{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.04402054292002935, "eval_steps": 42, "global_step": 210, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0002096216329525207, "eval_loss": 1.4208108186721802, "eval_runtime": 163.5996, "eval_samples_per_second": 49.114, "eval_steps_per_second": 6.143, "step": 1 }, { "epoch": 0.0006288648988575621, "grad_norm": 0.39683395624160767, "learning_rate": 3e-05, "loss": 1.4354, "step": 3 }, { "epoch": 0.0012577297977151242, "grad_norm": 0.4367702305316925, "learning_rate": 6e-05, "loss": 1.4342, "step": 6 }, { "epoch": 0.0018865946965726864, "grad_norm": 0.37110769748687744, "learning_rate": 9e-05, "loss": 1.3479, "step": 9 }, { "epoch": 0.0025154595954302483, "grad_norm": 0.36730247735977173, "learning_rate": 9.999588943391597e-05, "loss": 1.361, "step": 12 }, { "epoch": 0.0031443244942878103, "grad_norm": 0.35089993476867676, "learning_rate": 9.99743108100344e-05, "loss": 1.2814, "step": 15 }, { "epoch": 0.0037731893931453727, "grad_norm": 0.38672298192977905, "learning_rate": 9.993424445916923e-05, "loss": 1.3016, "step": 18 }, { "epoch": 0.004402054292002935, "grad_norm": 0.321205198764801, "learning_rate": 9.987570520365104e-05, "loss": 1.2163, "step": 21 }, { "epoch": 0.005030919190860497, "grad_norm": 0.3282659649848938, "learning_rate": 9.979871469976196e-05, "loss": 1.2312, "step": 24 }, { "epoch": 0.005659784089718059, "grad_norm": 0.3265891373157501, "learning_rate": 9.970330142972401e-05, "loss": 1.2505, "step": 27 }, { "epoch": 0.006288648988575621, "grad_norm": 0.3450949192047119, "learning_rate": 9.95895006911623e-05, "loss": 1.1982, "step": 30 }, { "epoch": 0.0069175138874331835, "grad_norm": 0.3153296411037445, "learning_rate": 9.945735458404681e-05, "loss": 1.1669, "step": 33 }, { "epoch": 0.0075463787862907455, "grad_norm": 0.3419332802295685, "learning_rate": 9.930691199511775e-05, "loss": 1.2492, "step": 36 }, { "epoch": 0.008175243685148307, "grad_norm": 0.2867058515548706, "learning_rate": 9.91382285798002e-05, "loss": 1.1652, "step": 39 }, { "epoch": 0.00880410858400587, "grad_norm": 0.3220820426940918, "learning_rate": 9.895136674161465e-05, "loss": 1.1996, "step": 42 }, { "epoch": 0.00880410858400587, "eval_loss": 1.1905547380447388, "eval_runtime": 165.0006, "eval_samples_per_second": 48.697, "eval_steps_per_second": 6.091, "step": 42 }, { "epoch": 0.009432973482863432, "grad_norm": 0.30218949913978577, "learning_rate": 9.874639560909117e-05, "loss": 1.1916, "step": 45 }, { "epoch": 0.010061838381720993, "grad_norm": 0.3343711197376251, "learning_rate": 9.852339101019574e-05, "loss": 1.1487, "step": 48 }, { "epoch": 0.010690703280578556, "grad_norm": 0.3268805146217346, "learning_rate": 9.828243544427796e-05, "loss": 1.1781, "step": 51 }, { "epoch": 0.011319568179436117, "grad_norm": 0.34138286113739014, "learning_rate": 9.802361805155097e-05, "loss": 1.1215, "step": 54 }, { "epoch": 0.01194843307829368, "grad_norm": 0.3412030339241028, "learning_rate": 9.774703458011453e-05, "loss": 1.2044, "step": 57 }, { "epoch": 0.012577297977151241, "grad_norm": 0.3168179392814636, "learning_rate": 9.745278735053343e-05, "loss": 1.1087, "step": 60 }, { "epoch": 0.013206162876008804, "grad_norm": 0.3409186005592346, "learning_rate": 9.714098521798465e-05, "loss": 1.2062, "step": 63 }, { "epoch": 0.013835027774866367, "grad_norm": 0.29721856117248535, "learning_rate": 9.681174353198687e-05, "loss": 1.1158, "step": 66 }, { "epoch": 0.014463892673723928, "grad_norm": 0.3369769752025604, "learning_rate": 9.64651840937276e-05, "loss": 1.1834, "step": 69 }, { "epoch": 0.015092757572581491, "grad_norm": 0.31268638372421265, "learning_rate": 9.610143511100354e-05, "loss": 1.1367, "step": 72 }, { "epoch": 0.015721622471439052, "grad_norm": 0.333360493183136, "learning_rate": 9.572063115079063e-05, "loss": 1.1805, "step": 75 }, { "epoch": 0.016350487370296613, "grad_norm": 0.2951726019382477, "learning_rate": 9.53229130894619e-05, "loss": 1.1459, "step": 78 }, { "epoch": 0.016979352269154178, "grad_norm": 0.36082521080970764, "learning_rate": 9.490842806067095e-05, "loss": 1.17, "step": 81 }, { "epoch": 0.01760821716801174, "grad_norm": 0.33050239086151123, "learning_rate": 9.44773294009206e-05, "loss": 1.1611, "step": 84 }, { "epoch": 0.01760821716801174, "eval_loss": 1.147385835647583, "eval_runtime": 165.0588, "eval_samples_per_second": 48.68, "eval_steps_per_second": 6.089, "step": 84 }, { "epoch": 0.0182370820668693, "grad_norm": 0.3466218411922455, "learning_rate": 9.40297765928369e-05, "loss": 1.181, "step": 87 }, { "epoch": 0.018865946965726865, "grad_norm": 0.3219515085220337, "learning_rate": 9.356593520616948e-05, "loss": 1.1236, "step": 90 }, { "epoch": 0.019494811864584426, "grad_norm": 0.35251516103744507, "learning_rate": 9.308597683653975e-05, "loss": 1.1749, "step": 93 }, { "epoch": 0.020123676763441987, "grad_norm": 0.3366820514202118, "learning_rate": 9.259007904196023e-05, "loss": 1.1411, "step": 96 }, { "epoch": 0.020752541662299548, "grad_norm": 0.3501774072647095, "learning_rate": 9.207842527714767e-05, "loss": 1.122, "step": 99 }, { "epoch": 0.021381406561157112, "grad_norm": 0.3260366916656494, "learning_rate": 9.155120482565521e-05, "loss": 1.1032, "step": 102 }, { "epoch": 0.022010271460014674, "grad_norm": 0.3130464255809784, "learning_rate": 9.10086127298478e-05, "loss": 1.1025, "step": 105 }, { "epoch": 0.022639136358872235, "grad_norm": 0.3319336473941803, "learning_rate": 9.045084971874738e-05, "loss": 1.1279, "step": 108 }, { "epoch": 0.0232680012577298, "grad_norm": 0.3683924973011017, "learning_rate": 8.987812213377424e-05, "loss": 1.1384, "step": 111 }, { "epoch": 0.02389686615658736, "grad_norm": 0.3043256998062134, "learning_rate": 8.929064185241213e-05, "loss": 1.106, "step": 114 }, { "epoch": 0.02452573105544492, "grad_norm": 0.3852984607219696, "learning_rate": 8.868862620982534e-05, "loss": 1.123, "step": 117 }, { "epoch": 0.025154595954302483, "grad_norm": 0.36246711015701294, "learning_rate": 8.807229791845673e-05, "loss": 1.1248, "step": 120 }, { "epoch": 0.025783460853160047, "grad_norm": 0.3592873215675354, "learning_rate": 8.744188498563641e-05, "loss": 1.0983, "step": 123 }, { "epoch": 0.02641232575201761, "grad_norm": 0.3461982309818268, "learning_rate": 8.679762062923175e-05, "loss": 1.1211, "step": 126 }, { "epoch": 0.02641232575201761, "eval_loss": 1.1246975660324097, "eval_runtime": 165.051, "eval_samples_per_second": 48.682, "eval_steps_per_second": 6.089, "step": 126 }, { "epoch": 0.02704119065087517, "grad_norm": 0.3487012982368469, "learning_rate": 8.613974319136958e-05, "loss": 1.1212, "step": 129 }, { "epoch": 0.027670055549732734, "grad_norm": 0.37176796793937683, "learning_rate": 8.54684960502629e-05, "loss": 1.0707, "step": 132 }, { "epoch": 0.028298920448590295, "grad_norm": 0.3854920268058777, "learning_rate": 8.478412753017433e-05, "loss": 1.1204, "step": 135 }, { "epoch": 0.028927785347447856, "grad_norm": 0.3561486303806305, "learning_rate": 8.408689080954998e-05, "loss": 1.1238, "step": 138 }, { "epoch": 0.029556650246305417, "grad_norm": 0.35637167096138, "learning_rate": 8.33770438273574e-05, "loss": 1.1415, "step": 141 }, { "epoch": 0.030185515145162982, "grad_norm": 0.37608301639556885, "learning_rate": 8.265484918766243e-05, "loss": 1.1374, "step": 144 }, { "epoch": 0.030814380044020543, "grad_norm": 0.3702179789543152, "learning_rate": 8.192057406248028e-05, "loss": 1.1001, "step": 147 }, { "epoch": 0.031443244942878104, "grad_norm": 0.35054582357406616, "learning_rate": 8.117449009293668e-05, "loss": 1.122, "step": 150 }, { "epoch": 0.03207210984173567, "grad_norm": 0.35157763957977295, "learning_rate": 8.041687328877567e-05, "loss": 1.1126, "step": 153 }, { "epoch": 0.032700974740593226, "grad_norm": 0.3567587435245514, "learning_rate": 7.964800392625129e-05, "loss": 1.1196, "step": 156 }, { "epoch": 0.03332983963945079, "grad_norm": 0.33376169204711914, "learning_rate": 7.886816644444098e-05, "loss": 1.1269, "step": 159 }, { "epoch": 0.033958704538308356, "grad_norm": 0.33005964756011963, "learning_rate": 7.807764934001874e-05, "loss": 1.0881, "step": 162 }, { "epoch": 0.03458756943716591, "grad_norm": 0.3465365171432495, "learning_rate": 7.727674506052743e-05, "loss": 1.0912, "step": 165 }, { "epoch": 0.03521643433602348, "grad_norm": 0.3523496985435486, "learning_rate": 7.646574989618938e-05, "loss": 1.0962, "step": 168 }, { "epoch": 0.03521643433602348, "eval_loss": 1.1077728271484375, "eval_runtime": 165.0063, "eval_samples_per_second": 48.695, "eval_steps_per_second": 6.091, "step": 168 }, { "epoch": 0.03584529923488104, "grad_norm": 0.3241373300552368, "learning_rate": 7.564496387029532e-05, "loss": 1.1243, "step": 171 }, { "epoch": 0.0364741641337386, "grad_norm": 0.33574438095092773, "learning_rate": 7.481469062821252e-05, "loss": 1.1031, "step": 174 }, { "epoch": 0.037103029032596165, "grad_norm": 0.38013267517089844, "learning_rate": 7.39752373250527e-05, "loss": 1.1191, "step": 177 }, { "epoch": 0.03773189393145373, "grad_norm": 0.35769525170326233, "learning_rate": 7.312691451204178e-05, "loss": 1.0992, "step": 180 }, { "epoch": 0.03836075883031129, "grad_norm": 0.32820338010787964, "learning_rate": 7.227003602163295e-05, "loss": 1.0621, "step": 183 }, { "epoch": 0.03898962372916885, "grad_norm": 0.32962238788604736, "learning_rate": 7.14049188514063e-05, "loss": 1.0828, "step": 186 }, { "epoch": 0.03961848862802641, "grad_norm": 0.34197545051574707, "learning_rate": 7.05318830467969e-05, "loss": 1.1168, "step": 189 }, { "epoch": 0.040247353526883974, "grad_norm": 0.37021368741989136, "learning_rate": 6.965125158269619e-05, "loss": 1.1374, "step": 192 }, { "epoch": 0.04087621842574154, "grad_norm": 0.3301369547843933, "learning_rate": 6.876335024396872e-05, "loss": 1.054, "step": 195 }, { "epoch": 0.041505083324599096, "grad_norm": 0.35479190945625305, "learning_rate": 6.786850750493006e-05, "loss": 1.0923, "step": 198 }, { "epoch": 0.04213394822345666, "grad_norm": 0.35211730003356934, "learning_rate": 6.696705440782938e-05, "loss": 1.1124, "step": 201 }, { "epoch": 0.042762813122314225, "grad_norm": 0.4026525616645813, "learning_rate": 6.605932444038229e-05, "loss": 1.1229, "step": 204 }, { "epoch": 0.04339167802117178, "grad_norm": 0.40407177805900574, "learning_rate": 6.514565341239861e-05, "loss": 1.066, "step": 207 }, { "epoch": 0.04402054292002935, "grad_norm": 0.3534255027770996, "learning_rate": 6.422637933155162e-05, "loss": 1.0462, "step": 210 }, { "epoch": 0.04402054292002935, "eval_loss": 1.0946003198623657, "eval_runtime": 165.0602, "eval_samples_per_second": 48.679, "eval_steps_per_second": 6.089, "step": 210 } ], "logging_steps": 3, "max_steps": 500, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 42, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 4.479610490191872e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }