| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.19860973187686196, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003972194637537239, | |
| "grad_norm": 0.4402098059654236, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8096, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003972194637537239, | |
| "eval_loss": 0.9886362552642822, | |
| "eval_runtime": 26.8876, | |
| "eval_samples_per_second": 3.942, | |
| "eval_steps_per_second": 1.971, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007944389275074478, | |
| "grad_norm": 0.387863427400589, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7586, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.011916583912611719, | |
| "grad_norm": 0.3501233160495758, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5331, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.015888778550148957, | |
| "grad_norm": 0.48034775257110596, | |
| "learning_rate": 4e-05, | |
| "loss": 0.621, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.019860973187686197, | |
| "grad_norm": 0.4095894694328308, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7758, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.019860973187686197, | |
| "eval_loss": 0.9715360403060913, | |
| "eval_runtime": 26.9366, | |
| "eval_samples_per_second": 3.935, | |
| "eval_steps_per_second": 1.968, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.023833167825223437, | |
| "grad_norm": 0.606968343257904, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7785, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.027805362462760674, | |
| "grad_norm": 1.2388267517089844, | |
| "learning_rate": 7e-05, | |
| "loss": 0.9735, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.031777557100297914, | |
| "grad_norm": 1.5921814441680908, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9913, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03574975173783516, | |
| "grad_norm": 2.2070536613464355, | |
| "learning_rate": 9e-05, | |
| "loss": 1.5737, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.039721946375372394, | |
| "grad_norm": 0.39528149366378784, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4109, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.039721946375372394, | |
| "eval_loss": 0.6905767917633057, | |
| "eval_runtime": 27.0421, | |
| "eval_samples_per_second": 3.92, | |
| "eval_steps_per_second": 1.96, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04369414101290963, | |
| "grad_norm": 0.9849607944488525, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.5394, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.047666335650446874, | |
| "grad_norm": 0.7983195185661316, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.5027, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05163853028798411, | |
| "grad_norm": 1.2366907596588135, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.749, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.05561072492552135, | |
| "grad_norm": 0.5725651979446411, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.4126, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05958291956305859, | |
| "grad_norm": 0.7858136892318726, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.4712, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05958291956305859, | |
| "eval_loss": 0.43833062052726746, | |
| "eval_runtime": 27.0611, | |
| "eval_samples_per_second": 3.917, | |
| "eval_steps_per_second": 1.959, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06355511420059583, | |
| "grad_norm": 0.487553209066391, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.2656, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.06752730883813307, | |
| "grad_norm": 0.596966564655304, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.2773, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.07149950347567031, | |
| "grad_norm": 0.7163795828819275, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.3517, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07547169811320754, | |
| "grad_norm": 0.4074513614177704, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.3021, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.07944389275074479, | |
| "grad_norm": 0.6424429416656494, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.3123, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07944389275074479, | |
| "eval_loss": 0.35487380623817444, | |
| "eval_runtime": 27.0826, | |
| "eval_samples_per_second": 3.914, | |
| "eval_steps_per_second": 1.957, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08341608738828203, | |
| "grad_norm": 0.6385417580604553, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.3987, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08738828202581926, | |
| "grad_norm": 0.4243467450141907, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.2488, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0913604766633565, | |
| "grad_norm": 0.34767284989356995, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.2428, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.09533267130089375, | |
| "grad_norm": 0.6192077994346619, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.3342, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09930486593843098, | |
| "grad_norm": 0.47093358635902405, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.2855, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09930486593843098, | |
| "eval_loss": 0.30741116404533386, | |
| "eval_runtime": 27.0911, | |
| "eval_samples_per_second": 3.913, | |
| "eval_steps_per_second": 1.956, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.10327706057596822, | |
| "grad_norm": 0.5661317110061646, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.3159, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.10724925521350546, | |
| "grad_norm": 0.3478457033634186, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.2156, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1112214498510427, | |
| "grad_norm": 0.3984580338001251, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.265, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.11519364448857994, | |
| "grad_norm": 0.5232352614402771, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.2667, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.11916583912611718, | |
| "grad_norm": 0.5363503694534302, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11916583912611718, | |
| "eval_loss": 0.2915026843547821, | |
| "eval_runtime": 27.0837, | |
| "eval_samples_per_second": 3.914, | |
| "eval_steps_per_second": 1.957, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12313803376365443, | |
| "grad_norm": 0.3917446434497833, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.2119, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.12711022840119166, | |
| "grad_norm": 0.2668308615684509, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.2285, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.13108242303872888, | |
| "grad_norm": 0.32772475481033325, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.2214, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.13505461767626614, | |
| "grad_norm": 0.4210818111896515, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.2204, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.13902681231380337, | |
| "grad_norm": 0.470721960067749, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.261, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.13902681231380337, | |
| "eval_loss": 0.2774577736854553, | |
| "eval_runtime": 27.0865, | |
| "eval_samples_per_second": 3.913, | |
| "eval_steps_per_second": 1.957, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.14299900695134063, | |
| "grad_norm": 0.3494637608528137, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.2237, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14697120158887786, | |
| "grad_norm": 0.49559351801872253, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.2608, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "grad_norm": 0.4698236882686615, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.2241, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.15491559086395235, | |
| "grad_norm": 0.6071957349777222, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.2778, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.15888778550148958, | |
| "grad_norm": 0.302901953458786, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.2584, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15888778550148958, | |
| "eval_loss": 0.2731941044330597, | |
| "eval_runtime": 27.0826, | |
| "eval_samples_per_second": 3.914, | |
| "eval_steps_per_second": 1.957, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1628599801390268, | |
| "grad_norm": 0.7419318556785583, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.3068, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.16683217477656406, | |
| "grad_norm": 0.5678064227104187, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.2748, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1708043694141013, | |
| "grad_norm": 0.3631245791912079, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.2302, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.17477656405163852, | |
| "grad_norm": 0.48080864548683167, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.2325, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.17874875868917578, | |
| "grad_norm": 0.3569527566432953, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.1848, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.17874875868917578, | |
| "eval_loss": 0.2712867259979248, | |
| "eval_runtime": 27.0882, | |
| "eval_samples_per_second": 3.913, | |
| "eval_steps_per_second": 1.957, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.182720953326713, | |
| "grad_norm": 0.42595577239990234, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.2049, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.18669314796425024, | |
| "grad_norm": 0.5140714645385742, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.2283, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.1906653426017875, | |
| "grad_norm": 0.6511046290397644, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.2613, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.19463753723932473, | |
| "grad_norm": 0.6832799911499023, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.2708, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.19860973187686196, | |
| "grad_norm": 0.32798025012016296, | |
| "learning_rate": 0.0, | |
| "loss": 0.2238, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19860973187686196, | |
| "eval_loss": 0.2707710564136505, | |
| "eval_runtime": 27.0853, | |
| "eval_samples_per_second": 3.914, | |
| "eval_steps_per_second": 1.957, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.69912849629184e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |