| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.016548072149594573, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00033096144299189144, | |
| "grad_norm": 1.7213324308395386, | |
| "learning_rate": 1e-05, | |
| "loss": 5.8203, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00033096144299189144, | |
| "eval_loss": 1.4616162776947021, | |
| "eval_runtime": 255.2761, | |
| "eval_samples_per_second": 4.987, | |
| "eval_steps_per_second": 2.495, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006619228859837829, | |
| "grad_norm": 1.3458422422409058, | |
| "learning_rate": 2e-05, | |
| "loss": 5.5279, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0009928843289756743, | |
| "grad_norm": 1.8355010747909546, | |
| "learning_rate": 3e-05, | |
| "loss": 6.4283, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0013238457719675658, | |
| "grad_norm": 1.6977440118789673, | |
| "learning_rate": 4e-05, | |
| "loss": 5.8083, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0016548072149594572, | |
| "grad_norm": 1.6104906797409058, | |
| "learning_rate": 5e-05, | |
| "loss": 5.7421, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016548072149594572, | |
| "eval_loss": 1.4400510787963867, | |
| "eval_runtime": 257.5858, | |
| "eval_samples_per_second": 4.942, | |
| "eval_steps_per_second": 2.473, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0019857686579513485, | |
| "grad_norm": 1.9484833478927612, | |
| "learning_rate": 6e-05, | |
| "loss": 5.6831, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0023167301009432402, | |
| "grad_norm": 1.7339823246002197, | |
| "learning_rate": 7e-05, | |
| "loss": 5.6721, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0026476915439351315, | |
| "grad_norm": 2.2821269035339355, | |
| "learning_rate": 8e-05, | |
| "loss": 5.8779, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.002978652986927023, | |
| "grad_norm": 2.1454436779022217, | |
| "learning_rate": 9e-05, | |
| "loss": 5.5083, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0033096144299189145, | |
| "grad_norm": 1.9616013765335083, | |
| "learning_rate": 0.0001, | |
| "loss": 4.9761, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0033096144299189145, | |
| "eval_loss": 1.2095388174057007, | |
| "eval_runtime": 257.6811, | |
| "eval_samples_per_second": 4.94, | |
| "eval_steps_per_second": 2.472, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0036405758729108058, | |
| "grad_norm": 2.1629104614257812, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 5.3772, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.003971537315902697, | |
| "grad_norm": 2.3832602500915527, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 4.6844, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004302498758894589, | |
| "grad_norm": 1.8890061378479004, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 4.5607, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0046334602018864805, | |
| "grad_norm": 2.4125006198883057, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 4.352, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.004964421644878371, | |
| "grad_norm": 2.396585464477539, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 4.6298, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004964421644878371, | |
| "eval_loss": 0.9934529066085815, | |
| "eval_runtime": 256.971, | |
| "eval_samples_per_second": 4.954, | |
| "eval_steps_per_second": 2.479, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005295383087870263, | |
| "grad_norm": 2.7325778007507324, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 3.8196, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.005626344530862155, | |
| "grad_norm": 2.316589832305908, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 4.0056, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.005957305973854046, | |
| "grad_norm": 2.2376809120178223, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.798, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006288267416845937, | |
| "grad_norm": 2.7583084106445312, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 3.6498, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.006619228859837829, | |
| "grad_norm": 1.9092457294464111, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.6041, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006619228859837829, | |
| "eval_loss": 0.8894630074501038, | |
| "eval_runtime": 257.2366, | |
| "eval_samples_per_second": 4.949, | |
| "eval_steps_per_second": 2.476, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006950190302829721, | |
| "grad_norm": 2.0496015548706055, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 3.731, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0072811517458216115, | |
| "grad_norm": 2.2720491886138916, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 3.4111, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.007612113188813503, | |
| "grad_norm": 2.0112087726593018, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 3.4615, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.007943074631805394, | |
| "grad_norm": 1.7832343578338623, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 3.4218, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.008274036074797287, | |
| "grad_norm": 2.1423120498657227, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.4775, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.008274036074797287, | |
| "eval_loss": 0.8443974852561951, | |
| "eval_runtime": 257.4444, | |
