| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.014971180477580658, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002994236095516131, | |
| "grad_norm": 16.735450744628906, | |
| "learning_rate": 2e-05, | |
| "loss": 10.969, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002994236095516131, | |
| "eval_loss": 11.57262897491455, | |
| "eval_runtime": 82.9715, | |
| "eval_samples_per_second": 8.485, | |
| "eval_steps_per_second": 8.485, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005988472191032262, | |
| "grad_norm": 10.167157173156738, | |
| "learning_rate": 4e-05, | |
| "loss": 10.7129, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008982708286548395, | |
| "grad_norm": 15.215925216674805, | |
| "learning_rate": 6e-05, | |
| "loss": 10.434, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0011976944382064525, | |
| "grad_norm": 15.04845142364502, | |
| "learning_rate": 8e-05, | |
| "loss": 9.5194, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0011976944382064525, | |
| "eval_loss": 11.429813385009766, | |
| "eval_runtime": 85.0662, | |
| "eval_samples_per_second": 8.276, | |
| "eval_steps_per_second": 8.276, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0014971180477580657, | |
| "grad_norm": 17.0161075592041, | |
| "learning_rate": 0.0001, | |
| "loss": 10.8734, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.001796541657309679, | |
| "grad_norm": 17.905702590942383, | |
| "learning_rate": 0.00012, | |
| "loss": 11.8788, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002095965266861292, | |
| "grad_norm": 21.928312301635742, | |
| "learning_rate": 0.00014, | |
| "loss": 10.4667, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.002395388876412905, | |
| "grad_norm": 13.790472030639648, | |
| "learning_rate": 0.00016, | |
| "loss": 9.8745, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.002395388876412905, | |
| "eval_loss": 9.637001037597656, | |
| "eval_runtime": 85.1568, | |
| "eval_samples_per_second": 8.267, | |
| "eval_steps_per_second": 8.267, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.002694812485964518, | |
| "grad_norm": 12.036263465881348, | |
| "learning_rate": 0.00018, | |
| "loss": 7.8039, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0029942360955161314, | |
| "grad_norm": 13.046463012695312, | |
| "learning_rate": 0.0002, | |
| "loss": 7.8042, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0032936597050677446, | |
| "grad_norm": 24.538808822631836, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 8.7757, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.003593083314619358, | |
| "grad_norm": 26.114961624145508, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 5.6577, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003593083314619358, | |
| "eval_loss": 5.31056022644043, | |
| "eval_runtime": 85.1933, | |
| "eval_samples_per_second": 8.264, | |
| "eval_steps_per_second": 8.264, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003892506924170971, | |
| "grad_norm": 17.937646865844727, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 4.7382, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.004191930533722584, | |
| "grad_norm": 21.441980361938477, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 5.3976, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0044913541432741975, | |
| "grad_norm": 32.05086898803711, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 3.6275, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00479077775282581, | |
| "grad_norm": 43.66167449951172, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 2.7349, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.00479077775282581, | |
| "eval_loss": 4.855716705322266, | |
| "eval_runtime": 83.6415, | |
| "eval_samples_per_second": 8.417, | |
| "eval_steps_per_second": 8.417, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.005090201362377423, | |
| "grad_norm": 26.169757843017578, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 4.8565, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.005389624971929036, | |
| "grad_norm": 27.142152786254883, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 4.9723, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00568904858148065, | |
| "grad_norm": 18.188777923583984, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 2.6307, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.005988472191032263, | |
| "grad_norm": 21.539011001586914, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.5522, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005988472191032263, | |
| "eval_loss": 4.069602012634277, | |
| "eval_runtime": 83.15, | |
| "eval_samples_per_second": 8.467, | |
| "eval_steps_per_second": 8.467, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006287895800583876, | |
| "grad_norm": 18.60074806213379, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 4.4333, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.006587319410135489, | |
| "grad_norm": 15.545724868774414, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 3.6316, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0068867430196871025, | |
| "grad_norm": 23.216941833496094, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 3.2189, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.007186166629238716, | |
| "grad_norm": 16.716583251953125, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 3.0378, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007186166629238716, | |
| "eval_loss": 3.7071402072906494, | |
| "eval_runtime": 83.2372, | |
| "eval_samples_per_second": 8.458, | |
| "eval_steps_per_second": 8.458, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007485590238790329, | |
| "grad_norm": 18.92926597595215, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 2.2366, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.007785013848341942, | |
| "grad_norm": 30.20500946044922, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 3.3568, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.008084437457893555, | |
| "grad_norm": 15.139212608337402, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 3.0559, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008383861067445169, | |
| "grad_norm": 20.157346725463867, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 3.2366, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.008383861067445169, | |
| "eval_loss": 3.518523931503296, | |
| "eval_runtime": 83.1412, | |
| "eval_samples_per_second": 8.468, | |
| "eval_steps_per_second": 8.468, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.008683284676996781, | |
| "grad_norm": 16.283584594726562, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 3.5697, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.008982708286548395, | |
| "grad_norm": 32.41582489013672, | |
| "learning_rate": 0.0001, | |
| "loss": 3.4125, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009282131896100007, | |
| "grad_norm": 26.877944946289062, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 3.3671, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.00958155550565162, | |
| "grad_norm": 27.856897354125977, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 3.5875, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.00958155550565162, | |
| "eval_loss": 3.401846408843994, | |
| "eval_runtime": 83.4215, | |
| "eval_samples_per_second": 8.439, | |
| "eval_steps_per_second": 8.439, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.009880979115203234, | |
| "grad_norm": 41.4135627746582, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 3.9928, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.010180402724754846, | |
| "grad_norm": 29.63922882080078, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 2.1258, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.01047982633430646, | |
| "grad_norm": 34.11370086669922, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 2.8219, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.010779249943858073, | |
| "grad_norm": 19.921571731567383, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 4.0769, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010779249943858073, | |
| "eval_loss": 3.3533554077148438, | |
| "eval_runtime": 83.2266, | |
| "eval_samples_per_second": 8.459, | |
| "eval_steps_per_second": 8.459, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.011078673553409687, | |
| "grad_norm": 21.9058895111084, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 2.9877, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0113780971629613, | |
| "grad_norm": 20.251365661621094, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 2.4354, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.011677520772512913, | |
| "grad_norm": 31.047348022460938, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 3.9592, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.011976944382064526, | |
| "grad_norm": 20.160734176635742, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 2.8831, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.011976944382064526, | |
| "eval_loss": 3.2888846397399902, | |
| "eval_runtime": 83.3941, | |
| "eval_samples_per_second": 8.442, | |
| "eval_steps_per_second": 8.442, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01227636799161614, | |
| "grad_norm": 22.80976104736328, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 3.3648, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.012575791601167752, | |
| "grad_norm": 40.02175521850586, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 3.0131, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.012875215210719364, | |
| "grad_norm": 40.88200759887695, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 4.2676, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.013174638820270979, | |
| "grad_norm": 44.29335021972656, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 3.5078, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.013174638820270979, | |
| "eval_loss": 3.2594785690307617, | |
| "eval_runtime": 83.3165, | |
| "eval_samples_per_second": 8.45, | |
| "eval_steps_per_second": 8.45, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.013474062429822591, | |
| "grad_norm": 34.80168151855469, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 4.7818, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013773486039374205, | |
| "grad_norm": 25.594444274902344, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 3.0296, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.014072909648925817, | |
| "grad_norm": 39.9898567199707, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 1.9811, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.014372333258477431, | |
| "grad_norm": 58.41755294799805, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 2.8259, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.014372333258477431, | |
| "eval_loss": 3.2478675842285156, | |
| "eval_runtime": 83.2404, | |
| "eval_samples_per_second": 8.457, | |
| "eval_steps_per_second": 8.457, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.014671756868029044, | |
| "grad_norm": 47.00809860229492, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 3.4149, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.014971180477580658, | |
| "grad_norm": 37.111209869384766, | |
| "learning_rate": 0.0, | |
| "loss": 3.3272, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.85471820890112e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |