{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.06763897696047347, "eval_steps": 100, "global_step": 400, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0001690974424011837, "eval_loss": 0.6459227800369263, "eval_runtime": 201.5767, "eval_samples_per_second": 12.353, "eval_steps_per_second": 6.176, "step": 1 }, { "epoch": 0.0008454872120059184, "grad_norm": 0.15712441504001617, "learning_rate": 1.6666666666666667e-05, "loss": 0.5307, "step": 5 }, { "epoch": 0.0016909744240118367, "grad_norm": 0.20046602189540863, "learning_rate": 3.3333333333333335e-05, "loss": 0.5676, "step": 10 }, { "epoch": 0.0025364616360177552, "grad_norm": 0.20849193632602692, "learning_rate": 5e-05, "loss": 0.5701, "step": 15 }, { "epoch": 0.0033819488480236735, "grad_norm": 0.27390506863594055, "learning_rate": 6.666666666666667e-05, "loss": 0.6423, "step": 20 }, { "epoch": 0.004227436060029592, "grad_norm": 0.22099897265434265, "learning_rate": 8.333333333333334e-05, "loss": 0.5055, "step": 25 }, { "epoch": 0.0050729232720355105, "grad_norm": 0.2435217648744583, "learning_rate": 0.0001, "loss": 0.5144, "step": 30 }, { "epoch": 0.005918410484041429, "grad_norm": 0.29201892018318176, "learning_rate": 9.995494831023409e-05, "loss": 0.5433, "step": 35 }, { "epoch": 0.006763897696047347, "grad_norm": 0.40463411808013916, "learning_rate": 9.981987442712633e-05, "loss": 0.5898, "step": 40 }, { "epoch": 0.007609384908053266, "grad_norm": 0.3479752838611603, "learning_rate": 9.959502176294383e-05, "loss": 0.5306, "step": 45 }, { "epoch": 0.008454872120059184, "grad_norm": 0.6131821274757385, "learning_rate": 9.928079551738543e-05, "loss": 0.5225, "step": 50 }, { "epoch": 0.009300359332065102, "grad_norm": 0.1976812481880188, "learning_rate": 9.887776194738432e-05, "loss": 0.4832, "step": 55 }, { "epoch": 0.010145846544071021, "grad_norm": 0.19775013625621796, "learning_rate": 9.838664734667495e-05, "loss": 0.5004, "step": 60 }, { "epoch": 0.01099133375607694, "grad_norm": 0.19548407196998596, "learning_rate": 9.780833673696254e-05, "loss": 0.5248, "step": 65 }, { "epoch": 0.011836820968082858, "grad_norm": 0.2102847397327423, "learning_rate": 9.714387227305422e-05, "loss": 0.4872, "step": 70 }, { "epoch": 0.012682308180088777, "grad_norm": 0.20248602330684662, "learning_rate": 9.639445136482548e-05, "loss": 0.5021, "step": 75 }, { "epoch": 0.013527795392094694, "grad_norm": 0.2491309642791748, "learning_rate": 9.55614245194068e-05, "loss": 0.5686, "step": 80 }, { "epoch": 0.014373282604100613, "grad_norm": 0.2637186050415039, "learning_rate": 9.464629290747842e-05, "loss": 0.4991, "step": 85 }, { "epoch": 0.015218769816106531, "grad_norm": 0.31047478318214417, "learning_rate": 9.365070565805941e-05, "loss": 0.5182, "step": 90 }, { "epoch": 0.01606425702811245, "grad_norm": 0.39293938875198364, "learning_rate": 9.257645688666556e-05, "loss": 0.5507, "step": 95 }, { "epoch": 0.016909744240118367, "grad_norm": 0.7622083425521851, "learning_rate": 9.142548246219212e-05, "loss": 0.5148, "step": 100 }, { "epoch": 0.016909744240118367, "eval_loss": 0.49994635581970215, "eval_runtime": 203.6154, "eval_samples_per_second": 12.229, "eval_steps_per_second": 6.114, "step": 100 }, { "epoch": 0.017755231452124286, "grad_norm": 0.1981024593114853, "learning_rate": 9.019985651834703e-05, "loss": 0.4192, "step": 105 }, { "epoch": 0.018600718664130204, "grad_norm": 0.19838833808898926, "learning_rate": 8.890178771592199e-05, "loss": 0.4863, "step": 110 }, { "epoch": 0.019446205876136123, "grad_norm": 0.21621914207935333, "learning_rate": 8.753361526263621e-05, "loss": 0.5245, "step": 115 }, { "epoch": 0.020291693088142042, "grad_norm": 0.2197190523147583, "learning_rate": 8.609780469772623e-05, "loss": 0.4785, "step": 120 }, { "epoch": 0.02113718030014796, "grad_norm": 0.2307160645723343, "learning_rate": 8.459694344887732e-05, "loss": 0.4939, "step": 125 }, { "epoch": 0.02198266751215388, "grad_norm": 0.2556055784225464, "learning_rate": 8.303373616950408e-05, "loss": 0.5202, "step": 130 }, { "epoch": 0.022828154724159798, "grad_norm": 0.3053711950778961, "learning_rate": 8.141099986478212e-05, "loss": 0.5207, "step": 135 }, { "epoch": 0.023673641936165717, "grad_norm": 0.3650396168231964, "learning_rate": 7.973165881521434e-05, "loss": 0.5146, "step": 140 }, { "epoch": 0.024519129148171635, "grad_norm": 0.4481881856918335, "learning_rate": 7.799873930687978e-05, "loss": 0.5183, "step": 145 }, { "epoch": 0.025364616360177554, "grad_norm": 0.7726387977600098, "learning_rate": 7.621536417786159e-05, "loss": 0.4749, "step": 150 }, { "epoch": 0.02621010357218347, "grad_norm": 0.18763859570026398, "learning_rate": 7.438474719068173e-05, "loss": 0.4664, "step": 155 }, { "epoch": 0.027055590784189388, "grad_norm": 0.1818513423204422, "learning_rate": 7.251018724088367e-05, "loss": 0.4774, "step": 160 }, { "epoch": 0.027901077996195307, "grad_norm": 0.23504586517810822, "learning_rate": 7.059506241219965e-05, "loss": 0.4938, "step": 165 }, { "epoch": 0.028746565208201225, "grad_norm": 0.22553101181983948, "learning_rate": 6.864282388901544e-05, "loss": 0.4936, "step": 170 }, { "epoch": 0.029592052420207144, "grad_norm": 0.21830540895462036, "learning_rate": 6.665698973710288e-05, "loss": 0.4305, "step": 175 }, { "epoch": 0.030437539632213063, "grad_norm": 0.2550351619720459, "learning_rate": 6.464113856382752e-05, "loss": 0.4829, "step": 180 }, { "epoch": 0.03128302684421898, "grad_norm": 0.3035644590854645, "learning_rate": 6.259890306925627e-05, "loss": 0.4895, "step": 185 }, { "epoch": 0.0321285140562249, "grad_norm": 0.3919617533683777, "learning_rate": 6.0533963499786314e-05, "loss": 0.5451, "step": 190 }, { "epoch": 0.032974001268230815, "grad_norm": 0.4303884208202362, "learning_rate": 5.8450041016092464e-05, "loss": 0.4667, "step": 195 }, { "epoch": 0.033819488480236734, "grad_norm": 0.829088032245636, "learning_rate": 5.6350890987343944e-05, "loss": 0.4154, "step": 200 }, { "epoch": 0.033819488480236734, "eval_loss": 0.47849199175834656, "eval_runtime": 203.4816, "eval_samples_per_second": 12.237, "eval_steps_per_second": 6.118, "step": 200 }, { "epoch": 0.03466497569224265, "grad_norm": 0.1753915548324585, "learning_rate": 5.4240296223775465e-05, "loss": 0.4032, "step": 205 }, { "epoch": 0.03551046290424857, "grad_norm": 0.19739754498004913, "learning_rate": 5.212206015980742e-05, "loss": 0.4729, "step": 210 }, { "epoch": 0.03635595011625449, "grad_norm": 0.22465206682682037, "learning_rate": 5e-05, "loss": 0.4631, "step": 215 }, { "epoch": 0.03720143732826041, "grad_norm": 0.24705617129802704, "learning_rate": 4.78779398401926e-05, "loss": 0.4618, "step": 220 }, { "epoch": 0.03804692454026633, "grad_norm": 0.28556302189826965, "learning_rate": 4.575970377622456e-05, "loss": 0.5512, "step": 225 }, { "epoch": 0.038892411752272246, "grad_norm": 0.3080470561981201, "learning_rate": 4.364910901265606e-05, "loss": 0.4949, "step": 230 }, { "epoch": 0.039737898964278165, "grad_norm": 0.3479246199131012, "learning_rate": 4.1549958983907555e-05, "loss": 0.4499, "step": 235 }, { "epoch": 0.040583386176284084, "grad_norm": 0.3600875735282898, "learning_rate": 3.94660365002137e-05, "loss": 0.4454, "step": 240 }, { "epoch": 0.04142887338829, "grad_norm": 0.5033133625984192, "learning_rate": 3.740109693074375e-05, "loss": 0.4829, "step": 245 }, { "epoch": 0.04227436060029592, "grad_norm": 0.9750732779502869, "learning_rate": 3.5358861436172485e-05, "loss": 0.4892, "step": 250 }, { "epoch": 0.04311984781230184, "grad_norm": 0.18351560831069946, "learning_rate": 3.334301026289712e-05, "loss": 0.4062, "step": 255 }, { "epoch": 0.04396533502430776, "grad_norm": 0.21633638441562653, "learning_rate": 3.135717611098458e-05, "loss": 0.4706, "step": 260 }, { "epoch": 0.04481082223631368, "grad_norm": 0.21196366846561432, "learning_rate": 2.9404937587800375e-05, "loss": 0.4781, "step": 265 }, { "epoch": 0.045656309448319596, "grad_norm": 0.25018104910850525, "learning_rate": 2.748981275911633e-05, "loss": 0.4947, "step": 270 }, { "epoch": 0.046501796660325515, "grad_norm": 0.27837511897087097, "learning_rate": 2.5615252809318284e-05, "loss": 0.4824, "step": 275 }, { "epoch": 0.04734728387233143, "grad_norm": 0.28721874952316284, "learning_rate": 2.3784635822138424e-05, "loss": 0.4753, "step": 280 }, { "epoch": 0.04819277108433735, "grad_norm": 0.3178447484970093, "learning_rate": 2.2001260693120233e-05, "loss": 0.4643, "step": 285 }, { "epoch": 0.04903825829634327, "grad_norm": 0.3714665174484253, "learning_rate": 2.026834118478567e-05, "loss": 0.4892, "step": 290 }, { "epoch": 0.04988374550834919, "grad_norm": 0.470114141702652, "learning_rate": 1.858900013521788e-05, "loss": 0.5121, "step": 295 }, { "epoch": 0.05072923272035511, "grad_norm": 0.8095857501029968, "learning_rate": 1.6966263830495936e-05, "loss": 0.4473, "step": 300 }, { "epoch": 0.05072923272035511, "eval_loss": 0.4689193367958069, "eval_runtime": 203.7914, "eval_samples_per_second": 12.218, "eval_steps_per_second": 6.109, "step": 300 }, { "epoch": 0.05157471993236102, "grad_norm": 0.17946785688400269, "learning_rate": 1.5403056551122697e-05, "loss": 0.4154, "step": 305 }, { "epoch": 0.05242020714436694, "grad_norm": 0.23259112238883972, "learning_rate": 1.3902195302273779e-05, "loss": 0.4642, "step": 310 }, { "epoch": 0.05326569435637286, "grad_norm": 0.21611733734607697, "learning_rate": 1.246638473736378e-05, "loss": 0.4635, "step": 315 }, { "epoch": 0.054111181568378776, "grad_norm": 0.24835757911205292, "learning_rate": 1.1098212284078036e-05, "loss": 0.4903, "step": 320 }, { "epoch": 0.054956668780384695, "grad_norm": 0.26646360754966736, "learning_rate": 9.800143481652979e-06, "loss": 0.4367, "step": 325 }, { "epoch": 0.05580215599239061, "grad_norm": 0.3183721899986267, "learning_rate": 8.574517537807897e-06, "loss": 0.4668, "step": 330 }, { "epoch": 0.05664764320439653, "grad_norm": 0.3372812867164612, "learning_rate": 7.423543113334436e-06, "loss": 0.4071, "step": 335 }, { "epoch": 0.05749313041640245, "grad_norm": 0.4076240360736847, "learning_rate": 6.349294341940593e-06, "loss": 0.4916, "step": 340 }, { "epoch": 0.05833861762840837, "grad_norm": 0.5152222514152527, "learning_rate": 5.353707092521582e-06, "loss": 0.5302, "step": 345 }, { "epoch": 0.05918410484041429, "grad_norm": 0.8785345554351807, "learning_rate": 4.43857548059321e-06, "loss": 0.4065, "step": 350 }, { "epoch": 0.06002959205242021, "grad_norm": 0.18611513078212738, "learning_rate": 3.605548635174533e-06, "loss": 0.4258, "step": 355 }, { "epoch": 0.060875079264426125, "grad_norm": 0.220343679189682, "learning_rate": 2.85612772694579e-06, "loss": 0.4736, "step": 360 }, { "epoch": 0.061720566476432044, "grad_norm": 0.22353102266788483, "learning_rate": 2.191663263037458e-06, "loss": 0.4437, "step": 365 }, { "epoch": 0.06256605368843796, "grad_norm": 0.24478361010551453, "learning_rate": 1.6133526533250565e-06, "loss": 0.4699, "step": 370 }, { "epoch": 0.06341154090044387, "grad_norm": 0.2732827663421631, "learning_rate": 1.1222380526156928e-06, "loss": 0.4612, "step": 375 }, { "epoch": 0.0642570281124498, "grad_norm": 0.2980816066265106, "learning_rate": 7.192044826145771e-07, "loss": 0.4447, "step": 380 }, { "epoch": 0.06510251532445571, "grad_norm": 0.3370252549648285, "learning_rate": 4.049782370561583e-07, "loss": 0.4983, "step": 385 }, { "epoch": 0.06594800253646163, "grad_norm": 0.38385066390037537, "learning_rate": 1.8012557287367392e-07, "loss": 0.4681, "step": 390 }, { "epoch": 0.06679348974846755, "grad_norm": 0.4089011251926422, "learning_rate": 4.5051689765929214e-08, "loss": 0.4885, "step": 395 }, { "epoch": 0.06763897696047347, "grad_norm": 0.8262264728546143, "learning_rate": 0.0, "loss": 0.4751, "step": 400 }, { "epoch": 0.06763897696047347, "eval_loss": 0.4665312170982361, "eval_runtime": 203.2583, "eval_samples_per_second": 12.25, "eval_steps_per_second": 6.125, "step": 400 } ], "logging_steps": 5, "max_steps": 400, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.3762794998464512e+17, "train_batch_size": 2, "trial_name": null, "trial_params": null }