| "eval_samples_per_second": 4.945, | |
| "eval_steps_per_second": 2.474, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.008604997517789177, | |
| "grad_norm": 1.674811601638794, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 3.4598, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.008935958960781068, | |
| "grad_norm": 2.091567039489746, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 3.7612, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.009266920403772961, | |
| "grad_norm": 2.1018764972686768, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 3.6964, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.009597881846764852, | |
| "grad_norm": 1.832294225692749, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 3.6292, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.009928843289756743, | |
| "grad_norm": 2.0582776069641113, | |
| "learning_rate": 5e-05, | |
| "loss": 2.8792, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009928843289756743, | |
| "eval_loss": 0.811681866645813, | |
| "eval_runtime": 257.1914, | |
| "eval_samples_per_second": 4.95, | |
| "eval_steps_per_second": 2.477, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010259804732748635, | |
| "grad_norm": 1.9213557243347168, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 3.4662, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.010590766175740526, | |
| "grad_norm": 2.288318395614624, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 3.3137, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.010921727618732417, | |
| "grad_norm": 1.9459996223449707, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 3.4205, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01125268906172431, | |
| "grad_norm": 1.8996334075927734, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 3.5199, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0115836505047162, | |
| "grad_norm": 2.4128260612487793, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 3.3873, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0115836505047162, | |
| "eval_loss": 0.7911669611930847, | |
| "eval_runtime": 256.9175, | |
| "eval_samples_per_second": 4.955, | |
| "eval_steps_per_second": 2.479, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.011914611947708093, | |
| "grad_norm": 2.779157876968384, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 3.2642, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.012245573390699984, | |
| "grad_norm": 2.04453444480896, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 3.3588, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.012576534833691875, | |
| "grad_norm": 1.9138102531433105, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 3.1433, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.012907496276683767, | |
| "grad_norm": 2.5308830738067627, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 3.2754, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.013238457719675658, | |
| "grad_norm": 2.0766677856445312, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 3.1162, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.013238457719675658, | |
| "eval_loss": 0.7798255085945129, | |
| "eval_runtime": 257.4302, | |
| "eval_samples_per_second": 4.945, | |
| "eval_steps_per_second": 2.474, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.013569419162667549, | |
| "grad_norm": 2.0006625652313232, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 3.2087, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.013900380605659441, | |
| "grad_norm": 2.1602704524993896, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 3.177, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.014231342048651332, | |
| "grad_norm": 1.935172438621521, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 2.6165, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.014562303491643223, | |
| "grad_norm": 1.9336456060409546, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.9167, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.014893264934635116, | |
| "grad_norm": 1.7702152729034424, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 3.3973, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.014893264934635116, | |
| "eval_loss": 0.7749022245407104, | |
| "eval_runtime": 257.2066, | |
| "eval_samples_per_second": 4.949, | |
| "eval_steps_per_second": 2.477, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.015224226377627006, | |
| "grad_norm": 1.742775559425354, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 3.1991, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.015555187820618897, | |
| "grad_norm": 2.1770782470703125, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 2.8691, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.015886149263610788, | |
| "grad_norm": 2.1110646724700928, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.4055, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.016217110706602682, | |
| "grad_norm": 1.619126796722412, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 3.075, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.016548072149594573, | |
| "grad_norm": 2.098407030105591, | |
| "learning_rate": 0.0, | |
| "loss": 2.9933, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.016548072149594573, | |
| "eval_loss": 0.7738050818443298, | |
| "eval_runtime": 256.8475, | |
| "eval_samples_per_second": 4.956, | |
| "eval_steps_per_second": 2.48, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.28697269714944e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |