| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.97323295692179, | |
| "eval_steps": 2000, | |
| "global_step": 38000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9895441237975744e-05, | |
| "loss": 2.5556, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9790882475951487e-05, | |
| "loss": 2.1664, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.968632371392723e-05, | |
| "loss": 2.1184, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.958176495190297e-05, | |
| "loss": 2.0691, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.9477206189878712e-05, | |
| "loss": 2.0324, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.9372647427854455e-05, | |
| "loss": 2.0135, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.9268088665830197e-05, | |
| "loss": 2.0028, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.916352990380594e-05, | |
| "loss": 1.99, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.9059493935591804e-05, | |
| "loss": 1.9608, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.8954935173567547e-05, | |
| "loss": 1.9402, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_bleu": 0.2597, | |
| "eval_gen_len": 18.3893, | |
| "eval_loss": 1.7414125204086304, | |
| "eval_runtime": 71.3769, | |
| "eval_samples_per_second": 45.127, | |
| "eval_steps_per_second": 1.415, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.885089920535341e-05, | |
| "loss": 1.9321, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.8746340443329153e-05, | |
| "loss": 1.8953, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.8641781681304896e-05, | |
| "loss": 1.9034, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.853722291928064e-05, | |
| "loss": 1.8885, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.8432664157256378e-05, | |
| "loss": 1.8809, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.832810539523212e-05, | |
| "loss": 1.8842, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.8223546633207864e-05, | |
| "loss": 1.8553, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.8118987871183606e-05, | |
| "loss": 1.8627, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.801442910915935e-05, | |
| "loss": 1.8589, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.7909870347135092e-05, | |
| "loss": 1.8503, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_bleu": 0.2671, | |
| "eval_gen_len": 18.3657, | |
| "eval_loss": 1.6917428970336914, | |
| "eval_runtime": 71.0352, | |
| "eval_samples_per_second": 45.344, | |
| "eval_steps_per_second": 1.422, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.7805311585110835e-05, | |
| "loss": 1.8452, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.7700752823086574e-05, | |
| "loss": 1.8537, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.759619406106232e-05, | |
| "loss": 1.8557, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.749163529903806e-05, | |
| "loss": 1.8429, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.7387076537013802e-05, | |
| "loss": 1.8429, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.7282517774989545e-05, | |
| "loss": 1.8295, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.7177959012965288e-05, | |
| "loss": 1.8243, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.707340025094103e-05, | |
| "loss": 1.8388, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.6968841488916773e-05, | |
| "loss": 1.8347, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.6864282726892516e-05, | |
| "loss": 1.8342, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_bleu": 0.2695, | |
| "eval_gen_len": 18.2366, | |
| "eval_loss": 1.6622676849365234, | |
| "eval_runtime": 70.9517, | |
| "eval_samples_per_second": 45.397, | |
| "eval_steps_per_second": 1.424, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.675972396486826e-05, | |
| "loss": 1.8022, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.6655165202843998e-05, | |
| "loss": 1.8215, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.6550606440819744e-05, | |
| "loss": 1.8081, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.6446047678795483e-05, | |
| "loss": 1.8093, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.6341488916771226e-05, | |
| "loss": 1.7879, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.623693015474697e-05, | |
| "loss": 1.8126, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.6132371392722712e-05, | |
| "loss": 1.7782, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.6027812630698454e-05, | |
| "loss": 1.7955, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.5923253868674197e-05, | |
| "loss": 1.7975, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.581869510664994e-05, | |
| "loss": 1.8201, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_bleu": 0.2721, | |
| "eval_gen_len": 18.2549, | |
| "eval_loss": 1.643432378768921, | |
| "eval_runtime": 71.6597, | |
| "eval_samples_per_second": 44.949, | |
| "eval_steps_per_second": 1.409, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.5714136344625683e-05, | |
| "loss": 1.8129, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.5609577582601422e-05, | |
| "loss": 1.7791, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.5505018820577165e-05, | |
| "loss": 1.7857, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.5400460058552907e-05, | |
| "loss": 1.7819, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.529590129652865e-05, | |
| "loss": 1.7665, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.5191342534504391e-05, | |
| "loss": 1.7671, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.5086783772480136e-05, | |
| "loss": 1.7696, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.4982747804265998e-05, | |
| "loss": 1.781, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.4878189042241742e-05, | |
| "loss": 1.751, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.4773630280217483e-05, | |
| "loss": 1.7567, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_bleu": 0.2744, | |
| "eval_gen_len": 18.1981, | |
| "eval_loss": 1.629482626914978, | |
| "eval_runtime": 71.9531, | |
| "eval_samples_per_second": 44.765, | |
| "eval_steps_per_second": 1.404, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.4669071518193226e-05, | |
| "loss": 1.7625, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.4564512756168969e-05, | |
| "loss": 1.7674, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.4459953994144712e-05, | |
| "loss": 1.7418, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1.4355395232120453e-05, | |
| "loss": 1.7596, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.4250836470096194e-05, | |
| "loss": 1.7466, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.4146277708071938e-05, | |
| "loss": 1.7583, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.4041718946047679e-05, | |
| "loss": 1.7522, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1.3937160184023422e-05, | |
| "loss": 1.7332, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.3832601421999166e-05, | |
| "loss": 1.7557, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.3728042659974907e-05, | |
| "loss": 1.745, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_bleu": 0.2747, | |
| "eval_gen_len": 18.1897, | |
| "eval_loss": 1.6197348833084106, | |
| "eval_runtime": 70.3482, | |
| "eval_samples_per_second": 45.787, | |
| "eval_steps_per_second": 1.436, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.3623483897950648e-05, | |
| "loss": 1.744, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.3518925135926391e-05, | |
| "loss": 1.7341, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.3414366373902134e-05, | |
| "loss": 1.7437, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.3309807611877877e-05, | |
| "loss": 1.7423, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.3205248849853618e-05, | |
| "loss": 1.7377, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.3100690087829362e-05, | |
| "loss": 1.7399, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.2996131325805103e-05, | |
| "loss": 1.7488, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.2891572563780846e-05, | |
| "loss": 1.7621, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.278753659556671e-05, | |
| "loss": 1.7282, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.268297783354245e-05, | |
| "loss": 1.7492, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_bleu": 0.2753, | |
| "eval_gen_len": 18.1773, | |
| "eval_loss": 1.608814001083374, | |
| "eval_runtime": 71.3668, | |
| "eval_samples_per_second": 45.133, | |
| "eval_steps_per_second": 1.415, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.2578419071518195e-05, | |
| "loss": 1.7392, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.2473860309493936e-05, | |
| "loss": 1.7378, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.2369301547469679e-05, | |
| "loss": 1.727, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.226474278544542e-05, | |
| "loss": 1.7366, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.2160184023421164e-05, | |
| "loss": 1.7405, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.2055625261396906e-05, | |
| "loss": 1.7292, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 1.1951066499372648e-05, | |
| "loss": 1.7514, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.1846507737348391e-05, | |
| "loss": 1.7322, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.1742471769134253e-05, | |
| "loss": 1.7417, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 1.1637913007109998e-05, | |
| "loss": 1.7126, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_bleu": 0.2771, | |
| "eval_gen_len": 18.2412, | |
| "eval_loss": 1.6022411584854126, | |
| "eval_runtime": 71.0218, | |
| "eval_samples_per_second": 45.352, | |
| "eval_steps_per_second": 1.422, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.1533354245085739e-05, | |
| "loss": 1.7279, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 1.1428795483061481e-05, | |
| "loss": 1.7124, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.1324236721037222e-05, | |
| "loss": 1.7307, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.1219677959012967e-05, | |
| "loss": 1.7178, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.1115119196988708e-05, | |
| "loss": 1.7, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.101056043496445e-05, | |
| "loss": 1.7064, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.0906001672940193e-05, | |
| "loss": 1.7242, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.0801442910915936e-05, | |
| "loss": 1.7111, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.0696884148891677e-05, | |
| "loss": 1.7125, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.0592325386867422e-05, | |
| "loss": 1.7251, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_bleu": 0.2773, | |
| "eval_gen_len": 18.1916, | |
| "eval_loss": 1.5964778661727905, | |
| "eval_runtime": 71.6245, | |
| "eval_samples_per_second": 44.971, | |
| "eval_steps_per_second": 1.41, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.0488289418653284e-05, | |
| "loss": 1.7147, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.0383730656629028e-05, | |
| "loss": 1.7224, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.027917189460477e-05, | |
| "loss": 1.717, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.017461313258051e-05, | |
| "loss": 1.7086, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.0070054370556253e-05, | |
| "loss": 1.7117, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.966018402342117e-06, | |
| "loss": 1.7005, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.86145964031786e-06, | |
| "loss": 1.6843, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.7569008782936e-06, | |
| "loss": 1.6922, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.652342116269344e-06, | |
| "loss": 1.6985, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.547783354245086e-06, | |
| "loss": 1.6824, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_bleu": 0.2763, | |
| "eval_gen_len": 18.1354, | |
| "eval_loss": 1.5918101072311401, | |
| "eval_runtime": 69.9735, | |
| "eval_samples_per_second": 46.032, | |
| "eval_steps_per_second": 1.443, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.443224592220829e-06, | |
| "loss": 1.7036, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 9.338665830196572e-06, | |
| "loss": 1.7075, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.234107068172313e-06, | |
| "loss": 1.7022, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.129548306148056e-06, | |
| "loss": 1.6877, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 9.024989544123798e-06, | |
| "loss": 1.6877, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.920430782099541e-06, | |
| "loss": 1.6853, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.815872020075284e-06, | |
| "loss": 1.7016, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.711313258051025e-06, | |
| "loss": 1.7158, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 8.606754496026768e-06, | |
| "loss": 1.695, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 8.50219573400251e-06, | |
| "loss": 1.6868, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_bleu": 0.2786, | |
| "eval_gen_len": 18.1912, | |
| "eval_loss": 1.5873745679855347, | |
| "eval_runtime": 70.6377, | |
| "eval_samples_per_second": 45.599, | |
| "eval_steps_per_second": 1.43, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 8.397636971978253e-06, | |
| "loss": 1.6916, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 8.293078209953996e-06, | |
| "loss": 1.7041, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.188519447929737e-06, | |
| "loss": 1.6802, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 8.08396068590548e-06, | |
| "loss": 1.6876, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 7.979401923881222e-06, | |
| "loss": 1.6922, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.874843161856965e-06, | |
| "loss": 1.7058, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 7.770284399832708e-06, | |
| "loss": 1.6939, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 7.665725637808449e-06, | |
| "loss": 1.6887, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 7.5616896695943135e-06, | |
| "loss": 1.6785, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 7.4571309075700546e-06, | |
| "loss": 1.7067, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_bleu": 0.2776, | |
| "eval_gen_len": 18.109, | |
| "eval_loss": 1.5828683376312256, | |
| "eval_runtime": 70.2744, | |
| "eval_samples_per_second": 45.835, | |
| "eval_steps_per_second": 1.437, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.352572145545797e-06, | |
| "loss": 1.6752, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 7.248013383521539e-06, | |
| "loss": 1.6866, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.143454621497282e-06, | |
| "loss": 1.6757, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 7.038895859473024e-06, | |
| "loss": 1.679, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.9343370974487666e-06, | |
| "loss": 1.6995, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 6.829778335424509e-06, | |
| "loss": 1.6851, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 6.725219573400251e-06, | |
| "loss": 1.6938, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 6.620660811375994e-06, | |
| "loss": 1.6919, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 6.516102049351736e-06, | |
| "loss": 1.6884, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 6.4115432873274786e-06, | |
| "loss": 1.6916, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_bleu": 0.2787, | |
| "eval_gen_len": 18.1813, | |
| "eval_loss": 1.5792288780212402, | |
| "eval_runtime": 70.8136, | |
| "eval_samples_per_second": 45.486, | |
| "eval_steps_per_second": 1.426, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 6.306984525303221e-06, | |
| "loss": 1.6838, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 6.202425763278963e-06, | |
| "loss": 1.6926, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 6.098389795064827e-06, | |
| "loss": 1.684, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 5.993831033040569e-06, | |
| "loss": 1.6785, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 5.889272271016312e-06, | |
| "loss": 1.68, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5.784713508992054e-06, | |
| "loss": 1.683, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 5.680154746967796e-06, | |
| "loss": 1.6782, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 5.575595984943539e-06, | |
| "loss": 1.6867, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 5.471037222919281e-06, | |
| "loss": 1.6677, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 5.366478460895024e-06, | |
| "loss": 1.6702, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_bleu": 0.279, | |
| "eval_gen_len": 18.1614, | |
| "eval_loss": 1.5775508880615234, | |
| "eval_runtime": 70.2225, | |
| "eval_samples_per_second": 45.868, | |
| "eval_steps_per_second": 1.438, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 5.261919698870766e-06, | |
| "loss": 1.6805, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 5.157360936846508e-06, | |
| "loss": 1.6956, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 5.05280217482225e-06, | |
| "loss": 1.6998, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.948766206608114e-06, | |
| "loss": 1.6684, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.844207444583857e-06, | |
| "loss": 1.6834, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.739648682559599e-06, | |
| "loss": 1.6769, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.6350899205353415e-06, | |
| "loss": 1.6872, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 4.531053952321205e-06, | |
| "loss": 1.6708, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.426495190296947e-06, | |
| "loss": 1.6631, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 4.321936428272689e-06, | |
| "loss": 1.6792, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_bleu": 0.2784, | |
| "eval_gen_len": 18.1822, | |
| "eval_loss": 1.5754998922348022, | |
| "eval_runtime": 70.3523, | |
| "eval_samples_per_second": 45.784, | |
| "eval_steps_per_second": 1.436, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 4.217377666248432e-06, | |
| "loss": 1.6546, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.112818904224175e-06, | |
| "loss": 1.6753, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 4.0082601421999166e-06, | |
| "loss": 1.6361, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 3.903701380175659e-06, | |
| "loss": 1.6696, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 3.799142618151401e-06, | |
| "loss": 1.6658, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 3.6945838561271435e-06, | |
| "loss": 1.6802, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 3.5900250941028862e-06, | |
| "loss": 1.6809, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 3.4854663320786286e-06, | |
| "loss": 1.6615, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.380907570054371e-06, | |
| "loss": 1.6761, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 3.276348808030113e-06, | |
| "loss": 1.65, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_bleu": 0.2791, | |
| "eval_gen_len": 18.1704, | |
| "eval_loss": 1.5742353200912476, | |
| "eval_runtime": 69.3182, | |
| "eval_samples_per_second": 46.467, | |
| "eval_steps_per_second": 1.457, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 3.1717900460058555e-06, | |
| "loss": 1.6806, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.067231283981598e-06, | |
| "loss": 1.6687, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.9626725219573405e-06, | |
| "loss": 1.6667, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.858113759933083e-06, | |
| "loss": 1.6667, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 2.753554997908825e-06, | |
| "loss": 1.6679, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.648996235884567e-06, | |
| "loss": 1.6717, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 2.5444374738603094e-06, | |
| "loss": 1.6806, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.439878711836052e-06, | |
| "loss": 1.6491, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.3353199498117944e-06, | |
| "loss": 1.6879, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 2.2307611877875367e-06, | |
| "loss": 1.6708, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "eval_bleu": 0.2797, | |
| "eval_gen_len": 18.1627, | |
| "eval_loss": 1.5723748207092285, | |
| "eval_runtime": 70.4906, | |
| "eval_samples_per_second": 45.694, | |
| "eval_steps_per_second": 1.433, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 2.126202425763279e-06, | |
| "loss": 1.6777, | |
| "step": 34200 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 2.0216436637390218e-06, | |
| "loss": 1.6832, | |
| "step": 34400 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 1.9176076955248853e-06, | |
| "loss": 1.6621, | |
| "step": 34600 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.8130489335006274e-06, | |
| "loss": 1.6588, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 1.70849017147637e-06, | |
| "loss": 1.6655, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.6039314094521122e-06, | |
| "loss": 1.6618, | |
| "step": 35200 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.4993726474278547e-06, | |
| "loss": 1.6548, | |
| "step": 35400 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.394813885403597e-06, | |
| "loss": 1.6768, | |
| "step": 35600 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.2902551233793392e-06, | |
| "loss": 1.6629, | |
| "step": 35800 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 1.1856963613550817e-06, | |
| "loss": 1.6683, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "eval_bleu": 0.2799, | |
| "eval_gen_len": 18.1835, | |
| "eval_loss": 1.5721137523651123, | |
| "eval_runtime": 70.8793, | |
| "eval_samples_per_second": 45.443, | |
| "eval_steps_per_second": 1.425, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.081137599330824e-06, | |
| "loss": 1.6514, | |
| "step": 36200 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 9.765788373065663e-07, | |
| "loss": 1.6605, | |
| "step": 36400 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.720200752823086e-07, | |
| "loss": 1.6594, | |
| "step": 36600 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 7.67461313258051e-07, | |
| "loss": 1.6662, | |
| "step": 36800 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 6.629025512337935e-07, | |
| "loss": 1.6949, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 5.583437892095358e-07, | |
| "loss": 1.6592, | |
| "step": 37200 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.543078209953995e-07, | |
| "loss": 1.6552, | |
| "step": 37400 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.497490589711418e-07, | |
| "loss": 1.6605, | |
| "step": 37600 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 2.4519029694688416e-07, | |
| "loss": 1.6834, | |
| "step": 37800 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 1.4063153492262652e-07, | |
| "loss": 1.6643, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_bleu": 0.2801, | |
| "eval_gen_len": 18.1801, | |
| "eval_loss": 1.5713245868682861, | |
| "eval_runtime": 70.3463, | |
| "eval_samples_per_second": 45.788, | |
| "eval_steps_per_second": 1.436, | |
| "step": 38000 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 38256, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 7.865111955999744e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |