| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09030874301518316, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0018061748603036632, | |
| "grad_norm": 1.2947936058044434, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8216, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0018061748603036632, | |
| "eval_loss": 0.8639734983444214, | |
| "eval_runtime": 155.1336, | |
| "eval_samples_per_second": 24.044, | |
| "eval_steps_per_second": 6.014, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0036123497206073264, | |
| "grad_norm": 1.2276273965835571, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8113, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0054185245809109895, | |
| "grad_norm": 1.1577560901641846, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 0.8154, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007224699441214653, | |
| "grad_norm": 0.8684483766555786, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 0.7326, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.009030874301518315, | |
| "grad_norm": 0.7157322764396667, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.6237, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010837049161821979, | |
| "grad_norm": 0.6510230302810669, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.58, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.012643224022125641, | |
| "grad_norm": 0.552749514579773, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 0.5556, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.014449398882429305, | |
| "grad_norm": 0.4990958869457245, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.4981, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.016255573742732968, | |
| "grad_norm": 0.5622621178627014, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 0.546, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01806174860303663, | |
| "grad_norm": 0.5227020382881165, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.494, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019867923463340296, | |
| "grad_norm": 0.40778839588165283, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.4547, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.021674098323643958, | |
| "grad_norm": 0.36594679951667786, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.4738, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02348027318394762, | |
| "grad_norm": 0.3439774215221405, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 0.5147, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02348027318394762, | |
| "eval_loss": 0.4746288061141968, | |
| "eval_runtime": 155.2285, | |
| "eval_samples_per_second": 24.029, | |
| "eval_steps_per_second": 6.01, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.025286448044251283, | |
| "grad_norm": 0.36557599902153015, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.4698, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02709262290455495, | |
| "grad_norm": 0.3393869996070862, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 0.4489, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02889879776485861, | |
| "grad_norm": 0.3452422618865967, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.4892, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.030704972625162273, | |
| "grad_norm": 0.3072020411491394, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.4762, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.032511147485465935, | |
| "grad_norm": 0.2750939130783081, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.4658, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0343173223457696, | |
| "grad_norm": 0.24484656751155853, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 0.4467, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03612349720607326, | |
| "grad_norm": 0.24473077058792114, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.4508, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03792967206637693, | |
| "grad_norm": 0.22997206449508667, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 0.4423, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03973584692668059, | |
| "grad_norm": 0.25137704610824585, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.4633, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.041542021786984254, | |
| "grad_norm": 0.23138833045959473, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.4377, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.043348196647287916, | |
| "grad_norm": 0.2416175752878189, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.4491, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04515437150759158, | |
| "grad_norm": 0.23560528457164764, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 0.4388, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04696054636789524, | |
| "grad_norm": 0.22770123183727264, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4517, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04696054636789524, | |
| "eval_loss": 0.4333505928516388, | |
| "eval_runtime": 155.3537, | |
| "eval_samples_per_second": 24.01, | |
| "eval_steps_per_second": 6.006, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0487667212281989, | |
| "grad_norm": 0.23669229447841644, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 0.4397, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.050572896088502566, | |
| "grad_norm": 0.22281350195407867, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.4151, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.052379070948806235, | |
| "grad_norm": 0.22589480876922607, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.4299, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0541852458091099, | |
| "grad_norm": 0.21541890501976013, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.4296, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05599142066941356, | |
| "grad_norm": 0.2260853797197342, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 0.4267, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.05779759552971722, | |
| "grad_norm": 0.2251616269350052, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.4368, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.059603770390020884, | |
| "grad_norm": 0.22280317544937134, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 0.4432, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.061409945250324546, | |
| "grad_norm": 0.22349373996257782, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.4452, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.06321612011062822, | |
| "grad_norm": 0.21193207800388336, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.4131, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06502229497093187, | |
| "grad_norm": 0.20389844477176666, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.4067, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06682846983123554, | |
| "grad_norm": 0.22179871797561646, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 0.4175, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0686346446915392, | |
| "grad_norm": 0.21262727677822113, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.3996, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.07044081955184286, | |
| "grad_norm": 0.21592184901237488, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 0.4185, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07044081955184286, | |
| "eval_loss": 0.4238123595714569, | |
| "eval_runtime": 155.3671, | |
| "eval_samples_per_second": 24.008, | |
| "eval_steps_per_second": 6.005, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07224699441214652, | |
| "grad_norm": 0.23702038824558258, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.4505, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07405316927245019, | |
| "grad_norm": 0.21211156249046326, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.4021, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.07585934413275386, | |
| "grad_norm": 0.2211143970489502, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.4463, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07766551899305751, | |
| "grad_norm": 0.20987620949745178, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 0.4313, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.07947169385336118, | |
| "grad_norm": 0.2077188938856125, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.4092, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.08127786871366484, | |
| "grad_norm": 0.1921454817056656, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 0.4024, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08308404357396851, | |
| "grad_norm": 0.22376859188079834, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.4143, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.08489021843427216, | |
| "grad_norm": 0.2070239931344986, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.3753, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.08669639329457583, | |
| "grad_norm": 0.22126780450344086, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.4333, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08850256815487949, | |
| "grad_norm": 0.21691939234733582, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 0.4049, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.09030874301518316, | |
| "grad_norm": 0.20602606236934662, | |
| "learning_rate": 0.0, | |
| "loss": 0.3844, | |
| "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": 3.148081105207296e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |