| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08771929824561403, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0017543859649122807, | |
| "grad_norm": 8.301186561584473, | |
| "learning_rate": 2e-05, | |
| "loss": 14.8867, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0017543859649122807, | |
| "eval_loss": 3.213578701019287, | |
| "eval_runtime": 4.8661, | |
| "eval_samples_per_second": 49.321, | |
| "eval_steps_per_second": 24.66, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0035087719298245615, | |
| "grad_norm": 7.328826904296875, | |
| "learning_rate": 4e-05, | |
| "loss": 12.4698, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005263157894736842, | |
| "grad_norm": 8.940353393554688, | |
| "learning_rate": 6e-05, | |
| "loss": 13.4153, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007017543859649123, | |
| "grad_norm": 7.479343891143799, | |
| "learning_rate": 8e-05, | |
| "loss": 12.7646, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.008771929824561403, | |
| "grad_norm": 8.830097198486328, | |
| "learning_rate": 0.0001, | |
| "loss": 13.7101, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010526315789473684, | |
| "grad_norm": 8.61213493347168, | |
| "learning_rate": 0.00012, | |
| "loss": 12.4308, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.012280701754385965, | |
| "grad_norm": 7.831013202667236, | |
| "learning_rate": 0.00014, | |
| "loss": 12.4425, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.014035087719298246, | |
| "grad_norm": 7.707040309906006, | |
| "learning_rate": 0.00016, | |
| "loss": 11.3206, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.015789473684210527, | |
| "grad_norm": 8.187994003295898, | |
| "learning_rate": 0.00018, | |
| "loss": 11.6318, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.017543859649122806, | |
| "grad_norm": 6.9909467697143555, | |
| "learning_rate": 0.0002, | |
| "loss": 9.5057, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01929824561403509, | |
| "grad_norm": 8.407684326171875, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 11.0084, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.021052631578947368, | |
| "grad_norm": 7.068591117858887, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 9.5473, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02280701754385965, | |
| "grad_norm": 7.12372350692749, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 8.5598, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02280701754385965, | |
| "eval_loss": 2.0703299045562744, | |
| "eval_runtime": 4.222, | |
| "eval_samples_per_second": 56.845, | |
| "eval_steps_per_second": 28.422, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02456140350877193, | |
| "grad_norm": 6.699890613555908, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 7.9868, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02631578947368421, | |
| "grad_norm": 6.799102306365967, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 8.6767, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.028070175438596492, | |
| "grad_norm": 6.97033166885376, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 7.6695, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02982456140350877, | |
| "grad_norm": 7.434906005859375, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 7.1493, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.031578947368421054, | |
| "grad_norm": 7.40922737121582, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 7.2892, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03333333333333333, | |
| "grad_norm": 7.225566387176514, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 7.3546, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03508771929824561, | |
| "grad_norm": 8.341033935546875, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 7.8588, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03684210526315789, | |
| "grad_norm": 6.631705284118652, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 7.1027, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03859649122807018, | |
| "grad_norm": 7.108447551727295, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 6.9375, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.04035087719298246, | |
| "grad_norm": 6.394927501678467, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 6.8771, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.042105263157894736, | |
| "grad_norm": 5.567875862121582, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 6.5752, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.043859649122807015, | |
| "grad_norm": 6.8372626304626465, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 6.7518, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0456140350877193, | |
| "grad_norm": 5.824334144592285, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 7.0496, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0456140350877193, | |
| "eval_loss": 1.7120705842971802, | |
| "eval_runtime": 4.2762, | |
| "eval_samples_per_second": 56.125, | |
| "eval_steps_per_second": 28.062, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04736842105263158, | |
| "grad_norm": 6.506723880767822, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 6.5923, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04912280701754386, | |
| "grad_norm": 5.621894359588623, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 6.1451, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.05087719298245614, | |
| "grad_norm": 6.360438346862793, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 7.1938, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.05263157894736842, | |
| "grad_norm": 6.532863616943359, | |
| "learning_rate": 0.0001, | |
| "loss": 6.9762, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.054385964912280704, | |
| "grad_norm": 6.711584091186523, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 7.1213, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.056140350877192984, | |
| "grad_norm": 6.394320487976074, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 7.1146, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.05789473684210526, | |
| "grad_norm": 6.20180082321167, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 7.1802, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.05964912280701754, | |
| "grad_norm": 6.715718746185303, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 7.0011, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.06140350877192982, | |
| "grad_norm": 6.306797027587891, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 7.1968, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06315789473684211, | |
| "grad_norm": 7.334612846374512, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 6.2764, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06491228070175438, | |
| "grad_norm": 6.493529796600342, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 6.5648, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.06666666666666667, | |
| "grad_norm": 5.998836994171143, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 5.7859, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.06842105263157895, | |
| "grad_norm": 6.2978925704956055, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 6.1908, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06842105263157895, | |
| "eval_loss": 1.6301758289337158, | |
| "eval_runtime": 4.3085, | |
| "eval_samples_per_second": 55.704, | |
| "eval_steps_per_second": 27.852, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 6.921982765197754, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 7.4431, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07192982456140351, | |
| "grad_norm": 5.440037727355957, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 6.0957, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.07368421052631578, | |
| "grad_norm": 6.525667667388916, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 7.1308, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07543859649122807, | |
| "grad_norm": 6.312617301940918, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 6.5642, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.07719298245614035, | |
| "grad_norm": 6.488813400268555, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 5.9102, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.07894736842105263, | |
| "grad_norm": 6.369126319885254, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 6.7954, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08070175438596491, | |
| "grad_norm": 5.541263103485107, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 6.9496, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0824561403508772, | |
| "grad_norm": 5.264928340911865, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 6.3512, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.08421052631578947, | |
| "grad_norm": 5.724519729614258, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 6.058, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08596491228070176, | |
| "grad_norm": 5.485682010650635, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 5.7932, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.08771929824561403, | |
| "grad_norm": 6.535744667053223, | |
| "learning_rate": 0.0, | |
| "loss": 6.6898, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3264746618880000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |