| { | |
| "best_metric": 1.0988906621932983, | |
| "best_model_checkpoint": "./koen_mbartLarge_64p_run1/checkpoint-60000", | |
| "epoch": 3.64716302818736, | |
| "eval_steps": 2500, | |
| "global_step": 70000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9395, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.9869403959671946e-05, | |
| "loss": 1.7758, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.973880791934389e-05, | |
| "loss": 1.6618, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.9608211879015834e-05, | |
| "loss": 1.5894, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.947761583868777e-05, | |
| "loss": 1.5543, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_bleu": 25.2368, | |
| "eval_gen_len": 18.5097, | |
| "eval_loss": 1.5067745447158813, | |
| "eval_runtime": 2599.8469, | |
| "eval_samples_per_second": 14.764, | |
| "eval_steps_per_second": 0.923, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.9347019798359716e-05, | |
| "loss": 1.5387, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.921642375803166e-05, | |
| "loss": 1.5048, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.9085827717703604e-05, | |
| "loss": 1.4639, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.895523167737555e-05, | |
| "loss": 1.4439, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.882463563704749e-05, | |
| "loss": 1.4399, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_bleu": 27.0554, | |
| "eval_gen_len": 18.5539, | |
| "eval_loss": 1.397201418876648, | |
| "eval_runtime": 2613.4705, | |
| "eval_samples_per_second": 14.687, | |
| "eval_steps_per_second": 0.918, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.869403959671943e-05, | |
| "loss": 1.4247, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.856344355639137e-05, | |
| "loss": 1.4094, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.843284751606332e-05, | |
| "loss": 1.385, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.830225147573526e-05, | |
| "loss": 1.3793, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.81716554354072e-05, | |
| "loss": 1.3448, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_bleu": 28.8579, | |
| "eval_gen_len": 18.6315, | |
| "eval_loss": 1.313157558441162, | |
| "eval_runtime": 2624.1392, | |
| "eval_samples_per_second": 14.628, | |
| "eval_steps_per_second": 0.915, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.804105939507914e-05, | |
| "loss": 1.3364, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.7910463354751087e-05, | |
| "loss": 1.3282, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.7779867314423024e-05, | |
| "loss": 1.3243, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.764927127409497e-05, | |
| "loss": 1.3314, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.751867523376691e-05, | |
| "loss": 1.3205, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_bleu": 29.5611, | |
| "eval_gen_len": 18.7781, | |
| "eval_loss": 1.287328839302063, | |
| "eval_runtime": 2675.7352, | |
| "eval_samples_per_second": 14.346, | |
| "eval_steps_per_second": 0.897, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.7388079193438856e-05, | |
| "loss": 1.3111, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.72574831531108e-05, | |
| "loss": 1.3197, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.7126887112782744e-05, | |
| "loss": 1.315, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.699629107245468e-05, | |
| "loss": 1.2968, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.6865695032126625e-05, | |
| "loss": 1.2786, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_bleu": 30.3042, | |
| "eval_gen_len": 18.5644, | |
| "eval_loss": 1.2398545742034912, | |
| "eval_runtime": 2641.4423, | |
| "eval_samples_per_second": 14.532, | |
| "eval_steps_per_second": 0.909, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.673509899179857e-05, | |
| "loss": 1.2809, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.6604502951470513e-05, | |
| "loss": 1.2734, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.647390691114246e-05, | |
| "loss": 1.2721, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.63433108708144e-05, | |
| "loss": 1.2644, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.6212714830486345e-05, | |
| "loss": 1.2561, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_bleu": 30.5801, | |
| "eval_gen_len": 19.0186, | |
| "eval_loss": 1.2173043489456177, | |
| "eval_runtime": 2780.297, | |
| "eval_samples_per_second": 13.806, | |
| "eval_steps_per_second": 0.863, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.608211879015828e-05, | |
| "loss": 1.2521, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.595152274983023e-05, | |
| "loss": 1.2316, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.582092670950217e-05, | |
| "loss": 1.2483, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.5690330669174115e-05, | |
| "loss": 1.2446, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.555973462884606e-05, | |
| "loss": 1.2479, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_bleu": 30.8896, | |
| "eval_gen_len": 18.7636, | |
| "eval_loss": 1.2125173807144165, | |
| "eval_runtime": 2723.8664, | |
| "eval_samples_per_second": 14.092, | |
| "eval_steps_per_second": 0.881, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.5429138588518e-05, | |
| "loss": 1.2305, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.529854254818994e-05, | |
| "loss": 1.2481, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.5167946507861884e-05, | |
| "loss": 1.2257, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.503735046753383e-05, | |
| "loss": 1.2247, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.490675442720577e-05, | |
| "loss": 1.1891, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_bleu": 31.9834, | |
| "eval_gen_len": 18.7002, | |
| "eval_loss": 1.1775543689727783, | |
| "eval_runtime": 2647.526, | |
| "eval_samples_per_second": 14.498, | |
| "eval_steps_per_second": 0.907, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.4776158386877716e-05, | |
| "loss": 1.1961, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.464556234654966e-05, | |
| "loss": 1.185, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.45149663062216e-05, | |
| "loss": 1.1783, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.4384370265893535e-05, | |
| "loss": 1.1756, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.425377422556548e-05, | |
| "loss": 1.1943, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_bleu": 32.0205, | |
| "eval_gen_len": 18.7054, | |
| "eval_loss": 1.1650614738464355, | |
| "eval_runtime": 2679.0855, | |
| "eval_samples_per_second": 14.328, | |
| "eval_steps_per_second": 0.896, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.412317818523742e-05, | |
| "loss": 1.1649, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.399258214490937e-05, | |
| "loss": 1.166, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.386198610458131e-05, | |
| "loss": 1.1654, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.3731390064253255e-05, | |
| "loss": 1.1459, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.360079402392519e-05, | |
| "loss": 1.1375, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_bleu": 32.3658, | |
| "eval_gen_len": 18.6287, | |
| "eval_loss": 1.1492213010787964, | |
| "eval_runtime": 2667.8187, | |
| "eval_samples_per_second": 14.388, | |
| "eval_steps_per_second": 0.9, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.3470197983597136e-05, | |
| "loss": 1.1259, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 4.333960194326908e-05, | |
| "loss": 1.125, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.3209005902941024e-05, | |
| "loss": 1.1122, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 4.307840986261297e-05, | |
| "loss": 1.1284, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.294781382228491e-05, | |
| "loss": 1.1351, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_bleu": 32.339, | |
| "eval_gen_len": 18.7655, | |
| "eval_loss": 1.1459516286849976, | |
| "eval_runtime": 2734.8416, | |
| "eval_samples_per_second": 14.036, | |
| "eval_steps_per_second": 0.878, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 4.2817217781956856e-05, | |
| "loss": 1.1207, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 4.2686621741628794e-05, | |
| "loss": 1.1537, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.255602570130074e-05, | |
| "loss": 1.1415, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.242542966097268e-05, | |
| "loss": 1.0911, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.2294833620644626e-05, | |
| "loss": 1.0859, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_bleu": 31.5418, | |
| "eval_gen_len": 19.016, | |
| "eval_loss": 1.1623187065124512, | |
| "eval_runtime": 2768.8989, | |
| "eval_samples_per_second": 13.863, | |
| "eval_steps_per_second": 0.867, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.216423758031657e-05, | |
| "loss": 1.0833, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.2033641539988514e-05, | |
| "loss": 1.0713, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 4.190304549966045e-05, | |
| "loss": 1.105, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.1772449459332395e-05, | |
| "loss": 1.0965, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.164185341900434e-05, | |
| "loss": 1.0373, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_bleu": 32.672, | |
| "eval_gen_len": 18.7224, | |
| "eval_loss": 1.1382720470428467, | |
| "eval_runtime": 2697.4501, | |
| "eval_samples_per_second": 14.23, | |
| "eval_steps_per_second": 0.89, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.151125737867628e-05, | |
| "loss": 1.0403, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.138066133834823e-05, | |
| "loss": 1.0286, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 4.125006529802017e-05, | |
| "loss": 1.0442, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.111946925769211e-05, | |
| "loss": 1.0892, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.0988873217364046e-05, | |
| "loss": 1.0824, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_bleu": 33.2231, | |
| "eval_gen_len": 18.6697, | |
| "eval_loss": 1.1231993436813354, | |
| "eval_runtime": 2652.0512, | |
| "eval_samples_per_second": 14.474, | |
| "eval_steps_per_second": 0.905, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.085827717703599e-05, | |
| "loss": 1.0766, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 4.0727681136707934e-05, | |
| "loss": 1.0753, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 4.059708509637988e-05, | |
| "loss": 1.0256, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4.046648905605182e-05, | |
| "loss": 1.0166, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.0335893015723766e-05, | |
| "loss": 1.0242, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_bleu": 32.813, | |
| "eval_gen_len": 18.2553, | |
| "eval_loss": 1.1312956809997559, | |
| "eval_runtime": 2568.2472, | |
| "eval_samples_per_second": 14.946, | |
| "eval_steps_per_second": 0.934, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.02052969753957e-05, | |
| "loss": 1.0382, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.007470093506765e-05, | |
| "loss": 1.0839, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 3.994410489473959e-05, | |
| "loss": 1.0665, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.9813508854411535e-05, | |
| "loss": 1.0682, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.968291281408348e-05, | |
| "loss": 1.0649, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_bleu": 33.2021, | |
| "eval_gen_len": 18.7216, | |
| "eval_loss": 1.1182446479797363, | |
| "eval_runtime": 2685.9538, | |
| "eval_samples_per_second": 14.291, | |
| "eval_steps_per_second": 0.894, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.9552316773755423e-05, | |
| "loss": 1.0566, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.942172073342737e-05, | |
| "loss": 1.0567, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.9291124693099305e-05, | |
| "loss": 1.0616, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 3.916052865277125e-05, | |
| "loss": 1.0676, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.902993261244319e-05, | |
| "loss": 1.054, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_bleu": 33.0588, | |
| "eval_gen_len": 18.4992, | |
| "eval_loss": 1.132915735244751, | |
| "eval_runtime": 2616.3663, | |
| "eval_samples_per_second": 14.671, | |
| "eval_steps_per_second": 0.917, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 3.889933657211514e-05, | |
| "loss": 1.0592, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3.876874053178708e-05, | |
| "loss": 1.0559, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 3.8638144491459025e-05, | |
| "loss": 1.0332, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 3.850754845113096e-05, | |
| "loss": 1.0527, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.8376952410802906e-05, | |
| "loss": 1.0143, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_bleu": 33.2176, | |
| "eval_gen_len": 18.7156, | |
| "eval_loss": 1.1187158823013306, | |
| "eval_runtime": 2644.7329, | |
| "eval_samples_per_second": 14.514, | |
| "eval_steps_per_second": 0.907, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.824635637047485e-05, | |
| "loss": 1.0222, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 3.8115760330146794e-05, | |
| "loss": 1.0081, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 3.798516428981874e-05, | |
| "loss": 1.026, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.785456824949068e-05, | |
| "loss": 1.0138, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.772397220916262e-05, | |
| "loss": 1.0037, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_bleu": 33.3754, | |
| "eval_gen_len": 18.6443, | |
| "eval_loss": 1.1162101030349731, | |
| "eval_runtime": 2647.6533, | |
| "eval_samples_per_second": 14.498, | |
| "eval_steps_per_second": 0.906, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.7593376168834564e-05, | |
| "loss": 0.9654, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3.74627801285065e-05, | |
| "loss": 0.9404, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.7332184088178445e-05, | |
| "loss": 0.9193, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.720158804785039e-05, | |
| "loss": 0.9634, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.707099200752233e-05, | |
| "loss": 0.9928, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_bleu": 33.0727, | |
| "eval_gen_len": 18.6361, | |
| "eval_loss": 1.1306148767471313, | |
| "eval_runtime": 2648.6411, | |
| "eval_samples_per_second": 14.492, | |
| "eval_steps_per_second": 0.906, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 3.694039596719428e-05, | |
| "loss": 1.0123, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.6809799926866214e-05, | |
| "loss": 1.0289, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 3.667920388653816e-05, | |
| "loss": 0.9778, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 3.65486078462101e-05, | |
| "loss": 0.9642, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 3.6418011805882046e-05, | |
| "loss": 0.9497, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_bleu": 33.227, | |
| "eval_gen_len": 18.7638, | |
| "eval_loss": 1.1169679164886475, | |
| "eval_runtime": 2645.8519, | |
| "eval_samples_per_second": 14.508, | |
| "eval_steps_per_second": 0.907, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3.628741576555399e-05, | |
| "loss": 0.9503, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 3.6156819725225934e-05, | |
| "loss": 1.0029, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.602622368489788e-05, | |
| "loss": 1.0363, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 3.5895627644569816e-05, | |
| "loss": 1.0253, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 3.576503160424176e-05, | |
| "loss": 1.0157, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "eval_bleu": 33.685, | |
| "eval_gen_len": 18.5847, | |
| "eval_loss": 1.1071757078170776, | |
| "eval_runtime": 2620.1785, | |
| "eval_samples_per_second": 14.65, | |
| "eval_steps_per_second": 0.916, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.5634435563913704e-05, | |
| "loss": 0.9519, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 3.550383952358565e-05, | |
| "loss": 0.9401, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 3.537324348325759e-05, | |
| "loss": 0.9409, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 3.5242647442929536e-05, | |
| "loss": 0.9713, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.511205140260147e-05, | |
| "loss": 0.9876, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 33.6971, | |
| "eval_gen_len": 18.6873, | |
| "eval_loss": 1.103514313697815, | |
| "eval_runtime": 2615.1168, | |
| "eval_samples_per_second": 14.678, | |
| "eval_steps_per_second": 0.918, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 3.498145536227342e-05, | |
| "loss": 0.9875, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 3.485085932194536e-05, | |
| "loss": 0.9743, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 3.4720263281617305e-05, | |
| "loss": 0.9802, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.458966724128925e-05, | |
| "loss": 0.9829, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 3.445907120096119e-05, | |
| "loss": 0.9665, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_bleu": 33.8919, | |
| "eval_gen_len": 18.5258, | |
| "eval_loss": 1.0988906621932983, | |
| "eval_runtime": 2615.0744, | |
| "eval_samples_per_second": 14.678, | |
| "eval_steps_per_second": 0.918, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 3.432847516063313e-05, | |
| "loss": 0.9692, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 3.4197879120305075e-05, | |
| "loss": 0.9464, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 3.406728307997702e-05, | |
| "loss": 0.9004, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.393668703964896e-05, | |
| "loss": 0.9117, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 3.38060909993209e-05, | |
| "loss": 0.9197, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_bleu": 33.7036, | |
| "eval_gen_len": 18.5407, | |
| "eval_loss": 1.1059657335281372, | |
| "eval_runtime": 2661.0999, | |
| "eval_samples_per_second": 14.424, | |
| "eval_steps_per_second": 0.902, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 3.3675494958992844e-05, | |
| "loss": 0.9258, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 3.354489891866479e-05, | |
| "loss": 0.9684, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.3414302878336725e-05, | |
| "loss": 0.9429, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.328370683800867e-05, | |
| "loss": 0.9474, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.315311079768061e-05, | |
| "loss": 0.9427, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "eval_bleu": 33.7642, | |
| "eval_gen_len": 18.7, | |
| "eval_loss": 1.099454402923584, | |
| "eval_runtime": 2639.8151, | |
| "eval_samples_per_second": 14.541, | |
| "eval_steps_per_second": 0.909, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 3.302251475735256e-05, | |
| "loss": 0.9826, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 3.28919187170245e-05, | |
| "loss": 0.9501, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.2761322676696445e-05, | |
| "loss": 0.9358, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 3.263072663636839e-05, | |
| "loss": 0.912, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 3.250013059604033e-05, | |
| "loss": 0.8993, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_bleu": 33.1757, | |
| "eval_gen_len": 18.646, | |
| "eval_loss": 1.1364279985427856, | |
| "eval_runtime": 2620.3614, | |
| "eval_samples_per_second": 14.649, | |
| "eval_steps_per_second": 0.916, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 3.236953455571227e-05, | |
| "loss": 0.8973, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 3.2238938515384215e-05, | |
| "loss": 0.8767, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 3.210834247505616e-05, | |
| "loss": 0.8621, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 3.19777464347281e-05, | |
| "loss": 0.8715, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 3.184715039440005e-05, | |
| "loss": 0.8957, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "eval_bleu": 33.0954, | |
| "eval_gen_len": 18.3129, | |
| "eval_loss": 1.12513267993927, | |
| "eval_runtime": 2574.1588, | |
| "eval_samples_per_second": 14.912, | |
| "eval_steps_per_second": 0.932, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "step": 70000, | |
| "total_flos": 2.4272893620487455e+18, | |
| "train_loss": 1.1237463548932756, | |
| "train_runtime": 125442.5935, | |
| "train_samples_per_second": 24.48, | |
| "train_steps_per_second": 1.53 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 191930, | |
| "num_train_epochs": 10, | |
| "save_steps": 2500, | |
| "total_flos": 2.4272893620487455e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |