| { | |
| "best_metric": 29.2694, | |
| "best_model_checkpoint": "./ko-en_mbartLarge_exp20p_linear/checkpoint-60000", | |
| "epoch": 8.81772827474185, | |
| "eval_steps": 4000, | |
| "global_step": 76000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.9079, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.7987, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.648, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5348, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.9927062667755864e-05, | |
| "loss": 1.4769, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.985412533551173e-05, | |
| "loss": 1.4511, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.978118800326759e-05, | |
| "loss": 1.4201, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.970825067102346e-05, | |
| "loss": 1.3977, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_bleu": 22.7153, | |
| "eval_gen_len": 18.7135, | |
| "eval_loss": 1.3720011711120605, | |
| "eval_runtime": 1153.7, | |
| "eval_samples_per_second": 14.941, | |
| "eval_steps_per_second": 0.934, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.963531333877932e-05, | |
| "loss": 1.3792, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.956237600653519e-05, | |
| "loss": 1.3552, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.948943867429105e-05, | |
| "loss": 1.3468, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.941650134204692e-05, | |
| "loss": 1.3468, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.934356400980278e-05, | |
| "loss": 1.3236, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.927062667755864e-05, | |
| "loss": 1.2924, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.9197689345314504e-05, | |
| "loss": 1.3005, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.912475201307037e-05, | |
| "loss": 1.2824, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_bleu": 24.8579, | |
| "eval_gen_len": 18.7821, | |
| "eval_loss": 1.2633070945739746, | |
| "eval_runtime": 1161.9505, | |
| "eval_samples_per_second": 14.835, | |
| "eval_steps_per_second": 0.928, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.905181468082623e-05, | |
| "loss": 1.2758, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.89788773485821e-05, | |
| "loss": 1.2295, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.890594001633796e-05, | |
| "loss": 1.2196, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.883300268409383e-05, | |
| "loss": 1.2043, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.876006535184969e-05, | |
| "loss": 1.2144, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.868712801960555e-05, | |
| "loss": 1.195, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.861419068736142e-05, | |
| "loss": 1.184, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.854125335511728e-05, | |
| "loss": 1.1989, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_bleu": 26.2533, | |
| "eval_gen_len": 18.7975, | |
| "eval_loss": 1.2069392204284668, | |
| "eval_runtime": 1167.3322, | |
| "eval_samples_per_second": 14.766, | |
| "eval_steps_per_second": 0.923, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.846831602287315e-05, | |
| "loss": 1.186, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.839537869062901e-05, | |
| "loss": 1.1785, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 4.832244135838488e-05, | |
| "loss": 1.1659, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.824950402614074e-05, | |
| "loss": 1.1624, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.817656669389661e-05, | |
| "loss": 1.1785, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 4.810362936165247e-05, | |
| "loss": 1.1465, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.803069202940834e-05, | |
| "loss": 1.1617, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 4.79577546971642e-05, | |
| "loss": 1.1534, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_bleu": 26.1503, | |
| "eval_gen_len": 19.2075, | |
| "eval_loss": 1.190694808959961, | |
| "eval_runtime": 1177.3192, | |
| "eval_samples_per_second": 14.641, | |
| "eval_steps_per_second": 0.916, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 4.788481736492006e-05, | |
| "loss": 1.1745, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 4.781188003267593e-05, | |
| "loss": 1.1663, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.773894270043179e-05, | |
| "loss": 1.1029, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 4.766600536818765e-05, | |
| "loss": 1.0464, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 4.759306803594352e-05, | |
| "loss": 1.0348, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 4.752013070369938e-05, | |
| "loss": 1.0228, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.744719337145524e-05, | |
| "loss": 1.0503, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.737425603921111e-05, | |
| "loss": 1.0245, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_bleu": 27.8764, | |
| "eval_gen_len": 18.6046, | |
| "eval_loss": 1.146370768547058, | |
| "eval_runtime": 1146.7204, | |
| "eval_samples_per_second": 15.032, | |
| "eval_steps_per_second": 0.94, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.730131870696697e-05, | |
| "loss": 1.0493, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.722838137472284e-05, | |
| "loss": 1.0287, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 4.71554440424787e-05, | |
| "loss": 1.03, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.708250671023457e-05, | |
| "loss": 1.0354, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.700956937799043e-05, | |
| "loss": 1.0268, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4.69366320457463e-05, | |
| "loss": 1.039, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 4.686369471350216e-05, | |
| "loss": 1.0343, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 4.679075738125803e-05, | |
| "loss": 1.0186, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_bleu": 28.4585, | |
| "eval_gen_len": 18.6731, | |
| "eval_loss": 1.1286325454711914, | |
| "eval_runtime": 1151.5333, | |
| "eval_samples_per_second": 14.969, | |
| "eval_steps_per_second": 0.936, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.671782004901389e-05, | |
| "loss": 1.0383, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.664488271676976e-05, | |
| "loss": 1.0488, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 4.657194538452562e-05, | |
| "loss": 1.0572, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 4.649900805228149e-05, | |
| "loss": 1.0168, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.642607072003735e-05, | |
| "loss": 0.9238, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.635313338779321e-05, | |
| "loss": 0.9281, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.628019605554907e-05, | |
| "loss": 0.9225, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 4.620725872330493e-05, | |
| "loss": 0.9245, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_bleu": 28.4834, | |
| "eval_gen_len": 18.5428, | |
| "eval_loss": 1.1263633966445923, | |
| "eval_runtime": 1147.1298, | |
| "eval_samples_per_second": 15.026, | |
| "eval_steps_per_second": 0.94, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.61343213910608e-05, | |
| "loss": 0.9145, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 4.606138405881666e-05, | |
| "loss": 0.9228, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 4.598844672657253e-05, | |
| "loss": 0.9196, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.591550939432839e-05, | |
| "loss": 0.9138, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 4.584257206208426e-05, | |
| "loss": 0.9274, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 4.576963472984012e-05, | |
| "loss": 0.9077, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 4.569669739759599e-05, | |
| "loss": 0.9312, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 4.562376006535185e-05, | |
| "loss": 0.9343, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_bleu": 28.8235, | |
| "eval_gen_len": 18.7833, | |
| "eval_loss": 1.1182423830032349, | |
| "eval_runtime": 1155.993, | |
| "eval_samples_per_second": 14.911, | |
| "eval_steps_per_second": 0.933, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 4.555082273310772e-05, | |
| "loss": 0.9232, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 4.547788540086358e-05, | |
| "loss": 0.9308, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 4.540494806861945e-05, | |
| "loss": 0.9361, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 4.533201073637531e-05, | |
| "loss": 0.9139, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.525907340413118e-05, | |
| "loss": 0.919, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 4.518613607188704e-05, | |
| "loss": 0.8226, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 4.5113198739642907e-05, | |
| "loss": 0.8242, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 4.504026140739877e-05, | |
| "loss": 0.8215, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_bleu": 28.6134, | |
| "eval_gen_len": 18.5656, | |
| "eval_loss": 1.133116602897644, | |
| "eval_runtime": 1144.5462, | |
| "eval_samples_per_second": 15.06, | |
| "eval_steps_per_second": 0.942, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.496732407515463e-05, | |
| "loss": 0.833, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 4.489438674291049e-05, | |
| "loss": 0.8175, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 4.482144941066635e-05, | |
| "loss": 0.8313, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 4.474851207842222e-05, | |
| "loss": 0.832, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 4.467557474617808e-05, | |
| "loss": 0.8397, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 4.460263741393395e-05, | |
| "loss": 0.8324, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 4.452970008168981e-05, | |
| "loss": 0.8379, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 4.445676274944568e-05, | |
| "loss": 0.8456, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "eval_bleu": 28.7324, | |
| "eval_gen_len": 18.459, | |
| "eval_loss": 1.1203099489212036, | |
| "eval_runtime": 1135.006, | |
| "eval_samples_per_second": 15.187, | |
| "eval_steps_per_second": 0.95, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 4.438382541720154e-05, | |
| "loss": 0.8481, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.431088808495741e-05, | |
| "loss": 0.8376, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 4.423795075271327e-05, | |
| "loss": 0.8416, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 4.416501342046914e-05, | |
| "loss": 0.8462, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 4.4092076088225e-05, | |
| "loss": 0.846, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 4.401913875598087e-05, | |
| "loss": 0.8429, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 4.394620142373673e-05, | |
| "loss": 0.7648, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 4.3873264091492596e-05, | |
| "loss": 0.7437, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "eval_bleu": 28.7297, | |
| "eval_gen_len": 18.7835, | |
| "eval_loss": 1.1458019018173218, | |
| "eval_runtime": 1166.3289, | |
| "eval_samples_per_second": 14.779, | |
| "eval_steps_per_second": 0.924, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.380032675924846e-05, | |
| "loss": 0.7562, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 4.3727389427004326e-05, | |
| "loss": 0.752, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 4.365445209476019e-05, | |
| "loss": 0.7526, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 4.358151476251605e-05, | |
| "loss": 0.7779, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 4.350857743027191e-05, | |
| "loss": 0.7942, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.343564009802778e-05, | |
| "loss": 0.7954, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 4.336270276578364e-05, | |
| "loss": 0.7874, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 4.32897654335395e-05, | |
| "loss": 0.7829, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "eval_bleu": 28.8328, | |
| "eval_gen_len": 18.6052, | |
| "eval_loss": 1.1367288827896118, | |
| "eval_runtime": 1145.791, | |
| "eval_samples_per_second": 15.044, | |
| "eval_steps_per_second": 0.941, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.63, | |
| "learning_rate": 4.321682810129537e-05, | |
| "loss": 0.7853, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 4.314389076905123e-05, | |
| "loss": 0.7916, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 4.30709534368071e-05, | |
| "loss": 0.7914, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 4.299801610456296e-05, | |
| "loss": 0.8096, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 4.292507877231883e-05, | |
| "loss": 0.8128, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 4.285214144007469e-05, | |
| "loss": 0.7987, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 4.2779204107830557e-05, | |
| "loss": 0.8185, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 4.270626677558642e-05, | |
| "loss": 0.7434, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_bleu": 28.2106, | |
| "eval_gen_len": 18.4871, | |
| "eval_loss": 1.1697230339050293, | |
| "eval_runtime": 1139.5757, | |
| "eval_samples_per_second": 15.126, | |
| "eval_steps_per_second": 0.946, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 4.2633329443342286e-05, | |
| "loss": 0.6938, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 4.256039211109815e-05, | |
| "loss": 0.695, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 4.2487454778854015e-05, | |
| "loss": 0.7016, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 4.241451744660988e-05, | |
| "loss": 0.6836, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 4.234158011436574e-05, | |
| "loss": 0.6807, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 4.2268642782121606e-05, | |
| "loss": 0.674, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 4.219570544987747e-05, | |
| "loss": 0.6897, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 4.2122768117633335e-05, | |
| "loss": 0.7153, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "eval_bleu": 28.1455, | |
| "eval_gen_len": 18.7413, | |
| "eval_loss": 1.1770974397659302, | |
| "eval_runtime": 1156.7175, | |
| "eval_samples_per_second": 14.902, | |
| "eval_steps_per_second": 0.932, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 4.20498307853892e-05, | |
| "loss": 0.7177, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 4.197689345314506e-05, | |
| "loss": 0.716, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 4.190395612090092e-05, | |
| "loss": 0.721, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 4.183101878865679e-05, | |
| "loss": 0.7152, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 4.175808145641265e-05, | |
| "loss": 0.7012, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 4.168514412416852e-05, | |
| "loss": 0.7069, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 4.161220679192438e-05, | |
| "loss": 0.7088, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 4.1539269459680246e-05, | |
| "loss": 0.6996, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_bleu": 29.2694, | |
| "eval_gen_len": 18.5162, | |
| "eval_loss": 1.1514174938201904, | |
| "eval_runtime": 1140.5069, | |
| "eval_samples_per_second": 15.113, | |
| "eval_steps_per_second": 0.945, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 4.146633212743611e-05, | |
| "loss": 0.6796, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 4.1393394795191976e-05, | |
| "loss": 0.6009, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 4.132045746294784e-05, | |
| "loss": 0.5985, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 4.1247520130703705e-05, | |
| "loss": 0.5966, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 4.1174582798459566e-05, | |
| "loss": 0.6132, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 4.110164546621543e-05, | |
| "loss": 0.6048, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 4.1028708133971296e-05, | |
| "loss": 0.6038, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 4.095577080172716e-05, | |
| "loss": 0.6336, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "eval_bleu": 28.1465, | |
| "eval_gen_len": 18.5439, | |
| "eval_loss": 1.2213271856307983, | |
| "eval_runtime": 1140.4863, | |
| "eval_samples_per_second": 15.114, | |
| "eval_steps_per_second": 0.945, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 4.0882833469483025e-05, | |
| "loss": 0.6943, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 4.0809896137238886e-05, | |
| "loss": 0.7508, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 4.0736958804994754e-05, | |
| "loss": 0.7606, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 4.0664021472750616e-05, | |
| "loss": 0.7578, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 4.059108414050648e-05, | |
| "loss": 0.751, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 4.051814680826234e-05, | |
| "loss": 0.738, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 4.0445209476018206e-05, | |
| "loss": 0.7185, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 4.037227214377407e-05, | |
| "loss": 0.7218, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "eval_bleu": 28.2245, | |
| "eval_gen_len": 18.5246, | |
| "eval_loss": 1.1834964752197266, | |
| "eval_runtime": 1143.8445, | |
| "eval_samples_per_second": 15.069, | |
| "eval_steps_per_second": 0.942, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 4.0299334811529936e-05, | |
| "loss": 0.723, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 4.02263974792858e-05, | |
| "loss": 0.7128, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 4.0153460147041665e-05, | |
| "loss": 0.5863, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 4.0080522814797527e-05, | |
| "loss": 0.5939, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 4.0007585482553395e-05, | |
| "loss": 0.6018, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 3.9934648150309256e-05, | |
| "loss": 0.588, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 3.986171081806512e-05, | |
| "loss": 0.5816, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 3.9788773485820985e-05, | |
| "loss": 0.5934, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "eval_bleu": 28.3836, | |
| "eval_gen_len": 18.6717, | |
| "eval_loss": 1.2387008666992188, | |
| "eval_runtime": 1156.4248, | |
| "eval_samples_per_second": 14.905, | |
| "eval_steps_per_second": 0.932, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 3.971583615357685e-05, | |
| "loss": 0.5772, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 3.9642898821332715e-05, | |
| "loss": 0.576, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 3.9569961489088576e-05, | |
| "loss": 0.5707, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 3.9497024156844444e-05, | |
| "loss": 0.5771, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 3.9424086824600305e-05, | |
| "loss": 0.5699, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 3.9351149492356174e-05, | |
| "loss": 0.5673, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 3.9278212160112035e-05, | |
| "loss": 0.5779, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 3.9205274827867896e-05, | |
| "loss": 0.5723, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "eval_bleu": 28.5925, | |
| "eval_gen_len": 18.5566, | |
| "eval_loss": 1.232297658920288, | |
| "eval_runtime": 1148.9097, | |
| "eval_samples_per_second": 15.003, | |
| "eval_steps_per_second": 0.938, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "step": 76000, | |
| "total_flos": 2.635368494115324e+18, | |
| "train_loss": 0.5245111798738178, | |
| "train_runtime": 53209.1351, | |
| "train_samples_per_second": 103.666, | |
| "train_steps_per_second": 6.479 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 344760, | |
| "num_train_epochs": 40, | |
| "save_steps": 4000, | |
| "total_flos": 2.635368494115324e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |