| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07073386383731212, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007073386383731211, | |
| "eval_loss": 1.189466118812561, | |
| "eval_runtime": 313.795, | |
| "eval_samples_per_second": 7.588, | |
| "eval_steps_per_second": 0.95, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002122015915119363, | |
| "grad_norm": 0.5069853067398071, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2315, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004244031830238726, | |
| "grad_norm": 0.4807300567626953, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1486, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00636604774535809, | |
| "grad_norm": 0.5452137589454651, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1401, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00636604774535809, | |
| "eval_loss": 0.9386758804321289, | |
| "eval_runtime": 316.2435, | |
| "eval_samples_per_second": 7.529, | |
| "eval_steps_per_second": 0.942, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008488063660477453, | |
| "grad_norm": 0.46840620040893555, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8462, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010610079575596816, | |
| "grad_norm": 0.7444979548454285, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.6378, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01273209549071618, | |
| "grad_norm": 0.4480806887149811, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.4438, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01273209549071618, | |
| "eval_loss": 0.3728099763393402, | |
| "eval_runtime": 316.5306, | |
| "eval_samples_per_second": 7.522, | |
| "eval_steps_per_second": 0.941, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014854111405835544, | |
| "grad_norm": 0.46977323293685913, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.3508, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.016976127320954906, | |
| "grad_norm": 0.33589252829551697, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.3184, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01909814323607427, | |
| "grad_norm": 0.31081634759902954, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.3063, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01909814323607427, | |
| "eval_loss": 0.288199245929718, | |
| "eval_runtime": 316.9259, | |
| "eval_samples_per_second": 7.513, | |
| "eval_steps_per_second": 0.94, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.021220159151193633, | |
| "grad_norm": 0.2740694582462311, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.2743, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.023342175066312996, | |
| "grad_norm": 0.24397042393684387, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.2724, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02546419098143236, | |
| "grad_norm": 0.2719847857952118, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.2675, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02546419098143236, | |
| "eval_loss": 0.2569960951805115, | |
| "eval_runtime": 316.3134, | |
| "eval_samples_per_second": 7.527, | |
| "eval_steps_per_second": 0.942, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.027586206896551724, | |
| "grad_norm": 0.22842727601528168, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.2581, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.029708222811671087, | |
| "grad_norm": 0.22034530341625214, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.2418, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03183023872679045, | |
| "grad_norm": 0.27058666944503784, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.2442, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03183023872679045, | |
| "eval_loss": 0.23530101776123047, | |
| "eval_runtime": 316.3177, | |
| "eval_samples_per_second": 7.527, | |
| "eval_steps_per_second": 0.942, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03395225464190981, | |
| "grad_norm": 0.24335481226444244, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.2388, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03607427055702918, | |
| "grad_norm": 0.19779016077518463, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.2253, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.03819628647214854, | |
| "grad_norm": 0.23507341742515564, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.2155, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03819628647214854, | |
| "eval_loss": 0.22180727124214172, | |
| "eval_runtime": 316.2483, | |
| "eval_samples_per_second": 7.529, | |
| "eval_steps_per_second": 0.942, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.040318302387267906, | |
| "grad_norm": 0.2855474352836609, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.2275, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.042440318302387266, | |
| "grad_norm": 0.18945002555847168, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.2311, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04456233421750663, | |
| "grad_norm": 0.23679417371749878, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.2085, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04456233421750663, | |
| "eval_loss": 0.21260088682174683, | |
| "eval_runtime": 316.4279, | |
| "eval_samples_per_second": 7.525, | |
| "eval_steps_per_second": 0.942, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04668435013262599, | |
| "grad_norm": 0.22867415845394135, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.2051, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.04880636604774536, | |
| "grad_norm": 0.18394522368907928, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.2023, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.05092838196286472, | |
| "grad_norm": 0.20804443955421448, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.205, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05092838196286472, | |
| "eval_loss": 0.20696386694908142, | |
| "eval_runtime": 316.3831, | |
| "eval_samples_per_second": 7.526, | |
| "eval_steps_per_second": 0.942, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05305039787798409, | |
| "grad_norm": 0.20641203224658966, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.2053, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05517241379310345, | |
| "grad_norm": 0.23211583495140076, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.2086, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.057294429708222815, | |
| "grad_norm": 0.246817946434021, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.2028, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.057294429708222815, | |
| "eval_loss": 0.20408986508846283, | |
| "eval_runtime": 316.2491, | |
| "eval_samples_per_second": 7.529, | |
| "eval_steps_per_second": 0.942, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.059416445623342175, | |
| "grad_norm": 0.20014618337154388, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.2064, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "grad_norm": 0.19082359969615936, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.203, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0636604774535809, | |
| "grad_norm": 0.15606807172298431, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.1979, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0636604774535809, | |
| "eval_loss": 0.20284633338451385, | |
| "eval_runtime": 316.2258, | |
| "eval_samples_per_second": 7.529, | |
| "eval_steps_per_second": 0.942, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06578249336870026, | |
| "grad_norm": 0.19346192479133606, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.2053, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.06790450928381962, | |
| "grad_norm": 0.19841697812080383, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.1951, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.070026525198939, | |
| "grad_norm": 0.20000827312469482, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.207, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.070026525198939, | |
| "eval_loss": 0.2025783807039261, | |
| "eval_runtime": 316.4638, | |
| "eval_samples_per_second": 7.524, | |
| "eval_steps_per_second": 0.942, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.6039610198982656e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |