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "grad_norm": 0.10731685161590576, | |
| "learning_rate": 0.00012798290140309923, | |
| "loss": 1.5284, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.84375, | |
| "grad_norm": 0.10968785732984543, | |
| "learning_rate": 0.00012270762630343734, | |
| "loss": 1.5233, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.11469796299934387, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.5244, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.5962610244750977, | |
| "eval_runtime": 0.8296, | |
| "eval_samples_per_second": 7.233, | |
| "eval_steps_per_second": 1.205, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.15625, | |
| "grad_norm": 0.10656954348087311, | |
| "learning_rate": 0.00011197036553049625, | |
| "loss": 1.5151, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "grad_norm": 0.12216157466173172, | |
| "learning_rate": 0.00010654031292301432, | |
| "loss": 1.5119, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.46875, | |
| "grad_norm": 0.11443573981523514, | |
| "learning_rate": 0.00010109080914941824, | |
| "loss": 1.5107, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 0.11616319417953491, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 1.5165, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.78125, | |
| "grad_norm": 0.1112031489610672, | |
| "learning_rate": 9.019828596704394e-05, | |
| "loss": 1.509, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 5.9375, | |
| "grad_norm": 0.11148490756750107, | |
| "learning_rate": 8.478766138100834e-05, | |
| "loss": 1.5072, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.594440460205078, | |
| "eval_runtime": 0.8285, | |
| "eval_samples_per_second": 7.242, | |
| "eval_steps_per_second": 1.207, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 6.09375, | |
| "grad_norm": 0.11521880328655243, | |
| "learning_rate": 7.942227893077652e-05, | |
| "loss": 1.5015, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.11160895973443985, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 1.4991, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.40625, | |
| "grad_norm": 0.11868108808994293, | |
| "learning_rate": 6.889088592289093e-05, | |
| "loss": 1.5008, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "grad_norm": 0.11169662326574326, | |
| "learning_rate": 6.375619617162985e-05, | |
| "loss": 1.5006, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.71875, | |
| "grad_norm": 0.11667145788669586, | |
| "learning_rate": 5.872929701956054e-05, | |
| "loss": 1.4989, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 0.10703334957361221, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 1.501, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.594101667404175, | |
| "eval_runtime": 0.8304, | |
| "eval_samples_per_second": 7.225, | |
| "eval_steps_per_second": 1.204, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 7.03125, | |
| "grad_norm": 0.11598368734121323, | |
| "learning_rate": 4.9058306315915826e-05, | |
| "loss": 1.5005, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 7.1875, | |
| "grad_norm": 0.11565965414047241, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.4975, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.34375, | |
| "grad_norm": 0.10369732230901718, | |
| "learning_rate": 3.999287600755192e-05, | |
| "loss": 1.4969, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.10466954857110977, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 1.4949, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.65625, | |
| "grad_norm": 0.09984403848648071, | |
| "learning_rate": 3.164076979771287e-05, | |
| "loss": 1.4901, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "grad_norm": 0.10001492500305176, | |
| "learning_rate": 2.776360379402445e-05, | |
| "loss": 1.492, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.96875, | |
| "grad_norm": 0.10096081346273422, | |
| "learning_rate": 2.4101271875283817e-05, | |
| "loss": 1.4858, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.5943994522094727, | |
| "eval_runtime": 0.8348, | |
| "eval_samples_per_second": 7.187, | |
| "eval_steps_per_second": 1.198, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "grad_norm": 0.09807378798723221, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 1.4889, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.28125, | |
| "grad_norm": 0.1029752716422081, | |
| "learning_rate": 1.7464006691513623e-05, | |
| "loss": 1.4893, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 8.4375, | |
| "grad_norm": 0.0935431718826294, | |
| "learning_rate": 1.4508812932705363e-05, | |
| "loss": 1.4883, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.59375, | |
| "grad_norm": 0.09739290177822113, | |
| "learning_rate": 1.1807873565164506e-05, | |
| "loss": 1.4858, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.10472334176301956, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 1.4933, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.90625, | |
| "grad_norm": 0.09686731547117233, | |
| "learning_rate": 7.200108780781556e-06, | |
| "loss": 1.4917, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.5955541133880615, | |
| "eval_runtime": 0.8323, | |
| "eval_samples_per_second": 7.209, | |
| "eval_steps_per_second": 1.201, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 9.0625, | |
| "grad_norm": 0.09799981117248535, | |
| "learning_rate": 5.306987050489442e-06, | |
| "loss": 1.4904, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.21875, | |
| "grad_norm": 0.10055091232061386, | |
| "learning_rate": 3.6954863292237297e-06, | |
| "loss": 1.4813, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.09550856798887253, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 1.4873, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.53125, | |
| "grad_norm": 0.09861158579587936, | |
| "learning_rate": 1.3356667915121025e-06, | |
| "loss": 1.4911, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 9.6875, | |
| "grad_norm": 0.10015583783388138, | |
| "learning_rate": 5.943661777680354e-07, | |
| "loss": 1.4933, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.84375, | |
| "grad_norm": 0.09337816387414932, | |
| "learning_rate": 1.487021060236904e-07, | |
| "loss": 1.49, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.08882776647806168, | |
| "learning_rate": 0.0, | |
| "loss": 1.4886, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.595360517501831, | |
| "eval_runtime": 0.827, | |
| "eval_samples_per_second": 7.255, | |
| "eval_steps_per_second": 1.209, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 320, | |
| "total_flos": 9.830314275137126e+17, | |
| "train_loss": 1.5905956603586673, | |
| "train_runtime": 1497.0202, | |
| "train_samples_per_second": 109.25, | |
| "train_steps_per_second": 0.214 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 320, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.830314275137126e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |