{ "best_metric": 0.8650239058477381, "best_model_checkpoint": "relevant_model/run-1/checkpoint-53415", "epoch": 3.0, "eval_steps": 500, "global_step": 53415, "is_hyper_param_search": true, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.03, "learning_rate": 4.101665655728474e-06, "loss": 0.5111, "step": 500 }, { "epoch": 0.06, "learning_rate": 4.078498947704027e-06, "loss": 0.4216, "step": 1000 }, { "epoch": 0.08, "learning_rate": 4.0553322396795785e-06, "loss": 0.4711, "step": 1500 }, { "epoch": 0.11, "learning_rate": 4.0321655316551306e-06, "loss": 0.4472, "step": 2000 }, { "epoch": 0.14, "learning_rate": 4.0089988236306835e-06, "loss": 0.4429, "step": 2500 }, { "epoch": 0.17, "learning_rate": 3.985832115606236e-06, "loss": 0.5083, "step": 3000 }, { "epoch": 0.2, "learning_rate": 3.962665407581788e-06, "loss": 0.469, "step": 3500 }, { "epoch": 0.22, "learning_rate": 3.93949869955734e-06, "loss": 0.4424, "step": 4000 }, { "epoch": 0.25, "learning_rate": 3.916331991532892e-06, "loss": 0.472, "step": 4500 }, { "epoch": 0.28, "learning_rate": 3.893165283508445e-06, "loss": 0.4344, "step": 5000 }, { "epoch": 0.31, "learning_rate": 3.869998575483997e-06, "loss": 0.4255, "step": 5500 }, { "epoch": 0.34, "learning_rate": 3.846831867459549e-06, "loss": 0.4293, "step": 6000 }, { "epoch": 0.37, "learning_rate": 3.823665159435101e-06, "loss": 0.4581, "step": 6500 }, { "epoch": 0.39, "learning_rate": 3.8004984514106537e-06, "loss": 0.4235, "step": 7000 }, { "epoch": 0.42, "learning_rate": 3.777331743386206e-06, "loss": 0.4334, "step": 7500 }, { "epoch": 0.45, "learning_rate": 3.7541650353617583e-06, "loss": 0.3703, "step": 8000 }, { "epoch": 0.48, "learning_rate": 3.7309983273373104e-06, "loss": 0.4415, "step": 8500 }, { "epoch": 0.51, "learning_rate": 3.7078316193128626e-06, "loss": 0.4234, "step": 9000 }, { "epoch": 0.53, "learning_rate": 3.684664911288415e-06, "loss": 0.4184, "step": 9500 }, { "epoch": 0.56, "learning_rate": 3.661498203263967e-06, "loss": 0.3912, "step": 10000 }, { "epoch": 0.59, "learning_rate": 3.6383314952395193e-06, "loss": 0.4158, "step": 10500 }, { "epoch": 0.62, "learning_rate": 3.615164787215072e-06, "loss": 0.3966, "step": 11000 }, { "epoch": 0.65, "learning_rate": 3.591998079190624e-06, "loss": 0.4171, "step": 11500 }, { "epoch": 0.67, "learning_rate": 3.568831371166176e-06, "loss": 0.381, "step": 12000 }, { "epoch": 0.7, "learning_rate": 3.5456646631417285e-06, "loss": 0.3867, "step": 12500 }, { "epoch": 0.73, "learning_rate": 3.5224979551172806e-06, "loss": 0.4438, "step": 13000 }, { "epoch": 0.76, "learning_rate": 3.4993312470928328e-06, "loss": 0.3806, "step": 13500 }, { "epoch": 0.79, "learning_rate": 3.4761645390683853e-06, "loss": 0.391, "step": 14000 }, { "epoch": 0.81, "learning_rate": 3.4529978310439374e-06, "loss": 0.3947, "step": 14500 }, { "epoch": 0.84, "learning_rate": 3.42983112301949e-06, "loss": 0.397, "step": 15000 }, { "epoch": 0.87, "learning_rate": 3.406664414995042e-06, "loss": 0.4048, "step": 15500 }, { "epoch": 0.9, "learning_rate": 3.383497706970594e-06, "loss": 0.4648, "step": 16000 }, { "epoch": 0.93, "learning_rate": 3.3603309989461466e-06, "loss": 0.3751, "step": 16500 }, { "epoch": 0.95, "learning_rate": 3.3371642909216987e-06, "loss": 0.3951, "step": 17000 }, { "epoch": 0.98, "learning_rate": 3.313997582897251e-06, "loss": 0.4386, "step": 17500 }, { "epoch": 1.0, "eval_f1": 0.8416666666666667, "eval_loss": 0.3926413059234619, "eval_runtime": 7.7212, "eval_samples_per_second": 512.357, "eval_steps_per_second": 8.03, "step": 17805 }, { "epoch": 1.01, "learning_rate": 3.2908308748728034e-06, "loss": 0.3639, "step": 18000 }, { "epoch": 1.04, "learning_rate": 3.2676641668483555e-06, "loss": 0.372, "step": 18500 }, { "epoch": 1.07, "learning_rate": 3.2444974588239076e-06, "loss": 0.3249, "step": 19000 }, { "epoch": 1.1, "learning_rate": 3.22133075079946e-06, "loss": 0.3586, "step": 19500 }, { "epoch": 1.12, "learning_rate": 3.1981640427750122e-06, "loss": 0.3595, "step": 20000 }, { "epoch": 1.15, "learning_rate": 3.1749973347505643e-06, "loss": 0.3577, "step": 20500 }, { "epoch": 1.18, "learning_rate": 3.151830626726117e-06, "loss": 0.3505, "step": 21000 }, { "epoch": 1.21, "learning_rate": 3.128663918701669e-06, "loss": 0.3448, "step": 21500 }, { "epoch": 1.24, "learning_rate": 3.105497210677221e-06, "loss": 0.3726, "step": 22000 }, { "epoch": 1.26, "learning_rate": 3.0823305026527736e-06, "loss": 0.3522, "step": 22500 }, { "epoch": 1.29, "learning_rate": 3.0591637946283257e-06, "loss": 0.3814, "step": 23000 }, { "epoch": 1.32, "learning_rate": 3.0359970866038778e-06, "loss": 0.3641, "step": 23500 }, { "epoch": 1.35, "learning_rate": 3.0128303785794303e-06, "loss": 0.343, "step": 24000 }, { "epoch": 1.38, "learning_rate": 2.9896636705549824e-06, "loss": 0.331, "step": 24500 }, { "epoch": 1.4, "learning_rate": 2.966496962530535e-06, "loss": 0.3737, "step": 25000 }, { "epoch": 1.43, "learning_rate": 2.943330254506087e-06, "loss": 0.3437, "step": 25500 }, { "epoch": 1.46, "learning_rate": 2.920163546481639e-06, "loss": 0.3648, "step": 26000 }, { "epoch": 1.49, "learning_rate": 2.8969968384571917e-06, "loss": 0.3421, "step": 26500 }, { "epoch": 1.52, "learning_rate": 2.8738301304327438e-06, "loss": 0.3344, "step": 27000 }, { "epoch": 1.54, "learning_rate": 2.850663422408296e-06, "loss": 0.3533, "step": 27500 }, { "epoch": 1.57, "learning_rate": 2.8274967143838484e-06, "loss": 0.371, "step": 28000 }, { "epoch": 1.6, "learning_rate": 2.8043300063594005e-06, "loss": 0.3526, "step": 28500 }, { "epoch": 1.63, "learning_rate": 2.7811632983349526e-06, "loss": 0.3539, "step": 29000 }, { "epoch": 1.66, "learning_rate": 2.757996590310505e-06, "loss": 0.3692, "step": 29500 }, { "epoch": 1.68, "learning_rate": 2.7348298822860572e-06, "loss": 0.3104, "step": 30000 }, { "epoch": 1.71, "learning_rate": 2.7116631742616094e-06, "loss": 0.3371, "step": 30500 }, { "epoch": 1.74, "learning_rate": 2.688496466237162e-06, "loss": 0.3818, "step": 31000 }, { "epoch": 1.77, "learning_rate": 2.665329758212714e-06, "loss": 0.384, "step": 31500 }, { "epoch": 1.8, "learning_rate": 2.6421630501882657e-06, "loss": 0.3249, "step": 32000 }, { "epoch": 1.83, "learning_rate": 2.6189963421638186e-06, "loss": 0.3385, "step": 32500 }, { "epoch": 1.85, "learning_rate": 2.5958296341393703e-06, "loss": 0.3837, "step": 33000 }, { "epoch": 1.88, "learning_rate": 2.5726629261149224e-06, "loss": 0.3224, "step": 33500 }, { "epoch": 1.91, "learning_rate": 2.549496218090475e-06, "loss": 0.3555, "step": 34000 }, { "epoch": 1.94, "learning_rate": 2.526329510066027e-06, "loss": 0.3555, "step": 34500 }, { "epoch": 1.97, "learning_rate": 2.50316280204158e-06, "loss": 0.3487, "step": 35000 }, { "epoch": 1.99, "learning_rate": 2.4799960940171317e-06, "loss": 0.3847, "step": 35500 }, { "epoch": 2.0, "eval_f1": 0.8587530820711519, "eval_loss": 0.37445908784866333, "eval_runtime": 7.6778, "eval_samples_per_second": 515.253, "eval_steps_per_second": 8.075, "step": 35610 }, { "epoch": 2.02, "learning_rate": 2.4568293859926838e-06, "loss": 0.3372, "step": 36000 }, { "epoch": 2.05, "learning_rate": 2.4336626779682363e-06, "loss": 0.2759, "step": 36500 }, { "epoch": 2.08, "learning_rate": 2.4104959699437884e-06, "loss": 0.2711, "step": 37000 }, { "epoch": 2.11, "learning_rate": 2.3873292619193405e-06, "loss": 0.2879, "step": 37500 }, { "epoch": 2.13, "learning_rate": 2.364162553894893e-06, "loss": 0.3026, "step": 38000 }, { "epoch": 2.16, "learning_rate": 2.340995845870445e-06, "loss": 0.3108, "step": 38500 }, { "epoch": 2.19, "learning_rate": 2.3178291378459972e-06, "loss": 0.2571, "step": 39000 }, { "epoch": 2.22, "learning_rate": 2.2946624298215498e-06, "loss": 0.3241, "step": 39500 }, { "epoch": 2.25, "learning_rate": 2.271495721797102e-06, "loss": 0.2841, "step": 40000 }, { "epoch": 2.27, "learning_rate": 2.248329013772654e-06, "loss": 0.2532, "step": 40500 }, { "epoch": 2.3, "learning_rate": 2.2251623057482065e-06, "loss": 0.3058, "step": 41000 }, { "epoch": 2.33, "learning_rate": 2.2019955977237586e-06, "loss": 0.3036, "step": 41500 }, { "epoch": 2.36, "learning_rate": 2.1788288896993107e-06, "loss": 0.3087, "step": 42000 }, { "epoch": 2.39, "learning_rate": 2.1556621816748632e-06, "loss": 0.2701, "step": 42500 }, { "epoch": 2.42, "learning_rate": 2.1324954736504153e-06, "loss": 0.3051, "step": 43000 }, { "epoch": 2.44, "learning_rate": 2.1093287656259674e-06, "loss": 0.3005, "step": 43500 }, { "epoch": 2.47, "learning_rate": 2.08616205760152e-06, "loss": 0.304, "step": 44000 }, { "epoch": 2.5, "learning_rate": 2.062995349577072e-06, "loss": 0.2895, "step": 44500 }, { "epoch": 2.53, "learning_rate": 2.0398286415526246e-06, "loss": 0.2984, "step": 45000 }, { "epoch": 2.56, "learning_rate": 2.0166619335281767e-06, "loss": 0.3043, "step": 45500 }, { "epoch": 2.58, "learning_rate": 1.993495225503729e-06, "loss": 0.2959, "step": 46000 }, { "epoch": 2.61, "learning_rate": 1.9703285174792813e-06, "loss": 0.2765, "step": 46500 }, { "epoch": 2.64, "learning_rate": 1.9471618094548334e-06, "loss": 0.3272, "step": 47000 }, { "epoch": 2.67, "learning_rate": 1.923995101430386e-06, "loss": 0.2409, "step": 47500 }, { "epoch": 2.7, "learning_rate": 1.900828393405938e-06, "loss": 0.3259, "step": 48000 }, { "epoch": 2.72, "learning_rate": 1.8776616853814904e-06, "loss": 0.2934, "step": 48500 }, { "epoch": 2.75, "learning_rate": 1.8544949773570427e-06, "loss": 0.2824, "step": 49000 }, { "epoch": 2.78, "learning_rate": 1.8313282693325946e-06, "loss": 0.2821, "step": 49500 }, { "epoch": 2.81, "learning_rate": 1.8081615613081469e-06, "loss": 0.2954, "step": 50000 }, { "epoch": 2.84, "learning_rate": 1.7849948532836992e-06, "loss": 0.3067, "step": 50500 }, { "epoch": 2.86, "learning_rate": 1.7618281452592513e-06, "loss": 0.3159, "step": 51000 }, { "epoch": 2.89, "learning_rate": 1.7386614372348036e-06, "loss": 0.3052, "step": 51500 }, { "epoch": 2.92, "learning_rate": 1.715494729210356e-06, "loss": 0.2831, "step": 52000 }, { "epoch": 2.95, "learning_rate": 1.6923280211859083e-06, "loss": 0.3139, "step": 52500 }, { "epoch": 2.98, "learning_rate": 1.6691613131614604e-06, "loss": 0.2793, "step": 53000 }, { "epoch": 3.0, "eval_f1": 0.8650239058477381, "eval_loss": 0.42799684405326843, "eval_runtime": 7.6499, "eval_samples_per_second": 517.128, "eval_steps_per_second": 8.105, "step": 53415 } ], "logging_steps": 500, "max_steps": 89025, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 500, "total_flos": 8347941521695260.0, "train_batch_size": 4, "trial_name": null, "trial_params": { "learning_rate": 4.124832363752922e-06, "num_train_epochs": 5, "per_device_train_batch_size": 4, "seed": 40 } }