| { | |
| "best_metric": 0.8102121949195862, | |
| "best_model_checkpoint": "tmp/checkpoint-75000", | |
| "epoch": 1.9959017484099317, | |
| "eval_steps": 5000, | |
| "global_step": 150000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.9833674854299174e-05, | |
| "loss": 0.8584, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.9667349708598347e-05, | |
| "loss": 0.7979, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.950102456289752e-05, | |
| "loss": 0.8067, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.933469941719669e-05, | |
| "loss": 0.7832, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.916837427149586e-05, | |
| "loss": 0.7659, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.9002049125795035e-05, | |
| "loss": 0.7545, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.8835723980094214e-05, | |
| "loss": 0.7754, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.866939883439338e-05, | |
| "loss": 0.7307, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.850307368869255e-05, | |
| "loss": 0.7553, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.833674854299173e-05, | |
| "loss": 0.7398, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_bleu": 42.8862, | |
| "eval_chrf++": 64.0757, | |
| "eval_gen_len": 24.4369, | |
| "eval_loss": 0.925624430179596, | |
| "eval_runtime": 4935.4955, | |
| "eval_samples_per_second": 4.716, | |
| "eval_spbleu": 50.6556, | |
| "eval_steps_per_second": 0.943, | |
| "eval_ter": 45.4715, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.8170423397290895e-05, | |
| "loss": 0.7439, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.800409825159007e-05, | |
| "loss": 0.74, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.7837773105889246e-05, | |
| "loss": 0.7484, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.767144796018841e-05, | |
| "loss": 0.7585, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.750512281448759e-05, | |
| "loss": 0.7432, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.733879766878676e-05, | |
| "loss": 0.7337, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.717247252308593e-05, | |
| "loss": 0.7241, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.700614737738511e-05, | |
| "loss": 0.7357, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.683982223168428e-05, | |
| "loss": 0.709, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.6673497085983444e-05, | |
| "loss": 0.7343, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_bleu": 43.3821, | |
| "eval_chrf++": 64.4124, | |
| "eval_gen_len": 24.2454, | |
| "eval_loss": 0.9000950455665588, | |
| "eval_runtime": 4884.4167, | |
| "eval_samples_per_second": 4.765, | |
| "eval_spbleu": 51.0312, | |
| "eval_steps_per_second": 0.953, | |
| "eval_ter": 44.8341, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.650717194028262e-05, | |
| "loss": 0.7183, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.6340846794581795e-05, | |
| "loss": 0.7382, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.617452164888097e-05, | |
| "loss": 0.7187, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.600819650318014e-05, | |
| "loss": 0.7253, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.584187135747931e-05, | |
| "loss": 0.7165, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.5675546211778483e-05, | |
| "loss": 0.7136, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.5509221066077656e-05, | |
| "loss": 0.7334, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.534289592037683e-05, | |
| "loss": 0.7191, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.5176570774676e-05, | |
| "loss": 0.7443, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.501024562897517e-05, | |
| "loss": 0.7663, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bleu": 44.1569, | |
| "eval_chrf++": 64.9134, | |
| "eval_gen_len": 24.2657, | |
| "eval_loss": 0.8873258233070374, | |
| "eval_runtime": 4925.5075, | |
| "eval_samples_per_second": 4.725, | |
| "eval_spbleu": 51.7797, | |
| "eval_steps_per_second": 0.945, | |
| "eval_ter": 44.382, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.4843920483274344e-05, | |
| "loss": 0.7273, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.4677595337573516e-05, | |
| "loss": 0.7407, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.451127019187269e-05, | |
| "loss": 0.7115, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.434494504617187e-05, | |
| "loss": 0.7209, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.417861990047103e-05, | |
| "loss": 0.6983, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.4012294754770204e-05, | |
| "loss": 0.7341, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.384596960906938e-05, | |
| "loss": 0.7535, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.3679644463368555e-05, | |
| "loss": 0.7407, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.351331931766772e-05, | |
| "loss": 0.7338, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.33469941719669e-05, | |
| "loss": 0.7182, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_bleu": 44.4461, | |
| "eval_chrf++": 65.0814, | |
| "eval_gen_len": 24.2502, | |
| "eval_loss": 0.8806242346763611, | |
| "eval_runtime": 4916.0397, | |
| "eval_samples_per_second": 4.734, | |
| "eval_spbleu": 52.0701, | |
| "eval_steps_per_second": 0.947, | |
| "eval_ter": 44.0663, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.318066902626607e-05, | |
| "loss": 0.7131, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.301434388056524e-05, | |
| "loss": 0.6982, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.2848018734864416e-05, | |
| "loss": 0.7372, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.268169358916359e-05, | |
| "loss": 0.7102, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.251536844346276e-05, | |
| "loss": 0.6934, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.234904329776193e-05, | |
| "loss": 0.7227, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.2182718152061104e-05, | |
| "loss": 0.7017, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.2016393006360276e-05, | |
| "loss": 0.7204, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.185006786065945e-05, | |
| "loss": 0.7258, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.168374271495862e-05, | |
| "loss": 0.6895, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_bleu": 44.7245, | |
| "eval_chrf++": 65.33, | |
| "eval_gen_len": 24.2462, | |
| "eval_loss": 0.8754296898841858, | |
| "eval_runtime": 4904.4105, | |
| "eval_samples_per_second": 4.746, | |
| "eval_spbleu": 52.3392, | |
| "eval_steps_per_second": 0.949, | |
| "eval_ter": 43.8723, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.151741756925779e-05, | |
| "loss": 0.6957, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.1351092423556965e-05, | |
| "loss": 0.7138, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.1184767277856137e-05, | |
| "loss": 0.6806, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.101844213215531e-05, | |
| "loss": 0.7534, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.085211698645448e-05, | |
| "loss": 0.7019, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.068579184075366e-05, | |
| "loss": 0.717, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.0519466695052825e-05, | |
| "loss": 0.726, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.0353141549352e-05, | |
| "loss": 0.7133, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.0186816403651176e-05, | |
| "loss": 0.7015, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.002049125795034e-05, | |
| "loss": 0.7129, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_bleu": 45.3956, | |
| "eval_chrf++": 65.8044, | |
| "eval_gen_len": 24.3021, | |
| "eval_loss": 0.8627288937568665, | |
| "eval_runtime": 4927.5215, | |
| "eval_samples_per_second": 4.723, | |
| "eval_spbleu": 53.0583, | |
| "eval_steps_per_second": 0.945, | |
| "eval_ter": 43.2655, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.985416611224951e-05, | |
| "loss": 0.7059, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.968784096654869e-05, | |
| "loss": 0.6986, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.952151582084786e-05, | |
| "loss": 0.6817, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.9355190675147036e-05, | |
| "loss": 0.7082, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.918886552944621e-05, | |
| "loss": 0.7019, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.9022540383745374e-05, | |
| "loss": 0.6874, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.885621523804455e-05, | |
| "loss": 0.7153, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.8689890092343725e-05, | |
| "loss": 0.687, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.852356494664289e-05, | |
| "loss": 0.6805, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.835723980094207e-05, | |
| "loss": 0.7136, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_bleu": 45.0925, | |
| "eval_chrf++": 65.7049, | |
| "eval_gen_len": 24.512, | |
| "eval_loss": 0.8570441603660583, | |
| "eval_runtime": 4955.0167, | |
| "eval_samples_per_second": 4.697, | |
| "eval_spbleu": 52.6941, | |
| "eval_steps_per_second": 0.939, | |
| "eval_ter": 43.8842, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.819091465524124e-05, | |
| "loss": 0.7114, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.802458950954041e-05, | |
| "loss": 0.7042, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.7858264363839585e-05, | |
| "loss": 0.6823, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.769193921813876e-05, | |
| "loss": 0.6834, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3.752561407243793e-05, | |
| "loss": 0.6915, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.73592889267371e-05, | |
| "loss": 0.667, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.7192963781036273e-05, | |
| "loss": 0.6738, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.7026638635335446e-05, | |
| "loss": 0.6913, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.686031348963462e-05, | |
| "loss": 0.7139, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.669398834393379e-05, | |
| "loss": 0.7136, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_bleu": 45.7491, | |
| "eval_chrf++": 65.9888, | |
| "eval_gen_len": 24.3784, | |
| "eval_loss": 0.8480991125106812, | |
| "eval_runtime": 4954.9176, | |
| "eval_samples_per_second": 4.697, | |
| "eval_spbleu": 53.3676, | |
| "eval_steps_per_second": 0.939, | |
| "eval_ter": 43.1147, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.652766319823296e-05, | |
| "loss": 0.7109, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 3.6361338052532134e-05, | |
| "loss": 0.7257, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 3.619501290683131e-05, | |
| "loss": 0.7068, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.602868776113048e-05, | |
| "loss": 0.6881, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.586236261542965e-05, | |
| "loss": 0.6844, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.569603746972883e-05, | |
| "loss": 0.6768, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.5529712324027994e-05, | |
| "loss": 0.6934, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.5363387178327166e-05, | |
| "loss": 0.6804, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.5197062032626345e-05, | |
| "loss": 0.6842, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3.503073688692551e-05, | |
| "loss": 0.6946, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_bleu": 45.9172, | |
| "eval_chrf++": 66.1754, | |
| "eval_gen_len": 24.2755, | |
| "eval_loss": 0.8429935574531555, | |
| "eval_runtime": 4908.9662, | |
| "eval_samples_per_second": 4.741, | |
| "eval_spbleu": 53.5024, | |
| "eval_steps_per_second": 0.948, | |
| "eval_ter": 42.6775, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.486441174122469e-05, | |
| "loss": 0.6827, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.469808659552386e-05, | |
| "loss": 0.6812, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.453176144982303e-05, | |
| "loss": 0.6804, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 3.4365436304122206e-05, | |
| "loss": 0.7064, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 3.419911115842138e-05, | |
| "loss": 0.6562, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.403278601272055e-05, | |
| "loss": 0.6815, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3.386646086701972e-05, | |
| "loss": 0.6666, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3.3700135721318894e-05, | |
| "loss": 0.6797, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 3.3533810575618066e-05, | |
| "loss": 0.6696, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.336748542991724e-05, | |
| "loss": 0.6948, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_bleu": 46.2973, | |
| "eval_chrf++": 66.3697, | |
| "eval_gen_len": 24.2583, | |
| "eval_loss": 0.8373845219612122, | |
| "eval_runtime": 4895.6247, | |
| "eval_samples_per_second": 4.754, | |
| "eval_spbleu": 53.8466, | |
| "eval_steps_per_second": 0.951, | |
| "eval_ter": 42.478, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.320116028421641e-05, | |
| "loss": 0.6612, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3.303483513851558e-05, | |
| "loss": 0.671, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.2868509992814755e-05, | |
| "loss": 0.6626, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 3.270218484711393e-05, | |
| "loss": 0.682, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.25358597014131e-05, | |
| "loss": 0.6708, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 3.236953455571227e-05, | |
| "loss": 0.6599, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 3.220320941001144e-05, | |
| "loss": 0.7175, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.2036884264310615e-05, | |
| "loss": 0.675, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3.187055911860979e-05, | |
| "loss": 0.6816, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3.1704233972908966e-05, | |
| "loss": 0.6603, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_bleu": 46.623, | |
| "eval_chrf++": 66.6376, | |
| "eval_gen_len": 24.3667, | |
| "eval_loss": 0.8319695591926575, | |
| "eval_runtime": 4912.1959, | |
| "eval_samples_per_second": 4.738, | |
| "eval_spbleu": 54.0695, | |
| "eval_steps_per_second": 0.948, | |
| "eval_ter": 42.3781, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.153790882720814e-05, | |
| "loss": 0.6798, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3.13715836815073e-05, | |
| "loss": 0.6689, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3.120525853580648e-05, | |
| "loss": 0.6468, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.1038933390105654e-05, | |
| "loss": 0.6755, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.087260824440482e-05, | |
| "loss": 0.6492, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 3.0706283098704e-05, | |
| "loss": 0.6908, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 3.053995795300317e-05, | |
| "loss": 0.6672, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.037363280730234e-05, | |
| "loss": 0.7015, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.0207307661601515e-05, | |
| "loss": 0.6814, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.0040982515900683e-05, | |
| "loss": 0.656, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_bleu": 46.474, | |
| "eval_chrf++": 66.5341, | |
| "eval_gen_len": 24.398, | |
| "eval_loss": 0.8285811543464661, | |
| "eval_runtime": 4904.2201, | |
| "eval_samples_per_second": 4.746, | |
| "eval_spbleu": 54.0574, | |
| "eval_steps_per_second": 0.949, | |
| "eval_ter": 42.4323, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.987465737019986e-05, | |
| "loss": 0.6686, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.970833222449903e-05, | |
| "loss": 0.6617, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2.95420070787982e-05, | |
| "loss": 0.704, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2.9375681933097375e-05, | |
| "loss": 0.665, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.9209356787396547e-05, | |
| "loss": 0.6817, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.9043031641695716e-05, | |
| "loss": 0.6651, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.887670649599489e-05, | |
| "loss": 0.6384, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 2.8710381350294063e-05, | |
| "loss": 0.6934, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.854405620459324e-05, | |
| "loss": 0.6751, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.837773105889241e-05, | |
| "loss": 0.6586, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_bleu": 47.0019, | |
| "eval_chrf++": 66.8331, | |
| "eval_gen_len": 24.383, | |
| "eval_loss": 0.82272869348526, | |
| "eval_runtime": 4917.3831, | |
| "eval_samples_per_second": 4.733, | |
| "eval_spbleu": 54.5205, | |
| "eval_steps_per_second": 0.947, | |
| "eval_ter": 42.1062, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.821140591319158e-05, | |
| "loss": 0.655, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.8045080767490755e-05, | |
| "loss": 0.6751, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.7878755621789927e-05, | |
| "loss": 0.6978, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2.7712430476089096e-05, | |
| "loss": 0.6421, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.754610533038827e-05, | |
| "loss": 0.649, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.7379780184687444e-05, | |
| "loss": 0.6663, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 2.7213455038986612e-05, | |
| "loss": 0.6511, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.7047129893285788e-05, | |
| "loss": 0.6416, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.688080474758496e-05, | |
| "loss": 0.6747, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 2.6714479601884135e-05, | |
| "loss": 0.6594, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_bleu": 47.2356, | |
| "eval_chrf++": 67.0583, | |
| "eval_gen_len": 24.3131, | |
| "eval_loss": 0.8117721080780029, | |
| "eval_runtime": 4915.7413, | |
| "eval_samples_per_second": 4.735, | |
| "eval_spbleu": 54.6764, | |
| "eval_steps_per_second": 0.947, | |
| "eval_ter": 41.9177, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.6548154456183304e-05, | |
| "loss": 0.6709, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.6381829310482476e-05, | |
| "loss": 0.6606, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.621550416478165e-05, | |
| "loss": 0.6702, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.604917901908082e-05, | |
| "loss": 0.7017, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.5882853873379992e-05, | |
| "loss": 0.6466, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.5716528727679168e-05, | |
| "loss": 0.6268, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.5550203581978337e-05, | |
| "loss": 0.647, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.5383878436277515e-05, | |
| "loss": 0.6453, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.5217553290576684e-05, | |
| "loss": 0.6561, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.5051228144875856e-05, | |
| "loss": 0.6776, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 47.4319, | |
| "eval_chrf++": 67.1911, | |
| "eval_gen_len": 24.3139, | |
| "eval_loss": 0.8102121949195862, | |
| "eval_runtime": 4952.6673, | |
| "eval_samples_per_second": 4.699, | |
| "eval_spbleu": 54.9312, | |
| "eval_steps_per_second": 0.94, | |
| "eval_ter": 41.7875, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.4884902999175028e-05, | |
| "loss": 0.5579, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.47185778534742e-05, | |
| "loss": 0.483, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 2.4552252707773372e-05, | |
| "loss": 0.4868, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 2.4385927562072548e-05, | |
| "loss": 0.4575, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 2.4219602416371717e-05, | |
| "loss": 0.4634, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.405327727067089e-05, | |
| "loss": 0.4869, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.3886952124970064e-05, | |
| "loss": 0.4637, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 2.3720626979269233e-05, | |
| "loss": 0.4837, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.355430183356841e-05, | |
| "loss": 0.4763, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.338797668786758e-05, | |
| "loss": 0.4961, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_bleu": 47.3102, | |
| "eval_chrf++": 67.0563, | |
| "eval_gen_len": 24.3704, | |
| "eval_loss": 0.8673813343048096, | |
| "eval_runtime": 4986.641, | |
| "eval_samples_per_second": 4.667, | |
| "eval_spbleu": 54.8203, | |
| "eval_steps_per_second": 0.933, | |
| "eval_ter": 41.9193, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.3221651542166753e-05, | |
| "loss": 0.4783, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 2.3055326396465925e-05, | |
| "loss": 0.4864, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 2.2889001250765097e-05, | |
| "loss": 0.4725, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.272267610506427e-05, | |
| "loss": 0.4788, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.255635095936344e-05, | |
| "loss": 0.4957, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.2390025813662613e-05, | |
| "loss": 0.4969, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 2.2223700667961785e-05, | |
| "loss": 0.4715, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 2.205737552226096e-05, | |
| "loss": 0.4712, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 2.1891050376560133e-05, | |
| "loss": 0.4777, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 2.17247252308593e-05, | |
| "loss": 0.4749, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_bleu": 47.5408, | |
| "eval_chrf++": 67.2416, | |
| "eval_gen_len": 24.4305, | |
| "eval_loss": 0.8608251810073853, | |
| "eval_runtime": 4964.0014, | |
| "eval_samples_per_second": 4.689, | |
| "eval_spbleu": 55.0306, | |
| "eval_steps_per_second": 0.938, | |
| "eval_ter": 41.7098, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 2.1558400085158477e-05, | |
| "loss": 0.4771, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 2.139207493945765e-05, | |
| "loss": 0.473, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 2.122574979375682e-05, | |
| "loss": 0.4701, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.1059424648055993e-05, | |
| "loss": 0.4857, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.0893099502355165e-05, | |
| "loss": 0.4728, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 2.0726774356654337e-05, | |
| "loss": 0.4784, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.056044921095351e-05, | |
| "loss": 0.4623, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.039412406525268e-05, | |
| "loss": 0.4721, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 2.0227798919551854e-05, | |
| "loss": 0.4864, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.006147377385103e-05, | |
| "loss": 0.482, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_bleu": 47.7026, | |
| "eval_chrf++": 67.3263, | |
| "eval_gen_len": 24.3967, | |
| "eval_loss": 0.8620461225509644, | |
| "eval_runtime": 4933.9377, | |
| "eval_samples_per_second": 4.717, | |
| "eval_spbleu": 55.1523, | |
| "eval_steps_per_second": 0.943, | |
| "eval_ter": 41.7466, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.9895148628150198e-05, | |
| "loss": 0.4648, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1.972882348244937e-05, | |
| "loss": 0.4821, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.9562498336748545e-05, | |
| "loss": 0.4843, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.9396173191047717e-05, | |
| "loss": 0.4815, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.9229848045346886e-05, | |
| "loss": 0.4686, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.906352289964606e-05, | |
| "loss": 0.5067, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.8897197753945234e-05, | |
| "loss": 0.485, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.8730872608244406e-05, | |
| "loss": 0.4756, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.8564547462543578e-05, | |
| "loss": 0.4643, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.839822231684275e-05, | |
| "loss": 0.4752, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_bleu": 47.7488, | |
| "eval_chrf++": 67.3486, | |
| "eval_gen_len": 24.3134, | |
| "eval_loss": 0.8601581454277039, | |
| "eval_runtime": 4927.554, | |
| "eval_samples_per_second": 4.723, | |
| "eval_spbleu": 55.1946, | |
| "eval_steps_per_second": 0.945, | |
| "eval_ter": 41.5271, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.8231897171141922e-05, | |
| "loss": 0.4739, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.8065572025441097e-05, | |
| "loss": 0.4867, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.7899246879740266e-05, | |
| "loss": 0.4813, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.7732921734039438e-05, | |
| "loss": 0.4602, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.7566596588338614e-05, | |
| "loss": 0.4728, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.7400271442637782e-05, | |
| "loss": 0.4909, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.7233946296936958e-05, | |
| "loss": 0.4719, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.706762115123613e-05, | |
| "loss": 0.4806, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.6901296005535302e-05, | |
| "loss": 0.5015, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.6734970859834474e-05, | |
| "loss": 0.4804, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_bleu": 47.8177, | |
| "eval_chrf++": 67.3657, | |
| "eval_gen_len": 24.2798, | |
| "eval_loss": 0.8537631630897522, | |
| "eval_runtime": 4934.3311, | |
| "eval_samples_per_second": 4.717, | |
| "eval_spbleu": 55.1623, | |
| "eval_steps_per_second": 0.943, | |
| "eval_ter": 41.4437, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.6568645714133646e-05, | |
| "loss": 0.4657, | |
| "step": 100500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.6402320568432818e-05, | |
| "loss": 0.4614, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.6235995422731994e-05, | |
| "loss": 0.4718, | |
| "step": 101500 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.6069670277031162e-05, | |
| "loss": 0.4632, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.5903345131330335e-05, | |
| "loss": 0.4781, | |
| "step": 102500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.573701998562951e-05, | |
| "loss": 0.4657, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.5570694839928682e-05, | |
| "loss": 0.4723, | |
| "step": 103500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.540436969422785e-05, | |
| "loss": 0.4761, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.5238044548527025e-05, | |
| "loss": 0.4813, | |
| "step": 104500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.5071719402826198e-05, | |
| "loss": 0.4701, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_bleu": 48.0326, | |
| "eval_chrf++": 67.5497, | |
| "eval_gen_len": 24.3708, | |
| "eval_loss": 0.8546710014343262, | |
| "eval_runtime": 4962.677, | |
| "eval_samples_per_second": 4.69, | |
| "eval_spbleu": 55.4494, | |
| "eval_steps_per_second": 0.938, | |
| "eval_ter": 41.2829, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.490539425712537e-05, | |
| "loss": 0.4595, | |
| "step": 105500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4739069111424541e-05, | |
| "loss": 0.4567, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.4572743965723715e-05, | |
| "loss": 0.491, | |
| "step": 106500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.4406418820022888e-05, | |
| "loss": 0.4752, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.4240093674322059e-05, | |
| "loss": 0.4659, | |
| "step": 107500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.4073768528621231e-05, | |
| "loss": 0.4739, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.3907443382920405e-05, | |
| "loss": 0.4735, | |
| "step": 108500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.3741118237219577e-05, | |
| "loss": 0.4496, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.3574793091518747e-05, | |
| "loss": 0.469, | |
| "step": 109500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.3408467945817921e-05, | |
| "loss": 0.4795, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_bleu": 48.1775, | |
| "eval_chrf++": 67.6488, | |
| "eval_gen_len": 24.3566, | |
| "eval_loss": 0.8540457487106323, | |
| "eval_runtime": 4943.4782, | |
| "eval_samples_per_second": 4.708, | |
| "eval_spbleu": 55.631, | |
| "eval_steps_per_second": 0.942, | |
| "eval_ter": 41.0129, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.3242142800117093e-05, | |
| "loss": 0.4636, | |
| "step": 110500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.3075817654416267e-05, | |
| "loss": 0.4674, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.2909492508715437e-05, | |
| "loss": 0.4792, | |
| "step": 111500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.2743167363014611e-05, | |
| "loss": 0.4592, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.2576842217313783e-05, | |
| "loss": 0.4624, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.2410517071612955e-05, | |
| "loss": 0.4657, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.2244191925912129e-05, | |
| "loss": 0.4656, | |
| "step": 113500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.20778667802113e-05, | |
| "loss": 0.4699, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.1911541634510473e-05, | |
| "loss": 0.4621, | |
| "step": 114500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.1745216488809645e-05, | |
| "loss": 0.4535, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_bleu": 48.3184, | |
| "eval_chrf++": 67.7304, | |
| "eval_gen_len": 24.3597, | |
| "eval_loss": 0.8529399633407593, | |
| "eval_runtime": 4955.9284, | |
| "eval_samples_per_second": 4.696, | |
| "eval_spbleu": 55.7822, | |
| "eval_steps_per_second": 0.939, | |
| "eval_ter": 40.9906, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.1578891343108817e-05, | |
| "loss": 0.4726, | |
| "step": 115500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.141256619740799e-05, | |
| "loss": 0.4627, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.1246241051707161e-05, | |
| "loss": 0.4498, | |
| "step": 116500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.1079915906006334e-05, | |
| "loss": 0.469, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.0913590760305506e-05, | |
| "loss": 0.4799, | |
| "step": 117500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.074726561460468e-05, | |
| "loss": 0.4868, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 1.0580940468903852e-05, | |
| "loss": 0.454, | |
| "step": 118500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 1.0414615323203024e-05, | |
| "loss": 0.4537, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.0248290177502196e-05, | |
| "loss": 0.4738, | |
| "step": 119500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.0081965031801368e-05, | |
| "loss": 0.4667, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_bleu": 48.525, | |
| "eval_chrf++": 67.8207, | |
| "eval_gen_len": 24.3414, | |
| "eval_loss": 0.8485604524612427, | |
| "eval_runtime": 4979.6092, | |
| "eval_samples_per_second": 4.674, | |
| "eval_spbleu": 55.8666, | |
| "eval_steps_per_second": 0.935, | |
| "eval_ter": 40.8434, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.91563988610054e-06, | |
| "loss": 0.4624, | |
| "step": 120500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.749314740399714e-06, | |
| "loss": 0.4526, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.582989594698886e-06, | |
| "loss": 0.4493, | |
| "step": 121500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.416664448998058e-06, | |
| "loss": 0.4693, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.25033930329723e-06, | |
| "loss": 0.4617, | |
| "step": 122500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 9.084014157596404e-06, | |
| "loss": 0.4631, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 8.917689011895574e-06, | |
| "loss": 0.4585, | |
| "step": 123500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 8.751363866194748e-06, | |
| "loss": 0.4721, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.58503872049392e-06, | |
| "loss": 0.4579, | |
| "step": 124500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.418713574793092e-06, | |
| "loss": 0.4864, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_bleu": 48.5947, | |
| "eval_chrf++": 67.9122, | |
| "eval_gen_len": 24.3948, | |
| "eval_loss": 0.8485385775566101, | |
| "eval_runtime": 4982.8937, | |
| "eval_samples_per_second": 4.671, | |
| "eval_spbleu": 56.0282, | |
| "eval_steps_per_second": 0.934, | |
| "eval_ter": 40.8473, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.252388429092264e-06, | |
| "loss": 0.4511, | |
| "step": 125500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.086063283391436e-06, | |
| "loss": 0.486, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 7.919738137690608e-06, | |
| "loss": 0.4667, | |
| "step": 126500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 7.75341299198978e-06, | |
| "loss": 0.483, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 7.587087846288954e-06, | |
| "loss": 0.4654, | |
| "step": 127500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 7.4207627005881254e-06, | |
| "loss": 0.4786, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 7.254437554887298e-06, | |
| "loss": 0.4689, | |
| "step": 128500 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 7.0881124091864705e-06, | |
| "loss": 0.4581, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 6.921787263485643e-06, | |
| "loss": 0.4619, | |
| "step": 129500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 6.7554621177848155e-06, | |
| "loss": 0.4636, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_bleu": 48.6674, | |
| "eval_chrf++": 67.9629, | |
| "eval_gen_len": 24.3097, | |
| "eval_loss": 0.847720205783844, | |
| "eval_runtime": 4947.6125, | |
| "eval_samples_per_second": 4.704, | |
| "eval_spbleu": 56.0708, | |
| "eval_steps_per_second": 0.941, | |
| "eval_ter": 40.5643, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 6.589136972083988e-06, | |
| "loss": 0.4585, | |
| "step": 130500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 6.42281182638316e-06, | |
| "loss": 0.4582, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 6.2564866806823334e-06, | |
| "loss": 0.4577, | |
| "step": 131500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 6.090161534981505e-06, | |
| "loss": 0.4501, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 5.923836389280677e-06, | |
| "loss": 0.4609, | |
| "step": 132500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5.75751124357985e-06, | |
| "loss": 0.4721, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.591186097879022e-06, | |
| "loss": 0.4518, | |
| "step": 133500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.424860952178194e-06, | |
| "loss": 0.4592, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.258535806477367e-06, | |
| "loss": 0.4588, | |
| "step": 134500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 5.092210660776539e-06, | |
| "loss": 0.4537, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_bleu": 48.7544, | |
| "eval_chrf++": 68.0279, | |
| "eval_gen_len": 24.3883, | |
| "eval_loss": 0.8480924367904663, | |
| "eval_runtime": 4988.7356, | |
| "eval_samples_per_second": 4.665, | |
| "eval_spbleu": 56.2147, | |
| "eval_steps_per_second": 0.933, | |
| "eval_ter": 40.5898, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.925885515075712e-06, | |
| "loss": 0.4586, | |
| "step": 135500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 4.759560369374884e-06, | |
| "loss": 0.4499, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.593235223674056e-06, | |
| "loss": 0.4576, | |
| "step": 136500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.426910077973229e-06, | |
| "loss": 0.4364, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.260584932272401e-06, | |
| "loss": 0.4417, | |
| "step": 137500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.094259786571573e-06, | |
| "loss": 0.4531, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.927934640870746e-06, | |
| "loss": 0.4653, | |
| "step": 138500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.761609495169918e-06, | |
| "loss": 0.4492, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 3.5952843494690906e-06, | |
| "loss": 0.4693, | |
| "step": 139500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 3.4289592037682623e-06, | |
| "loss": 0.4425, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_bleu": 48.8847, | |
| "eval_chrf++": 68.1124, | |
| "eval_gen_len": 24.3471, | |
| "eval_loss": 0.8463260531425476, | |
| "eval_runtime": 4980.2186, | |
| "eval_samples_per_second": 4.673, | |
| "eval_spbleu": 56.286, | |
| "eval_steps_per_second": 0.935, | |
| "eval_ter": 40.42, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.262634058067435e-06, | |
| "loss": 0.4588, | |
| "step": 140500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 3.0963089123666073e-06, | |
| "loss": 0.466, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.92998376666578e-06, | |
| "loss": 0.4338, | |
| "step": 141500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 2.7636586209649524e-06, | |
| "loss": 0.4503, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.5973334752641244e-06, | |
| "loss": 0.4632, | |
| "step": 142500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.4310083295632965e-06, | |
| "loss": 0.4401, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 2.264683183862469e-06, | |
| "loss": 0.4524, | |
| "step": 143500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.0983580381616416e-06, | |
| "loss": 0.4529, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.932032892460814e-06, | |
| "loss": 0.4418, | |
| "step": 144500 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.7657077467599864e-06, | |
| "loss": 0.4572, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_bleu": 48.961, | |
| "eval_chrf++": 68.1363, | |
| "eval_gen_len": 24.3521, | |
| "eval_loss": 0.8435162901878357, | |
| "eval_runtime": 4939.3838, | |
| "eval_samples_per_second": 4.712, | |
| "eval_spbleu": 56.3386, | |
| "eval_steps_per_second": 0.942, | |
| "eval_ter": 40.41, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.5993826010591587e-06, | |
| "loss": 0.4204, | |
| "step": 145500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.433057455358331e-06, | |
| "loss": 0.446, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.2667323096575033e-06, | |
| "loss": 0.4625, | |
| "step": 146500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.1004071639566756e-06, | |
| "loss": 0.4449, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 9.340820182558481e-07, | |
| "loss": 0.4475, | |
| "step": 147500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 7.677568725550204e-07, | |
| "loss": 0.4274, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 6.014317268541928e-07, | |
| "loss": 0.4592, | |
| "step": 148500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.351065811533651e-07, | |
| "loss": 0.4615, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 2.687814354525375e-07, | |
| "loss": 0.4492, | |
| "step": 149500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.0245628975170984e-07, | |
| "loss": 0.4682, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 49.03, | |
| "eval_chrf++": 68.1861, | |
| "eval_gen_len": 24.3596, | |
| "eval_loss": 0.8448808789253235, | |
| "eval_runtime": 4968.8377, | |
| "eval_samples_per_second": 4.684, | |
| "eval_spbleu": 56.4006, | |
| "eval_steps_per_second": 0.937, | |
| "eval_ter": 40.3942, | |
| "step": 150000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 150308, | |
| "num_train_epochs": 2, | |
| "save_steps": 5000, | |
| "total_flos": 6.383678253160202e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |