|
{ |
|
"best_metric": 0.9008079171180725, |
|
"best_model_checkpoint": "./koja_mbartLarge_55p_run1/checkpoint-32000", |
|
"epoch": 3.381744617893052, |
|
"eval_steps": 8000, |
|
"global_step": 56000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.9849024699559153e-05, |
|
"loss": 1.5854, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 4.9698049399118304e-05, |
|
"loss": 1.3951, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 4.954707409867746e-05, |
|
"loss": 1.2938, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 4.939609879823661e-05, |
|
"loss": 1.2482, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 4.9245123497795764e-05, |
|
"loss": 1.2122, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 4.9094148197354915e-05, |
|
"loss": 1.178, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 4.8943172896914066e-05, |
|
"loss": 1.1483, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 4.8792197596473224e-05, |
|
"loss": 1.1408, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 4.8641222296032375e-05, |
|
"loss": 1.1053, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 4.8490246995591526e-05, |
|
"loss": 1.0876, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 4.833927169515067e-05, |
|
"loss": 1.0745, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 4.818829639470983e-05, |
|
"loss": 1.0707, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 4.803732109426898e-05, |
|
"loss": 1.0664, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 4.788634579382813e-05, |
|
"loss": 1.0426, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 4.773537049338728e-05, |
|
"loss": 1.0357, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 4.758439519294643e-05, |
|
"loss": 1.0254, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"eval_bleu": 7.1252, |
|
"eval_gen_len": 18.5403, |
|
"eval_loss": 1.0065276622772217, |
|
"eval_runtime": 2330.5491, |
|
"eval_samples_per_second": 14.21, |
|
"eval_steps_per_second": 0.888, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 4.743341989250559e-05, |
|
"loss": 1.0195, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 4.728244459206474e-05, |
|
"loss": 1.0094, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 4.713146929162389e-05, |
|
"loss": 0.9984, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 4.698049399118304e-05, |
|
"loss": 0.9693, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 4.6829518690742194e-05, |
|
"loss": 0.9945, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 4.667854339030135e-05, |
|
"loss": 0.989, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 4.65275680898605e-05, |
|
"loss": 0.9907, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 4.6376592789419654e-05, |
|
"loss": 0.97, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 4.6225617488978805e-05, |
|
"loss": 0.9642, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 4.6074642188537956e-05, |
|
"loss": 0.9615, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 4.5923666888097114e-05, |
|
"loss": 0.9485, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 4.5772691587656265e-05, |
|
"loss": 0.9492, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 4.5621716287215416e-05, |
|
"loss": 0.9568, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 4.547074098677457e-05, |
|
"loss": 0.9507, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 4.531976568633372e-05, |
|
"loss": 0.9437, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 4.516879038589287e-05, |
|
"loss": 0.9374, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"eval_bleu": 8.2047, |
|
"eval_gen_len": 18.2091, |
|
"eval_loss": 0.9199957847595215, |
|
"eval_runtime": 2175.6016, |
|
"eval_samples_per_second": 15.222, |
|
"eval_steps_per_second": 0.951, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4.501781508545202e-05, |
|
"loss": 0.9266, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.486683978501117e-05, |
|
"loss": 0.8505, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 4.471586448457032e-05, |
|
"loss": 0.8316, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 4.456488918412948e-05, |
|
"loss": 0.8079, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 4.441391388368863e-05, |
|
"loss": 0.7993, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 4.426293858324778e-05, |
|
"loss": 0.7984, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 4.411196328280693e-05, |
|
"loss": 0.789, |
|
"step": 19500 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"learning_rate": 4.3960987982366084e-05, |
|
"loss": 0.7766, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 4.381001268192524e-05, |
|
"loss": 0.7798, |
|
"step": 20500 |
|
}, |
|
{ |
|
"epoch": 1.27, |
|
"learning_rate": 4.365903738148439e-05, |
|
"loss": 0.7717, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"learning_rate": 4.3508062081043544e-05, |
|
"loss": 0.7581, |
|
"step": 21500 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 4.3357086780602695e-05, |
|
"loss": 0.7531, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"learning_rate": 4.3206111480161846e-05, |
|
"loss": 0.7578, |
|
"step": 22500 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 4.3055136179721003e-05, |
|
"loss": 0.7528, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 4.2904160879280154e-05, |
|
"loss": 0.7479, |
|
"step": 23500 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 4.2753185578839306e-05, |
|
"loss": 0.7429, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"eval_bleu": 8.4622, |
|
"eval_gen_len": 17.9606, |
|
"eval_loss": 0.9085537791252136, |
|
"eval_runtime": 2144.5036, |
|
"eval_samples_per_second": 15.443, |
|
"eval_steps_per_second": 0.965, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"learning_rate": 4.2602210278398457e-05, |
|
"loss": 0.7292, |
|
"step": 24500 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 4.245123497795761e-05, |
|
"loss": 0.7363, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 4.2300259677516765e-05, |
|
"loss": 0.7261, |
|
"step": 25500 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 4.2149284377075916e-05, |
|
"loss": 0.7214, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 4.199830907663506e-05, |
|
"loss": 0.7077, |
|
"step": 26500 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"learning_rate": 4.184733377619421e-05, |
|
"loss": 0.724, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"learning_rate": 4.169635847575337e-05, |
|
"loss": 0.7286, |
|
"step": 27500 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"learning_rate": 4.154538317531252e-05, |
|
"loss": 0.7224, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 4.139440787487167e-05, |
|
"loss": 0.7194, |
|
"step": 28500 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 4.124343257443082e-05, |
|
"loss": 0.7095, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"learning_rate": 4.1092457273989973e-05, |
|
"loss": 0.7065, |
|
"step": 29500 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"learning_rate": 4.094148197354913e-05, |
|
"loss": 0.7012, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 4.079050667310828e-05, |
|
"loss": 0.7045, |
|
"step": 30500 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"learning_rate": 4.063953137266743e-05, |
|
"loss": 0.7099, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"learning_rate": 4.0488556072226584e-05, |
|
"loss": 0.7084, |
|
"step": 31500 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"learning_rate": 4.0337580771785735e-05, |
|
"loss": 0.7039, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"eval_bleu": 8.9198, |
|
"eval_gen_len": 17.8841, |
|
"eval_loss": 0.9008079171180725, |
|
"eval_runtime": 2125.9264, |
|
"eval_samples_per_second": 15.578, |
|
"eval_steps_per_second": 0.974, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"learning_rate": 4.018660547134489e-05, |
|
"loss": 0.7029, |
|
"step": 32500 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 4.0035630170904044e-05, |
|
"loss": 0.6983, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 3.9884654870463195e-05, |
|
"loss": 0.6503, |
|
"step": 33500 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"learning_rate": 3.9733679570022346e-05, |
|
"loss": 0.6191, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 3.95827042695815e-05, |
|
"loss": 0.605, |
|
"step": 34500 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"learning_rate": 3.9431728969140655e-05, |
|
"loss": 0.6035, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"learning_rate": 3.9280753668699806e-05, |
|
"loss": 0.6008, |
|
"step": 35500 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 3.912977836825896e-05, |
|
"loss": 0.5947, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"learning_rate": 3.897880306781811e-05, |
|
"loss": 0.5812, |
|
"step": 36500 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"learning_rate": 3.882782776737726e-05, |
|
"loss": 0.587, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"learning_rate": 3.867685246693641e-05, |
|
"loss": 0.5857, |
|
"step": 37500 |
|
}, |
|
{ |
|
"epoch": 2.29, |
|
"learning_rate": 3.852587716649556e-05, |
|
"loss": 0.5659, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 2.32, |
|
"learning_rate": 3.837490186605471e-05, |
|
"loss": 0.5702, |
|
"step": 38500 |
|
}, |
|
{ |
|
"epoch": 2.36, |
|
"learning_rate": 3.822392656561386e-05, |
|
"loss": 0.5651, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"learning_rate": 3.807295126517302e-05, |
|
"loss": 0.5691, |
|
"step": 39500 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"learning_rate": 3.792197596473217e-05, |
|
"loss": 0.5616, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"eval_bleu": 8.7938, |
|
"eval_gen_len": 17.78, |
|
"eval_loss": 0.9209699630737305, |
|
"eval_runtime": 2116.9975, |
|
"eval_samples_per_second": 15.644, |
|
"eval_steps_per_second": 0.978, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"learning_rate": 3.777100066429132e-05, |
|
"loss": 0.5552, |
|
"step": 40500 |
|
}, |
|
{ |
|
"epoch": 2.48, |
|
"learning_rate": 3.7620025363850474e-05, |
|
"loss": 0.5461, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 2.51, |
|
"learning_rate": 3.7469050063409625e-05, |
|
"loss": 0.5546, |
|
"step": 41500 |
|
}, |
|
{ |
|
"epoch": 2.54, |
|
"learning_rate": 3.731807476296878e-05, |
|
"loss": 0.5383, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 2.57, |
|
"learning_rate": 3.7167099462527934e-05, |
|
"loss": 0.5452, |
|
"step": 42500 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"learning_rate": 3.7016124162087085e-05, |
|
"loss": 0.5318, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 2.63, |
|
"learning_rate": 3.6865148861646236e-05, |
|
"loss": 0.5407, |
|
"step": 43500 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 3.671417356120539e-05, |
|
"loss": 0.549, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 2.69, |
|
"learning_rate": 3.6563198260764545e-05, |
|
"loss": 0.5384, |
|
"step": 44500 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"learning_rate": 3.6412222960323696e-05, |
|
"loss": 0.5441, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 2.75, |
|
"learning_rate": 3.626124765988285e-05, |
|
"loss": 0.5317, |
|
"step": 45500 |
|
}, |
|
{ |
|
"epoch": 2.78, |
|
"learning_rate": 3.6110272359442e-05, |
|
"loss": 0.5352, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 2.81, |
|
"learning_rate": 3.595929705900115e-05, |
|
"loss": 0.5248, |
|
"step": 46500 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 3.5808321758560307e-05, |
|
"loss": 0.5309, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"learning_rate": 3.565734645811946e-05, |
|
"loss": 0.5296, |
|
"step": 47500 |
|
}, |
|
{ |
|
"epoch": 2.9, |
|
"learning_rate": 3.55063711576786e-05, |
|
"loss": 0.5355, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 2.9, |
|
"eval_bleu": 8.6506, |
|
"eval_gen_len": 17.9146, |
|
"eval_loss": 0.9399827122688293, |
|
"eval_runtime": 2117.673, |
|
"eval_samples_per_second": 15.639, |
|
"eval_steps_per_second": 0.977, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 2.93, |
|
"learning_rate": 3.535539585723775e-05, |
|
"loss": 0.5262, |
|
"step": 48500 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"learning_rate": 3.520442055679691e-05, |
|
"loss": 0.5247, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"learning_rate": 3.505344525635606e-05, |
|
"loss": 0.5315, |
|
"step": 49500 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 3.490246995591521e-05, |
|
"loss": 0.4932, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 3.05, |
|
"learning_rate": 3.4751494655474364e-05, |
|
"loss": 0.4595, |
|
"step": 50500 |
|
}, |
|
{ |
|
"epoch": 3.08, |
|
"learning_rate": 3.4600519355033515e-05, |
|
"loss": 0.45, |
|
"step": 51000 |
|
}, |
|
{ |
|
"epoch": 3.11, |
|
"learning_rate": 3.444954405459267e-05, |
|
"loss": 0.4477, |
|
"step": 51500 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 3.4298568754151824e-05, |
|
"loss": 0.4423, |
|
"step": 52000 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 3.4147593453710975e-05, |
|
"loss": 0.435, |
|
"step": 52500 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"learning_rate": 3.3996618153270126e-05, |
|
"loss": 0.4345, |
|
"step": 53000 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"learning_rate": 3.3845642852829277e-05, |
|
"loss": 0.4337, |
|
"step": 53500 |
|
}, |
|
{ |
|
"epoch": 3.26, |
|
"learning_rate": 3.3694667552388434e-05, |
|
"loss": 0.4295, |
|
"step": 54000 |
|
}, |
|
{ |
|
"epoch": 3.29, |
|
"learning_rate": 3.3543692251947585e-05, |
|
"loss": 0.4143, |
|
"step": 54500 |
|
}, |
|
{ |
|
"epoch": 3.32, |
|
"learning_rate": 3.3392716951506736e-05, |
|
"loss": 0.4203, |
|
"step": 55000 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"learning_rate": 3.324174165106589e-05, |
|
"loss": 0.4164, |
|
"step": 55500 |
|
}, |
|
{ |
|
"epoch": 3.38, |
|
"learning_rate": 3.309076635062504e-05, |
|
"loss": 0.4194, |
|
"step": 56000 |
|
}, |
|
{ |
|
"epoch": 3.38, |
|
"eval_bleu": 8.4235, |
|
"eval_gen_len": 17.6759, |
|
"eval_loss": 1.0053656101226807, |
|
"eval_runtime": 2100.231, |
|
"eval_samples_per_second": 15.769, |
|
"eval_steps_per_second": 0.986, |
|
"step": 56000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 165590, |
|
"num_train_epochs": 10, |
|
"save_steps": 8000, |
|
"total_flos": 1.9418522940310815e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|