Summarization / trainer_state.json
roy
gogamza: summarization
f63bba1
{
"best_metric": 1.9633301496505737,
"best_model_checkpoint": "output/checkpoint-6000",
"epoch": 10.0,
"global_step": 35480,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 1.1231131887499657e-07,
"loss": 2.472,
"step": 100
},
{
"epoch": 0.06,
"learning_rate": 2.5441897566929925e-05,
"loss": 2.3979,
"step": 200
},
{
"epoch": 0.08,
"learning_rate": 1.0307801958256833e-07,
"loss": 2.3356,
"step": 300
},
{
"epoch": 0.11,
"learning_rate": 4.2830331047183876e-05,
"loss": 2.3626,
"step": 400
},
{
"epoch": 0.14,
"learning_rate": 2.5245954827173344e-05,
"loss": 2.3347,
"step": 500
},
{
"epoch": 0.17,
"learning_rate": 7.54679092701715e-06,
"loss": 2.3057,
"step": 600
},
{
"epoch": 0.2,
"learning_rate": 1.007695167624979e-07,
"loss": 2.264,
"step": 700
},
{
"epoch": 0.23,
"learning_rate": 4.813811122230701e-05,
"loss": 2.2884,
"step": 800
},
{
"epoch": 0.25,
"learning_rate": 4.276145918519949e-05,
"loss": 2.3137,
"step": 900
},
{
"epoch": 0.28,
"learning_rate": 3.468839804192268e-05,
"loss": 2.3233,
"step": 1000
},
{
"epoch": 0.31,
"learning_rate": 2.514797816905896e-05,
"loss": 2.2754,
"step": 1100
},
{
"epoch": 0.34,
"learning_rate": 1.559264200813033e-05,
"loss": 2.2508,
"step": 1200
},
{
"epoch": 0.37,
"learning_rate": 7.54679092701715e-06,
"loss": 2.2591,
"step": 1300
},
{
"epoch": 0.39,
"learning_rate": 2.0749052704813964e-06,
"loss": 2.231,
"step": 1400
},
{
"epoch": 0.42,
"learning_rate": 1.007695167624979e-07,
"loss": 2.2262,
"step": 1500
},
{
"epoch": 0.45,
"learning_rate": 4.953951865520701e-05,
"loss": 2.2988,
"step": 1600
},
{
"epoch": 0.48,
"learning_rate": 4.813811122230701e-05,
"loss": 2.2338,
"step": 1700
},
{
"epoch": 0.51,
"learning_rate": 4.5849440623095696e-05,
"loss": 2.2921,
"step": 1800
},
{
"epoch": 0.54,
"learning_rate": 4.276145918519949e-05,
"loss": 2.2703,
"step": 1900
},
{
"epoch": 0.56,
"learning_rate": 3.8992836303516824e-05,
"loss": 2.2635,
"step": 2000
},
{
"epoch": 0.56,
"eval_gen_len": 20.0,
"eval_loss": 2.061249256134033,
"eval_rouge1": 21.7272,
"eval_rouge2": 5.8776,
"eval_rougeL": 21.533,
"eval_rougeLsum": 21.5147,
"eval_runtime": 165.5553,
"eval_samples_per_second": 42.33,
"eval_steps_per_second": 2.646,
"step": 2000
},
{
"epoch": 0.59,
"learning_rate": 3.468839804192268e-05,
"loss": 2.2318,
"step": 2100
},
{
"epoch": 0.62,
"learning_rate": 3.001356154885334e-05,
"loss": 2.1998,
"step": 2200
},
{
"epoch": 0.65,
"learning_rate": 2.514797816905896e-05,
"loss": 2.2198,
"step": 2300
},
{
"epoch": 0.68,
"learning_rate": 2.027862954317443e-05,
"loss": 2.1908,
"step": 2400
},
{
"epoch": 0.7,
"learning_rate": 1.559264200813033e-05,
"loss": 2.2039,
"step": 2500
},
{
"epoch": 0.73,
"learning_rate": 1.1270095436966314e-05,
"loss": 2.1967,
"step": 2600
},
{
"epoch": 0.76,
"learning_rate": 7.477102870300167e-06,
"loss": 2.1736,
"step": 2700
},
{
"epoch": 0.79,
"learning_rate": 4.359426885334149e-06,
"loss": 2.1954,
"step": 2800
},
{
"epoch": 0.82,
"learning_rate": 2.0368780217576533e-06,
"loss": 2.1828,
"step": 2900
},
{
"epoch": 0.85,
"learning_rate": 5.987105298975171e-07,
"loss": 2.1688,
"step": 3000
},
{
"epoch": 0.87,
"learning_rate": 1.0019237993230962e-07,
"loss": 2.1516,
"step": 3100
},
{
"epoch": 0.9,
"learning_rate": 4.9886953900836016e-05,
"loss": 2.2263,
"step": 3200
},
{
"epoch": 0.93,
"learning_rate": 4.953482257211965e-05,
"loss": 2.1939,
"step": 3300
},
{
"epoch": 0.96,
"learning_rate": 4.8946889017268244e-05,
"loss": 2.2039,
"step": 3400
},
{
"epoch": 0.99,
"learning_rate": 4.8128815357812196e-05,
"loss": 2.1897,
"step": 3500
},
{
"epoch": 1.01,
"learning_rate": 4.708848009029661e-05,
"loss": 2.1036,
"step": 3600
},
{
"epoch": 1.04,
"learning_rate": 4.583590221205278e-05,
"loss": 2.0092,
"step": 3700
},
{
"epoch": 1.07,
"learning_rate": 4.43831447327769e-05,
"loss": 2.0428,
"step": 3800
},
{
"epoch": 1.1,
"learning_rate": 4.2744198501152435e-05,
"loss": 2.0471,
"step": 3900
},
{
"epoch": 1.13,
"learning_rate": 4.093484746532906e-05,
"loss": 2.0365,
"step": 4000
},
{
"epoch": 1.13,
"eval_gen_len": 20.0,
"eval_loss": 2.034134864807129,
"eval_rouge1": 21.7168,
"eval_rouge2": 5.7378,
"eval_rougeL": 21.5563,
"eval_rougeLsum": 21.552,
"eval_runtime": 164.9689,
"eval_samples_per_second": 42.481,
"eval_steps_per_second": 2.655,
"step": 4000
},
{
"epoch": 1.16,
"learning_rate": 3.897251666487364e-05,
"loss": 2.0227,
"step": 4100
},
{
"epoch": 1.18,
"learning_rate": 3.68761044181137e-05,
"loss": 2.0152,
"step": 4200
},
{
"epoch": 1.21,
"learning_rate": 3.4665800321001055e-05,
"loss": 2.0071,
"step": 4300
},
{
"epoch": 1.24,
"learning_rate": 3.2362890810266e-05,
"loss": 2.0391,
"step": 4400
},
{
"epoch": 1.27,
"learning_rate": 2.9989554163395335e-05,
"loss": 2.0197,
"step": 4500
},
{
"epoch": 1.3,
"learning_rate": 2.7568646909696192e-05,
"loss": 1.9765,
"step": 4600
},
{
"epoch": 1.32,
"learning_rate": 2.512348370942422e-05,
"loss": 1.9761,
"step": 4700
},
{
"epoch": 1.35,
"learning_rate": 2.2677612820860213e-05,
"loss": 1.992,
"step": 4800
},
{
"epoch": 1.38,
"learning_rate": 2.0254589317710082e-05,
"loss": 2.0143,
"step": 4900
},
{
"epoch": 1.41,
"learning_rate": 1.7877748240868528e-05,
"loss": 1.9675,
"step": 5000
},
{
"epoch": 1.44,
"learning_rate": 1.556997986921904e-05,
"loss": 2.0144,
"step": 5100
},
{
"epoch": 1.47,
"learning_rate": 1.3353509273735138e-05,
"loss": 2.0131,
"step": 5200
},
{
"epoch": 1.49,
"learning_rate": 1.1249682277897386e-05,
"loss": 1.9536,
"step": 5300
},
{
"epoch": 1.52,
"learning_rate": 9.278759885744533e-06,
"loss": 1.989,
"step": 5400
},
{
"epoch": 1.55,
"learning_rate": 7.45972315732879e-06,
"loss": 1.9796,
"step": 5500
},
{
"epoch": 1.58,
"learning_rate": 5.810090410731417e-06,
"loss": 1.9478,
"step": 5600
},
{
"epoch": 1.61,
"learning_rate": 4.345748511083142e-06,
"loss": 1.9681,
"step": 5700
},
{
"epoch": 1.63,
"learning_rate": 3.08079987136829e-06,
"loss": 1.9642,
"step": 5800
},
{
"epoch": 1.66,
"learning_rate": 2.027426638478571e-06,
"loss": 1.983,
"step": 5900
},
{
"epoch": 1.69,
"learning_rate": 1.1957733724791018e-06,
"loss": 1.9847,
"step": 6000
},
{
"epoch": 1.69,
"eval_gen_len": 20.0,
"eval_loss": 1.9633301496505737,
"eval_rouge1": 22.6551,
"eval_rouge2": 6.1328,
"eval_rougeL": 22.457,
"eval_rougeLsum": 22.4619,
"eval_runtime": 165.0565,
"eval_samples_per_second": 42.458,
"eval_steps_per_second": 2.654,
"step": 6000
},
{
"epoch": 1.72,
"learning_rate": 5.938493489493758e-07,
"loss": 1.9596,
"step": 6100
},
{
"epoch": 1.75,
"learning_rate": 2.27451425281131e-07,
"loss": 1.9612,
"step": 6200
},
{
"epoch": 1.78,
"learning_rate": 1.0010821377276459e-07,
"loss": 1.9514,
"step": 6300
},
{
"epoch": 1.8,
"learning_rate": 4.99717224507643e-05,
"loss": 1.986,
"step": 6400
},
{
"epoch": 1.83,
"learning_rate": 4.988343334271749e-05,
"loss": 2.0322,
"step": 6500
},
{
"epoch": 1.86,
"learning_rate": 4.9735318318596204e-05,
"loss": 2.0483,
"step": 6600
},
{
"epoch": 1.89,
"learning_rate": 4.9527734200466905e-05,
"loss": 2.0378,
"step": 6700
},
{
"epoch": 1.92,
"learning_rate": 4.926118107665238e-05,
"loss": 2.012,
"step": 6800
},
{
"epoch": 1.94,
"learning_rate": 4.893983616044946e-05,
"loss": 2.0259,
"step": 6900
},
{
"epoch": 1.97,
"learning_rate": 4.8557983100045764e-05,
"loss": 2.0511,
"step": 7000
},
{
"epoch": 2.0,
"learning_rate": 4.811949724929791e-05,
"loss": 2.0332,
"step": 7100
},
{
"epoch": 2.03,
"learning_rate": 4.762543495902719e-05,
"loss": 1.8404,
"step": 7200
},
{
"epoch": 2.06,
"learning_rate": 4.707698646856561e-05,
"loss": 1.8565,
"step": 7300
},
{
"epoch": 2.09,
"learning_rate": 4.64754730383651e-05,
"loss": 1.8571,
"step": 7400
},
{
"epoch": 2.11,
"learning_rate": 4.582234376696538e-05,
"loss": 1.8639,
"step": 7500
},
{
"epoch": 2.14,
"learning_rate": 4.511917209998861e-05,
"loss": 1.9017,
"step": 7600
},
{
"epoch": 2.17,
"learning_rate": 4.43676520395711e-05,
"loss": 1.8625,
"step": 7700
},
{
"epoch": 2.2,
"learning_rate": 4.3569594063363784e-05,
"loss": 1.8604,
"step": 7800
},
{
"epoch": 2.23,
"learning_rate": 4.2726920762932964e-05,
"loss": 1.8748,
"step": 7900
},
{
"epoch": 2.25,
"learning_rate": 4.1841662212068846e-05,
"loss": 1.861,
"step": 8000
},
{
"epoch": 2.25,
"eval_gen_len": 20.0,
"eval_loss": 2.010927438735962,
"eval_rouge1": 21.5977,
"eval_rouge2": 5.9659,
"eval_rougeL": 21.4021,
"eval_rougeLsum": 21.3731,
"eval_runtime": 165.7344,
"eval_samples_per_second": 42.285,
"eval_steps_per_second": 2.643,
"step": 8000
},
{
"epoch": 2.28,
"learning_rate": 4.091595107615995e-05,
"loss": 1.8498,
"step": 8100
},
{
"epoch": 2.31,
"learning_rate": 3.99520174744154e-05,
"loss": 1.8755,
"step": 8200
},
{
"epoch": 2.34,
"learning_rate": 3.895218360731241e-05,
"loss": 1.8772,
"step": 8300
},
{
"epoch": 2.37,
"learning_rate": 3.791885816221194e-05,
"loss": 1.8761,
"step": 8400
},
{
"epoch": 2.4,
"learning_rate": 3.6854530510619846e-05,
"loss": 1.879,
"step": 8500
},
{
"epoch": 2.42,
"learning_rate": 3.576176471107291e-05,
"loss": 1.8655,
"step": 8600
},
{
"epoch": 2.45,
"learning_rate": 3.4643193332097226e-05,
"loss": 1.8636,
"step": 8700
},
{
"epoch": 2.48,
"learning_rate": 3.350151111012003e-05,
"loss": 1.896,
"step": 8800
},
{
"epoch": 2.51,
"learning_rate": 3.23394684576136e-05,
"loss": 1.8561,
"step": 8900
},
{
"epoch": 2.54,
"learning_rate": 3.115986483711075e-05,
"loss": 1.8555,
"step": 9000
},
{
"epoch": 2.56,
"learning_rate": 2.9965542017054346e-05,
"loss": 1.8629,
"step": 9100
},
{
"epoch": 2.59,
"learning_rate": 2.8759377225728234e-05,
"loss": 1.8816,
"step": 9200
},
{
"epoch": 2.62,
"learning_rate": 2.7544276219762262e-05,
"loss": 1.8372,
"step": 9300
},
{
"epoch": 2.65,
"learning_rate": 2.632316628391009e-05,
"loss": 1.8493,
"step": 9400
},
{
"epoch": 2.68,
"learning_rate": 2.5098989178963792e-05,
"loss": 1.8573,
"step": 9500
},
{
"epoch": 2.71,
"learning_rate": 2.3874694054794416e-05,
"loss": 1.8587,
"step": 9600
},
{
"epoch": 2.73,
"learning_rate": 2.2653230345591686e-05,
"loss": 1.8501,
"step": 9700
},
{
"epoch": 2.76,
"learning_rate": 2.143754066441865e-05,
"loss": 1.8655,
"step": 9800
},
{
"epoch": 2.79,
"learning_rate": 2.02305537141991e-05,
"loss": 1.8599,
"step": 9900
},
{
"epoch": 2.82,
"learning_rate": 1.9035177232215682e-05,
"loss": 1.8634,
"step": 10000
},
{
"epoch": 2.82,
"eval_gen_len": 20.0,
"eval_loss": 1.9686568975448608,
"eval_rouge1": 22.5506,
"eval_rouge2": 6.0881,
"eval_rougeL": 22.4021,
"eval_rougeLsum": 22.3998,
"eval_runtime": 164.864,
"eval_samples_per_second": 42.508,
"eval_steps_per_second": 2.657,
"step": 10000
},
{
"epoch": 2.85,
"learning_rate": 1.7854290985116118e-05,
"loss": 1.8382,
"step": 10100
},
{
"epoch": 2.87,
"learning_rate": 1.669073983130307e-05,
"loss": 1.8459,
"step": 10200
},
{
"epoch": 2.9,
"learning_rate": 1.5547326867421132e-05,
"loss": 1.8562,
"step": 10300
},
{
"epoch": 2.93,
"learning_rate": 1.4426806675451438e-05,
"loss": 1.8404,
"step": 10400
},
{
"epoch": 2.96,
"learning_rate": 1.3331878686682424e-05,
"loss": 1.8492,
"step": 10500
},
{
"epoch": 2.99,
"learning_rate": 1.2265180678543392e-05,
"loss": 1.8172,
"step": 10600
},
{
"epoch": 3.02,
"learning_rate": 1.1229282419967673e-05,
"loss": 1.7255,
"step": 10700
},
{
"epoch": 3.04,
"learning_rate": 1.022667948059414e-05,
"loss": 1.6157,
"step": 10800
},
{
"epoch": 3.07,
"learning_rate": 9.25978721872139e-06,
"loss": 1.6024,
"step": 10900
},
{
"epoch": 3.1,
"learning_rate": 8.330934962498027e-06,
"loss": 1.5984,
"step": 11000
},
{
"epoch": 3.13,
"learning_rate": 7.442360398367151e-06,
"loss": 1.6075,
"step": 11100
},
{
"epoch": 3.16,
"learning_rate": 6.596204180283686e-06,
"loss": 1.5859,
"step": 11200
},
{
"epoch": 3.18,
"learning_rate": 5.79450477269154e-06,
"loss": 1.5973,
"step": 11300
},
{
"epoch": 3.21,
"learning_rate": 5.039193539684164e-06,
"loss": 1.6132,
"step": 11400
},
{
"epoch": 3.24,
"learning_rate": 4.332090092179324e-06,
"loss": 1.6221,
"step": 11500
},
{
"epoch": 3.27,
"learning_rate": 3.6748979043170137e-06,
"loss": 1.6258,
"step": 11600
},
{
"epoch": 3.3,
"learning_rate": 3.0692002096410223e-06,
"loss": 1.6149,
"step": 11700
},
{
"epoch": 3.33,
"learning_rate": 2.516456186950684e-06,
"loss": 1.624,
"step": 11800
},
{
"epoch": 3.35,
"learning_rate": 2.0179974450113125e-06,
"loss": 1.6138,
"step": 11900
},
{
"epoch": 3.38,
"learning_rate": 1.575024814592075e-06,
"loss": 1.6036,
"step": 12000
},
{
"epoch": 3.38,
"eval_gen_len": 20.0,
"eval_loss": 1.983000636100769,
"eval_rouge1": 22.1876,
"eval_rouge2": 5.6559,
"eval_rougeL": 21.9989,
"eval_rougeLsum": 21.9753,
"eval_runtime": 166.2195,
"eval_samples_per_second": 42.161,
"eval_steps_per_second": 2.635,
"step": 12000
},
{
"epoch": 3.41,
"learning_rate": 1.188605455559466e-06,
"loss": 1.5965,
"step": 12100
},
{
"epoch": 3.44,
"learning_rate": 8.596702859957324e-07,
"loss": 1.5971,
"step": 12200
},
{
"epoch": 3.47,
"learning_rate": 5.890117395356814e-07,
"loss": 1.6098,
"step": 12300
},
{
"epoch": 3.49,
"learning_rate": 3.772818563246123e-07,
"loss": 1.6057,
"step": 12400
},
{
"epoch": 3.52,
"learning_rate": 2.2499071219653263e-07,
"loss": 1.611,
"step": 12500
},
{
"epoch": 3.55,
"learning_rate": 1.3250518985677592e-07,
"loss": 1.6154,
"step": 12600
},
{
"epoch": 3.58,
"learning_rate": 1.0004809502943347e-07,
"loss": 1.6232,
"step": 12700
},
{
"epoch": 3.61,
"learning_rate": 4.999307463393912e-05,
"loss": 1.6383,
"step": 12800
},
{
"epoch": 3.64,
"learning_rate": 4.997113651616064e-05,
"loss": 1.6506,
"step": 12900
},
{
"epoch": 3.66,
"learning_rate": 4.993418683760613e-05,
"loss": 1.6974,
"step": 13000
},
{
"epoch": 3.69,
"learning_rate": 4.988224785538034e-05,
"loss": 1.7072,
"step": 13100
},
{
"epoch": 3.72,
"learning_rate": 4.981535085558401e-05,
"loss": 1.6934,
"step": 13200
},
{
"epoch": 3.75,
"learning_rate": 4.97335361344684e-05,
"loss": 1.6793,
"step": 13300
},
{
"epoch": 3.78,
"learning_rate": 4.963685297416225e-05,
"loss": 1.7101,
"step": 13400
},
{
"epoch": 3.8,
"learning_rate": 4.952535961298611e-05,
"loss": 1.75,
"step": 13500
},
{
"epoch": 3.83,
"learning_rate": 4.939912321037175e-05,
"loss": 1.7315,
"step": 13600
},
{
"epoch": 3.86,
"learning_rate": 4.9258219806407825e-05,
"loss": 1.7341,
"step": 13700
},
{
"epoch": 3.89,
"learning_rate": 4.910273427603616e-05,
"loss": 1.7312,
"step": 13800
},
{
"epoch": 3.92,
"learning_rate": 4.893276027792627e-05,
"loss": 1.7162,
"step": 13900
},
{
"epoch": 3.95,
"learning_rate": 4.874840019805889e-05,
"loss": 1.7558,
"step": 14000
},
{
"epoch": 3.95,
"eval_gen_len": 20.0,
"eval_loss": 2.0177605152130127,
"eval_rouge1": 21.6845,
"eval_rouge2": 5.7382,
"eval_rougeL": 21.4848,
"eval_rougeLsum": 21.4995,
"eval_runtime": 165.4038,
"eval_samples_per_second": 42.369,
"eval_steps_per_second": 2.648,
"step": 14000
},
{
"epoch": 3.97,
"learning_rate": 4.854976508805247e-05,
"loss": 1.767,
"step": 14100
},
{
"epoch": 4.0,
"learning_rate": 4.833697459826981e-05,
"loss": 1.7569,
"step": 14200
},
{
"epoch": 4.03,
"learning_rate": 4.811015690574522e-05,
"loss": 1.5871,
"step": 14300
},
{
"epoch": 4.06,
"learning_rate": 4.7869448636975414e-05,
"loss": 1.6034,
"step": 14400
},
{
"epoch": 4.09,
"learning_rate": 4.761499478562084e-05,
"loss": 1.6308,
"step": 14500
},
{
"epoch": 4.11,
"learning_rate": 4.73469486251669e-05,
"loss": 1.5885,
"step": 14600
},
{
"epoch": 4.14,
"learning_rate": 4.7065471616597775e-05,
"loss": 1.612,
"step": 14700
},
{
"epoch": 4.17,
"learning_rate": 4.677073331113833e-05,
"loss": 1.6088,
"step": 14800
},
{
"epoch": 4.2,
"learning_rate": 4.646291124812277e-05,
"loss": 1.6277,
"step": 14900
},
{
"epoch": 4.23,
"learning_rate": 4.614219084805167e-05,
"loss": 1.6087,
"step": 15000
},
{
"epoch": 4.26,
"learning_rate": 4.580876530090152e-05,
"loss": 1.6462,
"step": 15100
},
{
"epoch": 4.28,
"learning_rate": 4.546283544975434e-05,
"loss": 1.6423,
"step": 15200
},
{
"epoch": 4.31,
"learning_rate": 4.51046096698173e-05,
"loss": 1.642,
"step": 15300
},
{
"epoch": 4.34,
"learning_rate": 4.473430374290533e-05,
"loss": 1.6263,
"step": 15400
},
{
"epoch": 4.37,
"learning_rate": 4.435214072746215e-05,
"loss": 1.6582,
"step": 15500
},
{
"epoch": 4.4,
"learning_rate": 4.3958350824198286e-05,
"loss": 1.6671,
"step": 15600
},
{
"epoch": 4.43,
"learning_rate": 4.355317123742669e-05,
"loss": 1.6473,
"step": 15700
},
{
"epoch": 4.45,
"learning_rate": 4.314106363710374e-05,
"loss": 1.6576,
"step": 15800
},
{
"epoch": 4.48,
"learning_rate": 4.271395127748479e-05,
"loss": 1.633,
"step": 15900
},
{
"epoch": 4.51,
"learning_rate": 4.227619881411398e-05,
"loss": 1.64,
"step": 16000
},
{
"epoch": 4.51,
"eval_gen_len": 20.0,
"eval_loss": 2.032703161239624,
"eval_rouge1": 22.2681,
"eval_rouge2": 5.8023,
"eval_rougeL": 22.0116,
"eval_rougeLsum": 21.9997,
"eval_runtime": 165.398,
"eval_samples_per_second": 42.371,
"eval_steps_per_second": 2.648,
"step": 16000
},
{
"epoch": 4.54,
"learning_rate": 4.18280699327066e-05,
"loss": 1.651,
"step": 16100
},
{
"epoch": 4.57,
"learning_rate": 4.1369834569344205e-05,
"loss": 1.6491,
"step": 16200
},
{
"epoch": 4.59,
"learning_rate": 4.090176874787515e-05,
"loss": 1.6302,
"step": 16300
},
{
"epoch": 4.62,
"learning_rate": 4.042415441364819e-05,
"loss": 1.6376,
"step": 16400
},
{
"epoch": 4.65,
"learning_rate": 3.993727926367911e-05,
"loss": 1.6322,
"step": 16500
},
{
"epoch": 4.68,
"learning_rate": 3.944143657335282e-05,
"loss": 1.6651,
"step": 16600
},
{
"epoch": 4.71,
"learning_rate": 3.8936925019765214e-05,
"loss": 1.6308,
"step": 16700
},
{
"epoch": 4.74,
"learning_rate": 3.842404850181127e-05,
"loss": 1.6703,
"step": 16800
},
{
"epoch": 4.76,
"learning_rate": 3.790311595712772e-05,
"loss": 1.6816,
"step": 16900
},
{
"epoch": 4.79,
"learning_rate": 3.737444117600056e-05,
"loss": 1.6568,
"step": 17000
},
{
"epoch": 4.82,
"learning_rate": 3.6838342612349524e-05,
"loss": 1.6697,
"step": 17100
},
{
"epoch": 4.85,
"learning_rate": 3.629514319190331e-05,
"loss": 1.6352,
"step": 17200
},
{
"epoch": 4.88,
"learning_rate": 3.57451701176813e-05,
"loss": 1.6724,
"step": 17300
},
{
"epoch": 4.9,
"learning_rate": 3.5188754672898564e-05,
"loss": 1.6444,
"step": 17400
},
{
"epoch": 4.93,
"learning_rate": 3.462623202141332e-05,
"loss": 1.6568,
"step": 17500
},
{
"epoch": 4.96,
"learning_rate": 3.4057941005836765e-05,
"loss": 1.6626,
"step": 17600
},
{
"epoch": 4.99,
"learning_rate": 3.34842239434269e-05,
"loss": 1.6246,
"step": 17700
},
{
"epoch": 5.02,
"learning_rate": 3.290542641988946e-05,
"loss": 1.4957,
"step": 17800
},
{
"epoch": 5.05,
"learning_rate": 3.232189708121e-05,
"loss": 1.3941,
"step": 17900
},
{
"epoch": 5.07,
"learning_rate": 3.173398742364255e-05,
"loss": 1.4248,
"step": 18000
},
{
"epoch": 5.07,
"eval_gen_len": 20.0,
"eval_loss": 2.0742380619049072,
"eval_rouge1": 21.8569,
"eval_rouge2": 5.5533,
"eval_rougeL": 21.6308,
"eval_rougeLsum": 21.638,
"eval_runtime": 166.3338,
"eval_samples_per_second": 42.132,
"eval_steps_per_second": 2.633,
"step": 18000
},
{
"epoch": 5.1,
"learning_rate": 3.114205158198149e-05,
"loss": 1.4175,
"step": 18100
},
{
"epoch": 5.13,
"learning_rate": 3.054644611624394e-05,
"loss": 1.3888,
"step": 18200
},
{
"epoch": 5.16,
"learning_rate": 2.9947529796891307e-05,
"loss": 1.4086,
"step": 18300
},
{
"epoch": 5.19,
"learning_rate": 2.9345663388719467e-05,
"loss": 1.3948,
"step": 18400
},
{
"epoch": 5.21,
"learning_rate": 2.8741209433547458e-05,
"loss": 1.4149,
"step": 18500
},
{
"epoch": 5.24,
"learning_rate": 2.8134532031835893e-05,
"loss": 1.4184,
"step": 18600
},
{
"epoch": 5.27,
"learning_rate": 2.7525996623366436e-05,
"loss": 1.43,
"step": 18700
},
{
"epoch": 5.3,
"learning_rate": 2.691596976711453e-05,
"loss": 1.4435,
"step": 18800
},
{
"epoch": 5.33,
"learning_rate": 2.630481892044803e-05,
"loss": 1.4092,
"step": 18900
},
{
"epoch": 5.36,
"learning_rate": 2.5692912217784543e-05,
"loss": 1.4241,
"step": 19000
},
{
"epoch": 5.38,
"learning_rate": 2.5080618248841106e-05,
"loss": 1.414,
"step": 19100
},
{
"epoch": 5.41,
"learning_rate": 2.4468305836609452e-05,
"loss": 1.4265,
"step": 19200
},
{
"epoch": 5.44,
"learning_rate": 2.3856343815190883e-05,
"loss": 1.4061,
"step": 19300
},
{
"epoch": 5.47,
"learning_rate": 2.324510080762437e-05,
"loss": 1.3992,
"step": 19400
},
{
"epoch": 5.5,
"learning_rate": 2.2634945003841808e-05,
"loss": 1.4221,
"step": 19500
},
{
"epoch": 5.52,
"learning_rate": 2.202624393888421e-05,
"loss": 1.3943,
"step": 19600
},
{
"epoch": 5.55,
"learning_rate": 2.1419364271512303e-05,
"loss": 1.4266,
"step": 19700
},
{
"epoch": 5.58,
"learning_rate": 2.0814671563345028e-05,
"loss": 1.4466,
"step": 19800
},
{
"epoch": 5.61,
"learning_rate": 2.021853765291574e-05,
"loss": 1.4312,
"step": 19900
},
{
"epoch": 5.64,
"learning_rate": 1.9619279130307914e-05,
"loss": 1.4263,
"step": 20000
},
{
"epoch": 5.64,
"eval_gen_len": 20.0,
"eval_loss": 2.0796058177948,
"eval_rouge1": 22.037,
"eval_rouge2": 5.454,
"eval_rougeL": 21.7861,
"eval_rougeLsum": 21.7976,
"eval_runtime": 165.631,
"eval_samples_per_second": 42.311,
"eval_steps_per_second": 2.644,
"step": 20000
},
{
"epoch": 5.67,
"learning_rate": 1.902329187088426e-05,
"loss": 1.4114,
"step": 20100
},
{
"epoch": 5.69,
"learning_rate": 1.8430934875084445e-05,
"loss": 1.4068,
"step": 20200
},
{
"epoch": 5.72,
"learning_rate": 1.7842564956613097e-05,
"loss": 1.4212,
"step": 20300
},
{
"epoch": 5.75,
"learning_rate": 1.7258536527508527e-05,
"loss": 1.4179,
"step": 20400
},
{
"epoch": 5.78,
"learning_rate": 1.6679201384658204e-05,
"loss": 1.4263,
"step": 20500
},
{
"epoch": 5.81,
"learning_rate": 1.6104908497889437e-05,
"loss": 1.4062,
"step": 20600
},
{
"epoch": 5.83,
"learning_rate": 1.5536003799763033e-05,
"loss": 1.4168,
"step": 20700
},
{
"epoch": 5.86,
"learning_rate": 1.497282997719642e-05,
"loss": 1.4502,
"step": 20800
},
{
"epoch": 5.89,
"learning_rate": 1.4415726265041886e-05,
"loss": 1.42,
"step": 20900
},
{
"epoch": 5.92,
"learning_rate": 1.3865028241744182e-05,
"loss": 1.421,
"step": 21000
},
{
"epoch": 5.95,
"learning_rate": 1.332106762720061e-05,
"loss": 1.4272,
"step": 21100
},
{
"epoch": 5.98,
"learning_rate": 1.2784172082945395e-05,
"loss": 1.4115,
"step": 21200
},
{
"epoch": 6.0,
"learning_rate": 1.2254665014778574e-05,
"loss": 1.3844,
"step": 21300
},
{
"epoch": 6.03,
"learning_rate": 1.1732865377958439e-05,
"loss": 1.2502,
"step": 21400
},
{
"epoch": 6.06,
"learning_rate": 1.1219087485074895e-05,
"loss": 1.232,
"step": 21500
},
{
"epoch": 6.09,
"learning_rate": 1.0713640816719211e-05,
"loss": 1.2675,
"step": 21600
},
{
"epoch": 6.12,
"learning_rate": 1.021682983506454e-05,
"loss": 1.2537,
"step": 21700
},
{
"epoch": 6.14,
"learning_rate": 9.728953800469257e-06,
"loss": 1.2703,
"step": 21800
},
{
"epoch": 6.17,
"learning_rate": 9.250306591213766e-06,
"loss": 1.2496,
"step": 21900
},
{
"epoch": 6.2,
"learning_rate": 8.781176526479135e-06,
"loss": 1.2478,
"step": 22000
},
{
"epoch": 6.2,
"eval_gen_len": 20.0,
"eval_loss": 2.1144540309906006,
"eval_rouge1": 21.5358,
"eval_rouge2": 5.3838,
"eval_rougeL": 21.3299,
"eval_rougeLsum": 21.3191,
"eval_runtime": 166.7361,
"eval_samples_per_second": 42.03,
"eval_steps_per_second": 2.627,
"step": 22000
},
{
"epoch": 6.23,
"learning_rate": 8.321846192674462e-06,
"loss": 1.2486,
"step": 22100
},
{
"epoch": 6.26,
"learning_rate": 7.877034040486934e-06,
"loss": 1.2412,
"step": 22200
},
{
"epoch": 6.29,
"learning_rate": 7.4380223596522265e-06,
"loss": 1.2552,
"step": 22300
},
{
"epoch": 6.31,
"learning_rate": 7.009619475593796e-06,
"loss": 1.2642,
"step": 22400
},
{
"epoch": 6.34,
"learning_rate": 6.592083442189996e-06,
"loss": 1.265,
"step": 22500
},
{
"epoch": 6.37,
"learning_rate": 6.1856657675346675e-06,
"loss": 1.2631,
"step": 22600
},
{
"epoch": 6.4,
"learning_rate": 5.790611262438083e-06,
"loss": 1.2736,
"step": 22700
},
{
"epoch": 6.43,
"learning_rate": 5.407157892962079e-06,
"loss": 1.2595,
"step": 22800
},
{
"epoch": 6.45,
"learning_rate": 5.035536637078171e-06,
"loss": 1.2661,
"step": 22900
},
{
"epoch": 6.48,
"learning_rate": 4.675971345535108e-06,
"loss": 1.2651,
"step": 23000
},
{
"epoch": 6.51,
"learning_rate": 4.328678607019489e-06,
"loss": 1.2499,
"step": 23100
},
{
"epoch": 6.54,
"learning_rate": 3.993867617690892e-06,
"loss": 1.263,
"step": 23200
},
{
"epoch": 6.57,
"learning_rate": 3.6717400551698886e-06,
"loss": 1.2546,
"step": 23300
},
{
"epoch": 6.6,
"learning_rate": 3.3624899570550363e-06,
"loss": 1.261,
"step": 23400
},
{
"epoch": 6.62,
"learning_rate": 3.066303604041807e-06,
"loss": 1.2418,
"step": 23500
},
{
"epoch": 6.65,
"learning_rate": 2.7833594077141535e-06,
"loss": 1.2601,
"step": 23600
},
{
"epoch": 6.68,
"learning_rate": 2.5138278030759714e-06,
"loss": 1.2397,
"step": 23700
},
{
"epoch": 6.71,
"learning_rate": 2.2578711458874663e-06,
"loss": 1.264,
"step": 23800
},
{
"epoch": 6.74,
"learning_rate": 2.0156436148680858e-06,
"loss": 1.2497,
"step": 23900
},
{
"epoch": 6.76,
"learning_rate": 1.787291118825041e-06,
"loss": 1.2469,
"step": 24000
},
{
"epoch": 6.76,
"eval_gen_len": 20.0,
"eval_loss": 2.113508462905884,
"eval_rouge1": 21.3962,
"eval_rouge2": 5.3698,
"eval_rougeL": 21.1842,
"eval_rougeLsum": 21.159,
"eval_runtime": 166.5668,
"eval_samples_per_second": 42.073,
"eval_steps_per_second": 2.63,
"step": 24000
},
{
"epoch": 6.79,
"learning_rate": 1.5729512087633123e-06,
"loss": 1.2383,
"step": 24100
},
{
"epoch": 6.82,
"learning_rate": 1.372752995030021e-06,
"loss": 1.2619,
"step": 24200
},
{
"epoch": 6.85,
"learning_rate": 1.1868170695432342e-06,
"loss": 1.2536,
"step": 24300
},
{
"epoch": 6.88,
"learning_rate": 1.0152554331518699e-06,
"loss": 1.2514,
"step": 24400
},
{
"epoch": 6.91,
"learning_rate": 8.596702859957324e-07,
"loss": 1.2585,
"step": 24500
},
{
"epoch": 6.93,
"learning_rate": 7.170123718266064e-07,
"loss": 1.2439,
"step": 24600
},
{
"epoch": 6.96,
"learning_rate": 5.890117395356814e-07,
"loss": 1.245,
"step": 24700
},
{
"epoch": 6.99,
"learning_rate": 4.7574549191759186e-07,
"loss": 1.2446,
"step": 24800
},
{
"epoch": 7.02,
"learning_rate": 3.772818563246123e-07,
"loss": 1.2194,
"step": 24900
},
{
"epoch": 7.05,
"learning_rate": 2.936801435690251e-07,
"loss": 1.2058,
"step": 25000
},
{
"epoch": 7.07,
"learning_rate": 2.2499071219653263e-07,
"loss": 1.1916,
"step": 25100
},
{
"epoch": 7.1,
"learning_rate": 1.7171816152852273e-07,
"loss": 1.2186,
"step": 25200
},
{
"epoch": 7.13,
"learning_rate": 1.328184223787534e-07,
"loss": 1.2151,
"step": 25300
},
{
"epoch": 7.16,
"learning_rate": 1.0892786169791435e-07,
"loss": 1.2098,
"step": 25400
},
{
"epoch": 7.19,
"learning_rate": 1.0006087026643128e-07,
"loss": 1.2011,
"step": 25500
},
{
"epoch": 7.22,
"learning_rate": 4.9998444254188996e-05,
"loss": 1.2333,
"step": 25600
},
{
"epoch": 7.24,
"learning_rate": 4.9993146582012154e-05,
"loss": 1.3057,
"step": 25700
},
{
"epoch": 7.27,
"learning_rate": 4.998409256370816e-05,
"loss": 1.2949,
"step": 25800
},
{
"epoch": 7.3,
"learning_rate": 4.997128356277887e-05,
"loss": 1.3107,
"step": 25900
},
{
"epoch": 7.33,
"learning_rate": 4.995472150821271e-05,
"loss": 1.3323,
"step": 26000
},
{
"epoch": 7.33,
"eval_gen_len": 20.0,
"eval_loss": 2.143944501876831,
"eval_rouge1": 21.5245,
"eval_rouge2": 5.444,
"eval_rougeL": 21.312,
"eval_rougeLsum": 21.3015,
"eval_runtime": 166.5992,
"eval_samples_per_second": 42.065,
"eval_steps_per_second": 2.629,
"step": 26000
},
{
"epoch": 7.36,
"learning_rate": 4.9934408894194186e-05,
"loss": 1.3458,
"step": 26100
},
{
"epoch": 7.38,
"learning_rate": 4.991034877972824e-05,
"loss": 1.3346,
"step": 26200
},
{
"epoch": 7.41,
"learning_rate": 4.988254478817961e-05,
"loss": 1.3482,
"step": 26300
},
{
"epoch": 7.44,
"learning_rate": 4.9851001106727165e-05,
"loss": 1.3519,
"step": 26400
},
{
"epoch": 7.47,
"learning_rate": 4.9815722485733305e-05,
"loss": 1.3699,
"step": 26500
},
{
"epoch": 7.5,
"learning_rate": 4.9776714238028576e-05,
"loss": 1.3496,
"step": 26600
},
{
"epoch": 7.53,
"learning_rate": 4.97339822381116e-05,
"loss": 1.3835,
"step": 26700
},
{
"epoch": 7.55,
"learning_rate": 4.968753292126438e-05,
"loss": 1.3604,
"step": 26800
},
{
"epoch": 7.58,
"learning_rate": 4.96373732825832e-05,
"loss": 1.3776,
"step": 26900
},
{
"epoch": 7.61,
"learning_rate": 4.9583510875925124e-05,
"loss": 1.3988,
"step": 27000
},
{
"epoch": 7.64,
"learning_rate": 4.952595381277048e-05,
"loss": 1.3795,
"step": 27100
},
{
"epoch": 7.67,
"learning_rate": 4.946471076100126e-05,
"loss": 1.3823,
"step": 27200
},
{
"epoch": 7.69,
"learning_rate": 4.939979094359581e-05,
"loss": 1.3671,
"step": 27300
},
{
"epoch": 7.72,
"learning_rate": 4.933120413723981e-05,
"loss": 1.4092,
"step": 27400
},
{
"epoch": 7.75,
"learning_rate": 4.925896067085404e-05,
"loss": 1.4064,
"step": 27500
},
{
"epoch": 7.78,
"learning_rate": 4.918307142403879e-05,
"loss": 1.4131,
"step": 27600
},
{
"epoch": 7.81,
"learning_rate": 4.91035478254355e-05,
"loss": 1.4069,
"step": 27700
},
{
"epoch": 7.84,
"learning_rate": 4.902040185100559e-05,
"loss": 1.4013,
"step": 27800
},
{
"epoch": 7.86,
"learning_rate": 4.893364602222699e-05,
"loss": 1.4102,
"step": 27900
},
{
"epoch": 7.89,
"learning_rate": 4.88432934042084e-05,
"loss": 1.4175,
"step": 28000
},
{
"epoch": 7.89,
"eval_gen_len": 20.0,
"eval_loss": 2.1455013751983643,
"eval_rouge1": 21.4843,
"eval_rouge2": 5.5418,
"eval_rougeL": 21.2359,
"eval_rougeLsum": 21.2323,
"eval_runtime": 165.871,
"eval_samples_per_second": 42.25,
"eval_steps_per_second": 2.641,
"step": 28000
},
{
"epoch": 7.92,
"learning_rate": 4.874935760372175e-05,
"loss": 1.407,
"step": 28100
},
{
"epoch": 7.95,
"learning_rate": 4.865284543447741e-05,
"loss": 1.4325,
"step": 28200
},
{
"epoch": 7.98,
"learning_rate": 4.855182171477396e-05,
"loss": 1.4169,
"step": 28300
},
{
"epoch": 8.0,
"learning_rate": 4.844725870716676e-05,
"loss": 1.4046,
"step": 28400
},
{
"epoch": 8.03,
"learning_rate": 4.833917215846013e-05,
"loss": 1.1944,
"step": 28500
},
{
"epoch": 8.06,
"learning_rate": 4.822757834609075e-05,
"loss": 1.2087,
"step": 28600
},
{
"epoch": 8.09,
"learning_rate": 4.811249407567628e-05,
"loss": 1.2257,
"step": 28700
},
{
"epoch": 8.12,
"learning_rate": 4.799393667848454e-05,
"loss": 1.2269,
"step": 28800
},
{
"epoch": 8.15,
"learning_rate": 4.787192400882345e-05,
"loss": 1.244,
"step": 28900
},
{
"epoch": 8.17,
"learning_rate": 4.774647444135227e-05,
"loss": 1.2302,
"step": 29000
},
{
"epoch": 8.2,
"learning_rate": 4.7617606868314405e-05,
"loss": 1.2461,
"step": 29100
},
{
"epoch": 8.23,
"learning_rate": 4.748534069669236e-05,
"loss": 1.2514,
"step": 29200
},
{
"epoch": 8.26,
"learning_rate": 4.7349695845285055e-05,
"loss": 1.2402,
"step": 29300
},
{
"epoch": 8.29,
"learning_rate": 4.721069274170819e-05,
"loss": 1.2567,
"step": 29400
},
{
"epoch": 8.31,
"learning_rate": 4.706835231931785e-05,
"loss": 1.2423,
"step": 29500
},
{
"epoch": 8.34,
"learning_rate": 4.6922696014058083e-05,
"loss": 1.2573,
"step": 29600
},
{
"epoch": 8.37,
"learning_rate": 4.677374576123271e-05,
"loss": 1.253,
"step": 29700
},
{
"epoch": 8.4,
"learning_rate": 4.6621523992201896e-05,
"loss": 1.2469,
"step": 29800
},
{
"epoch": 8.43,
"learning_rate": 4.646605363100417e-05,
"loss": 1.2911,
"step": 29900
},
{
"epoch": 8.46,
"learning_rate": 4.6307358090904045e-05,
"loss": 1.2541,
"step": 30000
},
{
"epoch": 8.46,
"eval_gen_len": 20.0,
"eval_loss": 2.217175006866455,
"eval_rouge1": 20.9025,
"eval_rouge2": 5.1166,
"eval_rougeL": 20.7062,
"eval_rougeLsum": 20.6916,
"eval_runtime": 165.6707,
"eval_samples_per_second": 42.301,
"eval_steps_per_second": 2.644,
"step": 30000
},
{
"epoch": 8.48,
"learning_rate": 4.6147096005744875e-05,
"loss": 1.2806,
"step": 30100
},
{
"epoch": 8.51,
"learning_rate": 4.598205393356323e-05,
"loss": 1.2891,
"step": 30200
},
{
"epoch": 8.54,
"learning_rate": 4.581385957105333e-05,
"loss": 1.2964,
"step": 30300
},
{
"epoch": 8.57,
"learning_rate": 4.564253824766751e-05,
"loss": 1.2694,
"step": 30400
},
{
"epoch": 8.6,
"learning_rate": 4.54681157637669e-05,
"loss": 1.2883,
"step": 30500
},
{
"epoch": 8.62,
"learning_rate": 4.5290618386736e-05,
"loss": 1.3061,
"step": 30600
},
{
"epoch": 8.65,
"learning_rate": 4.51100728470269e-05,
"loss": 1.3022,
"step": 30700
},
{
"epoch": 8.68,
"learning_rate": 4.492650633413379e-05,
"loss": 1.2958,
"step": 30800
},
{
"epoch": 8.71,
"learning_rate": 4.473994649249829e-05,
"loss": 1.2966,
"step": 30900
},
{
"epoch": 8.74,
"learning_rate": 4.455042141734634e-05,
"loss": 1.2956,
"step": 31000
},
{
"epoch": 8.77,
"learning_rate": 4.4357959650457124e-05,
"loss": 1.3057,
"step": 31100
},
{
"epoch": 8.79,
"learning_rate": 4.41625901758648e-05,
"loss": 1.3006,
"step": 31200
},
{
"epoch": 8.82,
"learning_rate": 4.396434241549363e-05,
"loss": 1.311,
"step": 31300
},
{
"epoch": 8.85,
"learning_rate": 4.37632462247271e-05,
"loss": 1.309,
"step": 31400
},
{
"epoch": 8.88,
"learning_rate": 4.355933188791186e-05,
"loss": 1.3324,
"step": 31500
},
{
"epoch": 8.91,
"learning_rate": 4.335263011379698e-05,
"loss": 1.3186,
"step": 31600
},
{
"epoch": 8.93,
"learning_rate": 4.314317203090931e-05,
"loss": 1.3298,
"step": 31700
},
{
"epoch": 8.96,
"learning_rate": 4.2930989182865715e-05,
"loss": 1.3148,
"step": 31800
},
{
"epoch": 8.99,
"learning_rate": 4.271611352362262e-05,
"loss": 1.3318,
"step": 31900
},
{
"epoch": 9.02,
"learning_rate": 4.2498577412663946e-05,
"loss": 1.1331,
"step": 32000
},
{
"epoch": 9.02,
"eval_gen_len": 20.0,
"eval_loss": 2.278179168701172,
"eval_rouge1": 20.9415,
"eval_rouge2": 5.2115,
"eval_rougeL": 20.7647,
"eval_rougeLsum": 20.7806,
"eval_runtime": 166.4103,
"eval_samples_per_second": 42.113,
"eval_steps_per_second": 2.632,
"step": 32000
},
{
"epoch": 9.05,
"learning_rate": 4.2278413610127834e-05,
"loss": 1.0677,
"step": 32100
},
{
"epoch": 9.08,
"learning_rate": 4.205565527187312e-05,
"loss": 1.0712,
"step": 32200
},
{
"epoch": 9.1,
"learning_rate": 4.183033594448618e-05,
"loss": 1.0928,
"step": 32300
},
{
"epoch": 9.13,
"learning_rate": 4.160248956022893e-05,
"loss": 1.0821,
"step": 32400
},
{
"epoch": 9.16,
"learning_rate": 4.137215043192875e-05,
"loss": 1.0988,
"step": 32500
},
{
"epoch": 9.19,
"learning_rate": 4.11393532478111e-05,
"loss": 1.103,
"step": 32600
},
{
"epoch": 9.22,
"learning_rate": 4.0904133066275636e-05,
"loss": 1.1109,
"step": 32700
},
{
"epoch": 9.24,
"learning_rate": 4.066652531061647e-05,
"loss": 1.0833,
"step": 32800
},
{
"epoch": 9.27,
"learning_rate": 4.0426565763687624e-05,
"loss": 1.0964,
"step": 32900
},
{
"epoch": 9.3,
"learning_rate": 4.0184290562514214e-05,
"loss": 1.1168,
"step": 33000
},
{
"epoch": 9.33,
"learning_rate": 3.9939736192850386e-05,
"loss": 1.1122,
"step": 33100
},
{
"epoch": 9.36,
"learning_rate": 3.969293948368467e-05,
"loss": 1.1165,
"step": 33200
},
{
"epoch": 9.39,
"learning_rate": 3.944393760169368e-05,
"loss": 1.1106,
"step": 33300
},
{
"epoch": 9.41,
"learning_rate": 3.9192768045644926e-05,
"loss": 1.1119,
"step": 33400
},
{
"epoch": 9.44,
"learning_rate": 3.8939468640749696e-05,
"loss": 1.1258,
"step": 33500
},
{
"epoch": 9.47,
"learning_rate": 3.868407753296665e-05,
"loss": 1.099,
"step": 33600
},
{
"epoch": 9.5,
"learning_rate": 3.842921766325347e-05,
"loss": 1.1246,
"step": 33700
},
{
"epoch": 9.53,
"learning_rate": 3.816977879359396e-05,
"loss": 1.1175,
"step": 33800
},
{
"epoch": 9.55,
"learning_rate": 3.790836413350648e-05,
"loss": 1.1313,
"step": 33900
},
{
"epoch": 9.58,
"learning_rate": 3.764501305107728e-05,
"loss": 1.1067,
"step": 34000
},
{
"epoch": 9.58,
"eval_gen_len": 20.0,
"eval_loss": 2.27223801612854,
"eval_rouge1": 21.5648,
"eval_rouge2": 5.4445,
"eval_rougeL": 21.3624,
"eval_rougeLsum": 21.3838,
"eval_runtime": 166.4194,
"eval_samples_per_second": 42.11,
"eval_steps_per_second": 2.632,
"step": 34000
},
{
"epoch": 9.61,
"learning_rate": 3.737976520601068e-05,
"loss": 1.1316,
"step": 34100
},
{
"epoch": 9.64,
"learning_rate": 3.7112660543656495e-05,
"loss": 1.137,
"step": 34200
},
{
"epoch": 9.67,
"learning_rate": 3.6843739288994395e-05,
"loss": 1.1319,
"step": 34300
},
{
"epoch": 9.7,
"learning_rate": 3.657304194057615e-05,
"loss": 1.1217,
"step": 34400
},
{
"epoch": 9.72,
"learning_rate": 3.630060926442673e-05,
"loss": 1.1465,
"step": 34500
},
{
"epoch": 9.75,
"learning_rate": 3.602648228790508e-05,
"loss": 1.1396,
"step": 34600
},
{
"epoch": 9.78,
"learning_rate": 3.5750702293525555e-05,
"loss": 1.132,
"step": 34700
},
{
"epoch": 9.81,
"learning_rate": 3.5473310812740905e-05,
"loss": 1.159,
"step": 34800
},
{
"epoch": 9.84,
"learning_rate": 3.519434961968779e-05,
"loss": 1.1484,
"step": 34900
},
{
"epoch": 9.86,
"learning_rate": 3.491386072489581e-05,
"loss": 1.1375,
"step": 35000
},
{
"epoch": 9.89,
"learning_rate": 3.4631886368960794e-05,
"loss": 1.1493,
"step": 35100
},
{
"epoch": 9.92,
"learning_rate": 3.434846901618357e-05,
"loss": 1.1374,
"step": 35200
},
{
"epoch": 9.95,
"learning_rate": 3.406365134817494e-05,
"loss": 1.1569,
"step": 35300
},
{
"epoch": 9.98,
"learning_rate": 3.3777476257428106e-05,
"loss": 1.1441,
"step": 35400
},
{
"epoch": 10.0,
"step": 35480,
"total_flos": 2.3762843604025344e+17,
"train_loss": 1.581618417182931,
"train_runtime": 14068.2834,
"train_samples_per_second": 40.346,
"train_steps_per_second": 2.522
}
],
"max_steps": 35480,
"num_train_epochs": 10,
"total_flos": 2.3762843604025344e+17,
"trial_name": null,
"trial_params": null
}