{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0, "eval_steps": 500, "global_step": 73151, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.01, "learning_rate": 4.965824117236949e-05, "loss": 5.6804, "step": 500 }, { "epoch": 0.01, "learning_rate": 4.9316482344738966e-05, "loss": 4.9782, "step": 1000 }, { "epoch": 0.02, "learning_rate": 4.897472351710845e-05, "loss": 4.8339, "step": 1500 }, { "epoch": 0.03, "learning_rate": 4.8632964689477936e-05, "loss": 4.6915, "step": 2000 }, { "epoch": 0.03, "learning_rate": 4.829188937950268e-05, "loss": 4.5389, "step": 2500 }, { "epoch": 0.04, "learning_rate": 4.7950130551872156e-05, "loss": 4.362, "step": 3000 }, { "epoch": 0.05, "learning_rate": 4.760837172424164e-05, "loss": 4.3116, "step": 3500 }, { "epoch": 0.05, "learning_rate": 4.7266612896611126e-05, "loss": 4.245, "step": 4000 }, { "epoch": 0.06, "learning_rate": 4.6924854068980604e-05, "loss": 4.2283, "step": 4500 }, { "epoch": 0.07, "learning_rate": 4.658309524135009e-05, "loss": 4.1251, "step": 5000 }, { "epoch": 0.08, "learning_rate": 4.624133641371957e-05, "loss": 4.0663, "step": 5500 }, { "epoch": 0.08, "learning_rate": 4.5899577586089046e-05, "loss": 4.0828, "step": 6000 }, { "epoch": 0.09, "learning_rate": 4.555918579376906e-05, "loss": 3.9893, "step": 6500 }, { "epoch": 0.1, "learning_rate": 4.5217426966138536e-05, "loss": 3.9494, "step": 7000 }, { "epoch": 0.1, "learning_rate": 4.487566813850802e-05, "loss": 3.8261, "step": 7500 }, { "epoch": 0.11, "learning_rate": 4.4533909310877506e-05, "loss": 3.8834, "step": 8000 }, { "epoch": 0.12, "learning_rate": 4.4192150483246984e-05, "loss": 3.8186, "step": 8500 }, { "epoch": 0.12, "learning_rate": 4.3851075173271725e-05, "loss": 3.791, "step": 9000 }, { "epoch": 0.13, "learning_rate": 4.350931634564121e-05, "loss": 3.7413, "step": 9500 }, { "epoch": 0.14, "learning_rate": 4.3167557518010696e-05, "loss": 3.631, "step": 10000 }, { "epoch": 0.14, "learning_rate": 4.2825798690380174e-05, "loss": 3.6658, "step": 10500 }, { "epoch": 0.15, "learning_rate": 4.248403986274966e-05, "loss": 3.6128, "step": 11000 }, { "epoch": 0.16, "learning_rate": 4.2142281035119144e-05, "loss": 3.5832, "step": 11500 }, { "epoch": 0.16, "learning_rate": 4.180052220748862e-05, "loss": 3.4998, "step": 12000 }, { "epoch": 0.17, "learning_rate": 4.14587633798581e-05, "loss": 3.5122, "step": 12500 }, { "epoch": 0.18, "learning_rate": 4.111768806988285e-05, "loss": 3.5041, "step": 13000 }, { "epoch": 0.18, "learning_rate": 4.0775929242252334e-05, "loss": 3.4191, "step": 13500 }, { "epoch": 0.19, "learning_rate": 4.0434853932277075e-05, "loss": 3.4023, "step": 14000 }, { "epoch": 0.2, "learning_rate": 4.0093095104646554e-05, "loss": 3.3738, "step": 14500 }, { "epoch": 0.21, "learning_rate": 3.975133627701604e-05, "loss": 3.3359, "step": 15000 }, { "epoch": 0.21, "learning_rate": 3.9409577449385524e-05, "loss": 3.2606, "step": 15500 }, { "epoch": 0.22, "learning_rate": 3.9068502139410265e-05, "loss": 3.3322, "step": 16000 }, { "epoch": 0.23, "learning_rate": 3.8726743311779744e-05, "loss": 3.2278, "step": 16500 }, { "epoch": 0.23, "learning_rate": 3.838498448414923e-05, "loss": 3.2608, "step": 17000 }, { "epoch": 0.24, "learning_rate": 3.8043225656518714e-05, "loss": 3.2201, "step": 17500 }, { "epoch": 0.25, "learning_rate": 3.770146682888819e-05, "loss": 3.2005, "step": 18000 }, { "epoch": 0.25, "learning_rate": 3.7360391518912934e-05, "loss": 3.1464, "step": 18500 }, { "epoch": 0.26, "learning_rate": 3.701863269128242e-05, "loss": 3.2039, "step": 19000 }, { "epoch": 0.27, "learning_rate": 3.6676873863651904e-05, "loss": 3.1524, "step": 19500 }, { "epoch": 0.27, "learning_rate": 3.633511503602138e-05, "loss": 3.1729, "step": 20000 }, { "epoch": 0.28, "learning_rate": 3.599335620839087e-05, "loss": 3.1225, "step": 20500 }, { "epoch": 0.29, "learning_rate": 3.5651597380760345e-05, "loss": 3.1449, "step": 21000 }, { "epoch": 0.29, "learning_rate": 3.530983855312983e-05, "loss": 3.1289, "step": 21500 }, { "epoch": 0.3, "learning_rate": 3.496807972549931e-05, "loss": 3.0806, "step": 22000 }, { "epoch": 0.31, "learning_rate": 3.462700441552406e-05, "loss": 3.0845, "step": 22500 }, { "epoch": 0.31, "learning_rate": 3.4285245587893535e-05, "loss": 3.12, "step": 23000 }, { "epoch": 0.32, "learning_rate": 3.394348676026302e-05, "loss": 3.0579, "step": 23500 }, { "epoch": 0.33, "learning_rate": 3.3601727932632506e-05, "loss": 3.0443, "step": 24000 }, { "epoch": 0.33, "learning_rate": 3.326065262265725e-05, "loss": 3.0636, "step": 24500 }, { "epoch": 0.34, "learning_rate": 3.2918893795026725e-05, "loss": 3.0179, "step": 25000 }, { "epoch": 0.35, "learning_rate": 3.257713496739621e-05, "loss": 2.9552, "step": 25500 }, { "epoch": 0.36, "learning_rate": 3.2235376139765695e-05, "loss": 3.0536, "step": 26000 }, { "epoch": 0.36, "learning_rate": 3.1893617312135174e-05, "loss": 2.911, "step": 26500 }, { "epoch": 0.37, "learning_rate": 3.1552542002159915e-05, "loss": 3.0095, "step": 27000 }, { "epoch": 0.38, "learning_rate": 3.12107831745294e-05, "loss": 2.9568, "step": 27500 }, { "epoch": 0.38, "learning_rate": 3.0869024346898885e-05, "loss": 2.9344, "step": 28000 }, { "epoch": 0.39, "learning_rate": 3.0527265519268364e-05, "loss": 2.9417, "step": 28500 }, { "epoch": 0.4, "learning_rate": 3.018619020929311e-05, "loss": 2.9011, "step": 29000 }, { "epoch": 0.4, "learning_rate": 2.984443138166259e-05, "loss": 2.8799, "step": 29500 }, { "epoch": 0.41, "learning_rate": 2.9502672554032075e-05, "loss": 2.9197, "step": 30000 }, { "epoch": 0.42, "learning_rate": 2.9160913726401557e-05, "loss": 2.8963, "step": 30500 }, { "epoch": 0.42, "learning_rate": 2.881915489877104e-05, "loss": 2.914, "step": 31000 }, { "epoch": 0.43, "learning_rate": 2.847807958879578e-05, "loss": 2.8748, "step": 31500 }, { "epoch": 0.44, "learning_rate": 2.8136320761165265e-05, "loss": 2.9159, "step": 32000 }, { "epoch": 0.44, "learning_rate": 2.7794561933534747e-05, "loss": 2.8798, "step": 32500 }, { "epoch": 0.45, "learning_rate": 2.745280310590423e-05, "loss": 2.8773, "step": 33000 }, { "epoch": 0.46, "learning_rate": 2.711104427827371e-05, "loss": 2.8688, "step": 33500 }, { "epoch": 0.46, "learning_rate": 2.676928545064319e-05, "loss": 2.8408, "step": 34000 }, { "epoch": 0.47, "learning_rate": 2.642752662301267e-05, "loss": 2.8473, "step": 34500 }, { "epoch": 0.48, "learning_rate": 2.6085767795382155e-05, "loss": 2.8303, "step": 35000 }, { "epoch": 0.49, "learning_rate": 2.57446924854069e-05, "loss": 2.7961, "step": 35500 }, { "epoch": 0.49, "learning_rate": 2.5402933657776385e-05, "loss": 2.8852, "step": 36000 }, { "epoch": 0.5, "learning_rate": 2.5061858347801127e-05, "loss": 2.819, "step": 36500 }, { "epoch": 0.51, "learning_rate": 2.4720099520170605e-05, "loss": 2.8543, "step": 37000 }, { "epoch": 0.51, "learning_rate": 2.437834069254009e-05, "loss": 2.8135, "step": 37500 }, { "epoch": 0.52, "learning_rate": 2.4036581864909572e-05, "loss": 2.8237, "step": 38000 }, { "epoch": 0.53, "learning_rate": 2.3695506554934317e-05, "loss": 2.7813, "step": 38500 }, { "epoch": 0.53, "learning_rate": 2.3353747727303798e-05, "loss": 2.7788, "step": 39000 }, { "epoch": 0.54, "learning_rate": 2.301198889967328e-05, "loss": 2.7682, "step": 39500 }, { "epoch": 0.55, "learning_rate": 2.267023007204276e-05, "loss": 2.7843, "step": 40000 }, { "epoch": 0.55, "learning_rate": 2.2328471244412243e-05, "loss": 2.7897, "step": 40500 }, { "epoch": 0.56, "learning_rate": 2.1987395934436988e-05, "loss": 2.7146, "step": 41000 }, { "epoch": 0.57, "learning_rate": 2.164563710680647e-05, "loss": 2.7909, "step": 41500 }, { "epoch": 0.57, "learning_rate": 2.130387827917595e-05, "loss": 2.7408, "step": 42000 }, { "epoch": 0.58, "learning_rate": 2.0962119451545433e-05, "loss": 2.7703, "step": 42500 }, { "epoch": 0.59, "learning_rate": 2.0620360623914915e-05, "loss": 2.7564, "step": 43000 }, { "epoch": 0.59, "learning_rate": 2.027928531393966e-05, "loss": 2.7858, "step": 43500 }, { "epoch": 0.6, "learning_rate": 1.993752648630914e-05, "loss": 2.7562, "step": 44000 }, { "epoch": 0.61, "learning_rate": 1.9595767658678623e-05, "loss": 2.7151, "step": 44500 }, { "epoch": 0.62, "learning_rate": 1.9254008831048105e-05, "loss": 2.7155, "step": 45000 }, { "epoch": 0.62, "learning_rate": 1.891225000341759e-05, "loss": 2.7111, "step": 45500 }, { "epoch": 0.63, "learning_rate": 1.857049117578707e-05, "loss": 2.7161, "step": 46000 }, { "epoch": 0.64, "learning_rate": 1.8228732348156553e-05, "loss": 2.7353, "step": 46500 }, { "epoch": 0.64, "learning_rate": 1.7886973520526035e-05, "loss": 2.671, "step": 47000 }, { "epoch": 0.65, "learning_rate": 1.754589821055078e-05, "loss": 2.708, "step": 47500 }, { "epoch": 0.66, "learning_rate": 1.720413938292026e-05, "loss": 2.7514, "step": 48000 }, { "epoch": 0.66, "learning_rate": 1.6862380555289743e-05, "loss": 2.6464, "step": 48500 }, { "epoch": 0.67, "learning_rate": 1.6521305245314485e-05, "loss": 2.7079, "step": 49000 }, { "epoch": 0.68, "learning_rate": 1.617954641768397e-05, "loss": 2.6883, "step": 49500 }, { "epoch": 0.68, "learning_rate": 1.583778759005345e-05, "loss": 2.686, "step": 50000 }, { "epoch": 0.69, "learning_rate": 1.5496028762422933e-05, "loss": 2.6826, "step": 50500 }, { "epoch": 0.7, "learning_rate": 1.5154269934792417e-05, "loss": 2.6855, "step": 51000 }, { "epoch": 0.7, "learning_rate": 1.48125111071619e-05, "loss": 2.607, "step": 51500 }, { "epoch": 0.71, "learning_rate": 1.447075227953138e-05, "loss": 2.7357, "step": 52000 }, { "epoch": 0.72, "learning_rate": 1.4128993451900863e-05, "loss": 2.644, "step": 52500 }, { "epoch": 0.72, "learning_rate": 1.3787918141925606e-05, "loss": 2.6631, "step": 53000 }, { "epoch": 0.73, "learning_rate": 1.344615931429509e-05, "loss": 2.6176, "step": 53500 }, { "epoch": 0.74, "learning_rate": 1.3104400486664573e-05, "loss": 2.6017, "step": 54000 }, { "epoch": 0.75, "learning_rate": 1.2762641659034053e-05, "loss": 2.615, "step": 54500 }, { "epoch": 0.75, "learning_rate": 1.2420882831403535e-05, "loss": 2.6685, "step": 55000 }, { "epoch": 0.76, "learning_rate": 1.2079124003773018e-05, "loss": 2.6363, "step": 55500 }, { "epoch": 0.77, "learning_rate": 1.17373651761425e-05, "loss": 2.6344, "step": 56000 }, { "epoch": 0.77, "learning_rate": 1.1396289866167243e-05, "loss": 2.6118, "step": 56500 }, { "epoch": 0.78, "learning_rate": 1.1054531038536727e-05, "loss": 2.617, "step": 57000 }, { "epoch": 0.79, "learning_rate": 1.0712772210906208e-05, "loss": 2.6345, "step": 57500 }, { "epoch": 0.79, "learning_rate": 1.037101338327569e-05, "loss": 2.6346, "step": 58000 }, { "epoch": 0.8, "learning_rate": 1.0029938073300433e-05, "loss": 2.6163, "step": 58500 }, { "epoch": 0.81, "learning_rate": 9.688179245669916e-06, "loss": 2.629, "step": 59000 }, { "epoch": 0.81, "learning_rate": 9.346420418039398e-06, "loss": 2.6725, "step": 59500 }, { "epoch": 0.82, "learning_rate": 9.00466159040888e-06, "loss": 2.6297, "step": 60000 }, { "epoch": 0.83, "learning_rate": 8.662902762778363e-06, "loss": 2.6274, "step": 60500 }, { "epoch": 0.83, "learning_rate": 8.321143935147847e-06, "loss": 2.6258, "step": 61000 }, { "epoch": 0.84, "learning_rate": 7.979385107517327e-06, "loss": 2.6075, "step": 61500 }, { "epoch": 0.85, "learning_rate": 7.63762627988681e-06, "loss": 2.5569, "step": 62000 }, { "epoch": 0.85, "learning_rate": 7.296550969911553e-06, "loss": 2.606, "step": 62500 }, { "epoch": 0.86, "learning_rate": 6.954792142281036e-06, "loss": 2.6188, "step": 63000 }, { "epoch": 0.87, "learning_rate": 6.6130333146505175e-06, "loss": 2.6118, "step": 63500 }, { "epoch": 0.87, "learning_rate": 6.27127448702e-06, "loss": 2.5898, "step": 64000 }, { "epoch": 0.88, "learning_rate": 5.929515659389482e-06, "loss": 2.6512, "step": 64500 }, { "epoch": 0.89, "learning_rate": 5.588440349414226e-06, "loss": 2.552, "step": 65000 }, { "epoch": 0.9, "learning_rate": 5.247365039438969e-06, "loss": 2.5514, "step": 65500 }, { "epoch": 0.9, "learning_rate": 4.905606211808451e-06, "loss": 2.5717, "step": 66000 }, { "epoch": 0.91, "learning_rate": 4.563847384177934e-06, "loss": 2.5928, "step": 66500 }, { "epoch": 0.92, "learning_rate": 4.2220885565474156e-06, "loss": 2.6304, "step": 67000 }, { "epoch": 0.92, "learning_rate": 3.880329728916898e-06, "loss": 2.6121, "step": 67500 }, { "epoch": 0.93, "learning_rate": 3.539254418941641e-06, "loss": 2.495, "step": 68000 }, { "epoch": 0.94, "learning_rate": 3.197495591311124e-06, "loss": 2.5219, "step": 68500 }, { "epoch": 0.94, "learning_rate": 2.855736763680606e-06, "loss": 2.6132, "step": 69000 }, { "epoch": 0.95, "learning_rate": 2.513977936050088e-06, "loss": 2.6146, "step": 69500 }, { "epoch": 0.96, "learning_rate": 2.1722191084195706e-06, "loss": 2.5907, "step": 70000 }, { "epoch": 0.96, "learning_rate": 1.830460280789053e-06, "loss": 2.4962, "step": 70500 }, { "epoch": 0.97, "learning_rate": 1.4887014531585353e-06, "loss": 2.6012, "step": 71000 }, { "epoch": 0.98, "learning_rate": 1.1469426255280174e-06, "loss": 2.5367, "step": 71500 }, { "epoch": 0.98, "learning_rate": 8.051837978974998e-07, "loss": 2.6087, "step": 72000 }, { "epoch": 0.99, "learning_rate": 4.641084879222431e-07, "loss": 2.5672, "step": 72500 }, { "epoch": 1.0, "learning_rate": 1.2234966029172534e-07, "loss": 2.5733, "step": 73000 }, { "epoch": 1.0, "step": 73151, "total_flos": 1.2183202602103603e+17, "train_loss": 3.037818258402621, "train_runtime": 6822.2882, "train_samples_per_second": 10.722, "train_steps_per_second": 10.722 } ], "logging_steps": 500, "max_steps": 73151, "num_train_epochs": 1, "save_steps": -73151, "total_flos": 1.2183202602103603e+17, "trial_name": null, "trial_params": null }