{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.994385527136619, "eval_steps": 500, "global_step": 1200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.024953212726138492, "grad_norm": 8.501412658362785, "learning_rate": 5e-06, "loss": 0.9043, "step": 10 }, { "epoch": 0.049906425452276984, "grad_norm": 4.686127489377169, "learning_rate": 5e-06, "loss": 0.7713, "step": 20 }, { "epoch": 0.07485963817841547, "grad_norm": 1.065750712888582, "learning_rate": 5e-06, "loss": 0.7261, "step": 30 }, { "epoch": 0.09981285090455397, "grad_norm": 1.1573725910946995, "learning_rate": 5e-06, "loss": 0.6939, "step": 40 }, { "epoch": 0.12476606363069245, "grad_norm": 1.0825665225776888, "learning_rate": 5e-06, "loss": 0.6773, "step": 50 }, { "epoch": 0.14971927635683094, "grad_norm": 0.8282989762012689, "learning_rate": 5e-06, "loss": 0.6551, "step": 60 }, { "epoch": 0.17467248908296942, "grad_norm": 0.8431312859314775, "learning_rate": 5e-06, "loss": 0.6458, "step": 70 }, { "epoch": 0.19962570180910794, "grad_norm": 0.7090142603822943, "learning_rate": 5e-06, "loss": 0.6379, "step": 80 }, { "epoch": 0.22457891453524642, "grad_norm": 0.6113443247263295, "learning_rate": 5e-06, "loss": 0.6258, "step": 90 }, { "epoch": 0.2495321272613849, "grad_norm": 0.5723628494735549, "learning_rate": 5e-06, "loss": 0.6294, "step": 100 }, { "epoch": 0.2744853399875234, "grad_norm": 0.5986798385585294, "learning_rate": 5e-06, "loss": 0.616, "step": 110 }, { "epoch": 0.2994385527136619, "grad_norm": 0.6703694452116971, "learning_rate": 5e-06, "loss": 0.61, "step": 120 }, { "epoch": 0.32439176543980036, "grad_norm": 0.48278079529549894, "learning_rate": 5e-06, "loss": 0.6088, "step": 130 }, { "epoch": 0.34934497816593885, "grad_norm": 0.5383252054302604, "learning_rate": 5e-06, "loss": 0.6097, "step": 140 }, { "epoch": 0.37429819089207733, "grad_norm": 0.5904470071991698, "learning_rate": 5e-06, "loss": 0.6053, "step": 150 }, { "epoch": 0.39925140361821587, "grad_norm": 0.5818363085286041, "learning_rate": 5e-06, "loss": 0.6014, "step": 160 }, { "epoch": 0.42420461634435436, "grad_norm": 0.6764973062341755, "learning_rate": 5e-06, "loss": 0.5993, "step": 170 }, { "epoch": 0.44915782907049284, "grad_norm": 0.6999555821582358, "learning_rate": 5e-06, "loss": 0.5991, "step": 180 }, { "epoch": 0.4741110417966313, "grad_norm": 0.7880847284498282, "learning_rate": 5e-06, "loss": 0.5961, "step": 190 }, { "epoch": 0.4990642545227698, "grad_norm": 0.7979122334833642, "learning_rate": 5e-06, "loss": 0.5916, "step": 200 }, { "epoch": 0.5240174672489083, "grad_norm": 0.5793813415501323, "learning_rate": 5e-06, "loss": 0.5921, "step": 210 }, { "epoch": 0.5489706799750468, "grad_norm": 0.5540868891147879, "learning_rate": 5e-06, "loss": 0.5806, "step": 220 }, { "epoch": 0.5739238927011853, "grad_norm": 0.5338207990715906, "learning_rate": 5e-06, "loss": 0.5851, "step": 230 }, { "epoch": 0.5988771054273238, "grad_norm": 0.5622641527094587, "learning_rate": 5e-06, "loss": 0.5848, "step": 240 }, { "epoch": 0.6238303181534622, "grad_norm": 0.5384076976816572, "learning_rate": 5e-06, "loss": 0.5791, "step": 250 }, { "epoch": 0.6487835308796007, "grad_norm": 0.6033419963307188, "learning_rate": 5e-06, "loss": 0.5842, "step": 260 }, { "epoch": 0.6737367436057392, "grad_norm": 0.5342190421176904, "learning_rate": 5e-06, "loss": 0.5802, "step": 270 }, { "epoch": 0.6986899563318777, "grad_norm": 0.5282395994026974, "learning_rate": 5e-06, "loss": 0.5769, "step": 280 }, { "epoch": 0.7236431690580162, "grad_norm": 0.6039215573643518, "learning_rate": 5e-06, "loss": 0.5811, "step": 290 }, { "epoch": 0.7485963817841547, "grad_norm": 0.5625470869148316, "learning_rate": 5e-06, "loss": 0.5748, "step": 300 }, { "epoch": 0.7735495945102931, "grad_norm": 0.6938197396793522, "learning_rate": 5e-06, "loss": 0.576, "step": 310 }, { "epoch": 0.7985028072364317, "grad_norm": 0.5199860087817598, "learning_rate": 5e-06, "loss": 0.5733, "step": 320 }, { "epoch": 0.8234560199625702, "grad_norm": 0.5347102211468582, "learning_rate": 5e-06, "loss": 0.5713, "step": 330 }, { "epoch": 0.8484092326887087, "grad_norm": 0.49920752644836863, "learning_rate": 5e-06, "loss": 0.5747, "step": 340 }, { "epoch": 0.8733624454148472, "grad_norm": 0.5742360043218735, "learning_rate": 5e-06, "loss": 0.5672, "step": 350 }, { "epoch": 0.8983156581409857, "grad_norm": 0.6082105330996729, "learning_rate": 5e-06, "loss": 0.5648, "step": 360 }, { "epoch": 0.9232688708671242, "grad_norm": 0.5296253264006645, "learning_rate": 5e-06, "loss": 0.5701, "step": 370 }, { "epoch": 0.9482220835932627, "grad_norm": 0.5862019120409135, "learning_rate": 5e-06, "loss": 0.5693, "step": 380 }, { "epoch": 0.9731752963194011, "grad_norm": 0.5004204006192343, "learning_rate": 5e-06, "loss": 0.5667, "step": 390 }, { "epoch": 0.9981285090455396, "grad_norm": 0.5111639493787322, "learning_rate": 5e-06, "loss": 0.5574, "step": 400 }, { "epoch": 0.9981285090455396, "eval_loss": 0.5630170702934265, "eval_runtime": 136.6299, "eval_samples_per_second": 79.002, "eval_steps_per_second": 0.622, "step": 400 }, { "epoch": 1.023081721771678, "grad_norm": 0.5280851311810567, "learning_rate": 5e-06, "loss": 0.5296, "step": 410 }, { "epoch": 1.0480349344978166, "grad_norm": 0.8751970650854215, "learning_rate": 5e-06, "loss": 0.5324, "step": 420 }, { "epoch": 1.072988147223955, "grad_norm": 0.5756582314387694, "learning_rate": 5e-06, "loss": 0.5245, "step": 430 }, { "epoch": 1.0979413599500936, "grad_norm": 0.5619488808417233, "learning_rate": 5e-06, "loss": 0.5277, "step": 440 }, { "epoch": 1.122894572676232, "grad_norm": 0.8057161437263283, "learning_rate": 5e-06, "loss": 0.5255, "step": 450 }, { "epoch": 1.1478477854023705, "grad_norm": 0.581447349352516, "learning_rate": 5e-06, "loss": 0.5246, "step": 460 }, { "epoch": 1.172800998128509, "grad_norm": 0.5907089920281934, "learning_rate": 5e-06, "loss": 0.5308, "step": 470 }, { "epoch": 1.1977542108546475, "grad_norm": 0.527607335995899, "learning_rate": 5e-06, "loss": 0.5314, "step": 480 }, { "epoch": 1.222707423580786, "grad_norm": 0.6625045440230309, "learning_rate": 5e-06, "loss": 0.5271, "step": 490 }, { "epoch": 1.2476606363069245, "grad_norm": 0.49276396345156864, "learning_rate": 5e-06, "loss": 0.5307, "step": 500 }, { "epoch": 1.272613849033063, "grad_norm": 0.47964009848881595, "learning_rate": 5e-06, "loss": 0.525, "step": 510 }, { "epoch": 1.2975670617592014, "grad_norm": 0.6903279013904727, "learning_rate": 5e-06, "loss": 0.5173, "step": 520 }, { "epoch": 1.32252027448534, "grad_norm": 0.6235250926674746, "learning_rate": 5e-06, "loss": 0.5241, "step": 530 }, { "epoch": 1.3474734872114784, "grad_norm": 0.7115742016246728, "learning_rate": 5e-06, "loss": 0.5209, "step": 540 }, { "epoch": 1.372426699937617, "grad_norm": 0.5305573388431144, "learning_rate": 5e-06, "loss": 0.527, "step": 550 }, { "epoch": 1.3973799126637554, "grad_norm": 0.5804195692188839, "learning_rate": 5e-06, "loss": 0.5253, "step": 560 }, { "epoch": 1.4223331253898939, "grad_norm": 0.49733338644989056, "learning_rate": 5e-06, "loss": 0.523, "step": 570 }, { "epoch": 1.4472863381160324, "grad_norm": 0.5635813062598766, "learning_rate": 5e-06, "loss": 0.518, "step": 580 }, { "epoch": 1.472239550842171, "grad_norm": 0.5233795326173047, "learning_rate": 5e-06, "loss": 0.5184, "step": 590 }, { "epoch": 1.4971927635683095, "grad_norm": 0.5340519491855418, "learning_rate": 5e-06, "loss": 0.5199, "step": 600 }, { "epoch": 1.522145976294448, "grad_norm": 0.6382502305796666, "learning_rate": 5e-06, "loss": 0.5152, "step": 610 }, { "epoch": 1.5470991890205865, "grad_norm": 0.6114232456070381, "learning_rate": 5e-06, "loss": 0.5202, "step": 620 }, { "epoch": 1.572052401746725, "grad_norm": 0.5267522341441786, "learning_rate": 5e-06, "loss": 0.5194, "step": 630 }, { "epoch": 1.5970056144728635, "grad_norm": 0.49682168460425896, "learning_rate": 5e-06, "loss": 0.5235, "step": 640 }, { "epoch": 1.621958827199002, "grad_norm": 0.5492682660691606, "learning_rate": 5e-06, "loss": 0.5224, "step": 650 }, { "epoch": 1.6469120399251405, "grad_norm": 0.5160022677697064, "learning_rate": 5e-06, "loss": 0.5136, "step": 660 }, { "epoch": 1.671865252651279, "grad_norm": 0.5108996597287156, "learning_rate": 5e-06, "loss": 0.5143, "step": 670 }, { "epoch": 1.6968184653774174, "grad_norm": 0.5398052585347178, "learning_rate": 5e-06, "loss": 0.5142, "step": 680 }, { "epoch": 1.721771678103556, "grad_norm": 0.5061450862678053, "learning_rate": 5e-06, "loss": 0.5131, "step": 690 }, { "epoch": 1.7467248908296944, "grad_norm": 0.5157055149054185, "learning_rate": 5e-06, "loss": 0.508, "step": 700 }, { "epoch": 1.7716781035558329, "grad_norm": 0.5204430984353897, "learning_rate": 5e-06, "loss": 0.5175, "step": 710 }, { "epoch": 1.7966313162819714, "grad_norm": 0.5357849253293278, "learning_rate": 5e-06, "loss": 0.5168, "step": 720 }, { "epoch": 1.8215845290081099, "grad_norm": 0.7132216372119276, "learning_rate": 5e-06, "loss": 0.5192, "step": 730 }, { "epoch": 1.8465377417342483, "grad_norm": 0.6769653937286553, "learning_rate": 5e-06, "loss": 0.515, "step": 740 }, { "epoch": 1.8714909544603868, "grad_norm": 0.47758034322098186, "learning_rate": 5e-06, "loss": 0.5113, "step": 750 }, { "epoch": 1.8964441671865253, "grad_norm": 0.45410954217292865, "learning_rate": 5e-06, "loss": 0.508, "step": 760 }, { "epoch": 1.9213973799126638, "grad_norm": 0.5514628147384041, "learning_rate": 5e-06, "loss": 0.5111, "step": 770 }, { "epoch": 1.9463505926388023, "grad_norm": 0.5437873076115038, "learning_rate": 5e-06, "loss": 0.515, "step": 780 }, { "epoch": 1.9713038053649408, "grad_norm": 0.5047758942757364, "learning_rate": 5e-06, "loss": 0.509, "step": 790 }, { "epoch": 1.9962570180910792, "grad_norm": 0.4712219343056804, "learning_rate": 5e-06, "loss": 0.5138, "step": 800 }, { "epoch": 1.9987523393636932, "eval_loss": 0.5372687578201294, "eval_runtime": 138.155, "eval_samples_per_second": 78.13, "eval_steps_per_second": 0.615, "step": 801 }, { "epoch": 2.0212102308172177, "grad_norm": 0.5647461852845252, "learning_rate": 5e-06, "loss": 0.478, "step": 810 }, { "epoch": 2.046163443543356, "grad_norm": 0.6651732696313954, "learning_rate": 5e-06, "loss": 0.4739, "step": 820 }, { "epoch": 2.0711166562694947, "grad_norm": 0.6344890617063265, "learning_rate": 5e-06, "loss": 0.4742, "step": 830 }, { "epoch": 2.096069868995633, "grad_norm": 0.5551443463065495, "learning_rate": 5e-06, "loss": 0.4757, "step": 840 }, { "epoch": 2.1210230817217717, "grad_norm": 0.5821542029735364, "learning_rate": 5e-06, "loss": 0.4745, "step": 850 }, { "epoch": 2.14597629444791, "grad_norm": 0.5036890967140374, "learning_rate": 5e-06, "loss": 0.4741, "step": 860 }, { "epoch": 2.1709295071740486, "grad_norm": 0.526252717008255, "learning_rate": 5e-06, "loss": 0.4752, "step": 870 }, { "epoch": 2.195882719900187, "grad_norm": 0.5161295092624307, "learning_rate": 5e-06, "loss": 0.4708, "step": 880 }, { "epoch": 2.2208359326263256, "grad_norm": 0.5979999195074699, "learning_rate": 5e-06, "loss": 0.474, "step": 890 }, { "epoch": 2.245789145352464, "grad_norm": 0.548385025393527, "learning_rate": 5e-06, "loss": 0.4725, "step": 900 }, { "epoch": 2.2707423580786026, "grad_norm": 0.5786454120324753, "learning_rate": 5e-06, "loss": 0.4721, "step": 910 }, { "epoch": 2.295695570804741, "grad_norm": 0.49033248942630286, "learning_rate": 5e-06, "loss": 0.4754, "step": 920 }, { "epoch": 2.3206487835308796, "grad_norm": 0.578944877860593, "learning_rate": 5e-06, "loss": 0.4771, "step": 930 }, { "epoch": 2.345601996257018, "grad_norm": 0.5444827802066028, "learning_rate": 5e-06, "loss": 0.478, "step": 940 }, { "epoch": 2.3705552089831565, "grad_norm": 0.5738362287944658, "learning_rate": 5e-06, "loss": 0.4706, "step": 950 }, { "epoch": 2.395508421709295, "grad_norm": 0.5497412052181665, "learning_rate": 5e-06, "loss": 0.4742, "step": 960 }, { "epoch": 2.4204616344354335, "grad_norm": 0.6321664132907767, "learning_rate": 5e-06, "loss": 0.4745, "step": 970 }, { "epoch": 2.445414847161572, "grad_norm": 0.5155397917374155, "learning_rate": 5e-06, "loss": 0.4714, "step": 980 }, { "epoch": 2.4703680598877105, "grad_norm": 0.48000015580082134, "learning_rate": 5e-06, "loss": 0.4737, "step": 990 }, { "epoch": 2.495321272613849, "grad_norm": 0.5646370042066121, "learning_rate": 5e-06, "loss": 0.4737, "step": 1000 }, { "epoch": 2.5202744853399874, "grad_norm": 0.5488929805385689, "learning_rate": 5e-06, "loss": 0.479, "step": 1010 }, { "epoch": 2.545227698066126, "grad_norm": 0.5690240798527287, "learning_rate": 5e-06, "loss": 0.4735, "step": 1020 }, { "epoch": 2.5701809107922644, "grad_norm": 0.5241263467703002, "learning_rate": 5e-06, "loss": 0.4759, "step": 1030 }, { "epoch": 2.595134123518403, "grad_norm": 0.4966621491972494, "learning_rate": 5e-06, "loss": 0.4741, "step": 1040 }, { "epoch": 2.6200873362445414, "grad_norm": 0.6751968467864414, "learning_rate": 5e-06, "loss": 0.4798, "step": 1050 }, { "epoch": 2.64504054897068, "grad_norm": 0.6383918524471734, "learning_rate": 5e-06, "loss": 0.4754, "step": 1060 }, { "epoch": 2.6699937616968183, "grad_norm": 0.5256722819070684, "learning_rate": 5e-06, "loss": 0.4726, "step": 1070 }, { "epoch": 2.694946974422957, "grad_norm": 0.5129557694221921, "learning_rate": 5e-06, "loss": 0.4738, "step": 1080 }, { "epoch": 2.7199001871490953, "grad_norm": 0.49845285895513675, "learning_rate": 5e-06, "loss": 0.4788, "step": 1090 }, { "epoch": 2.744853399875234, "grad_norm": 0.6463893459607853, "learning_rate": 5e-06, "loss": 0.4772, "step": 1100 }, { "epoch": 2.7698066126013723, "grad_norm": 0.5326072861407999, "learning_rate": 5e-06, "loss": 0.4756, "step": 1110 }, { "epoch": 2.7947598253275108, "grad_norm": 0.5085270847721544, "learning_rate": 5e-06, "loss": 0.4741, "step": 1120 }, { "epoch": 2.8197130380536493, "grad_norm": 0.5495621538149744, "learning_rate": 5e-06, "loss": 0.4745, "step": 1130 }, { "epoch": 2.8446662507797877, "grad_norm": 0.48958337112517436, "learning_rate": 5e-06, "loss": 0.4715, "step": 1140 }, { "epoch": 2.8696194635059262, "grad_norm": 0.5857338163975798, "learning_rate": 5e-06, "loss": 0.4747, "step": 1150 }, { "epoch": 2.8945726762320647, "grad_norm": 0.48260180409447406, "learning_rate": 5e-06, "loss": 0.4715, "step": 1160 }, { "epoch": 2.919525888958203, "grad_norm": 0.5058713259660398, "learning_rate": 5e-06, "loss": 0.4722, "step": 1170 }, { "epoch": 2.944479101684342, "grad_norm": 0.544953115514363, "learning_rate": 5e-06, "loss": 0.4765, "step": 1180 }, { "epoch": 2.96943231441048, "grad_norm": 0.4680283080397261, "learning_rate": 5e-06, "loss": 0.4756, "step": 1190 }, { "epoch": 2.994385527136619, "grad_norm": 0.52273100015014, "learning_rate": 5e-06, "loss": 0.4712, "step": 1200 }, { "epoch": 2.994385527136619, "eval_loss": 0.5305783748626709, "eval_runtime": 137.7646, "eval_samples_per_second": 78.351, "eval_steps_per_second": 0.617, "step": 1200 }, { "epoch": 2.994385527136619, "step": 1200, "total_flos": 2009625935216640.0, "train_loss": 0.5358249799410502, "train_runtime": 20182.0753, "train_samples_per_second": 30.484, "train_steps_per_second": 0.059 } ], "logging_steps": 10, "max_steps": 1200, "num_input_tokens_seen": 0, "num_train_epochs": 3, "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": 2009625935216640.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }