| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.0, | |
| "global_step": 57432, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.826577517760134e-05, | |
| "loss": 2.376, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.652458559687979e-05, | |
| "loss": 1.9479, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.478339601615824e-05, | |
| "loss": 1.8326, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.304568881459813e-05, | |
| "loss": 1.7996, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.1304499233876586e-05, | |
| "loss": 1.7476, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 3.956330965315504e-05, | |
| "loss": 1.7537, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3.782212007243349e-05, | |
| "loss": 1.7133, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.608093049171194e-05, | |
| "loss": 1.6888, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.434322329015183e-05, | |
| "loss": 1.6906, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.6714558601379395, | |
| "eval_runtime": 47.5103, | |
| "eval_samples_per_second": 9.577, | |
| "eval_steps_per_second": 1.2, | |
| "step": 4786 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.2602033709430284e-05, | |
| "loss": 1.6771, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.086084412870874e-05, | |
| "loss": 1.6319, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.9119654547987185e-05, | |
| "loss": 1.6327, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.737846496726564e-05, | |
| "loss": 1.6368, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.563727538654409e-05, | |
| "loss": 1.6329, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 2.3899568184983983e-05, | |
| "loss": 1.6143, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.2158378604262433e-05, | |
| "loss": 1.6044, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.0417189023540884e-05, | |
| "loss": 1.6295, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.8675999442819337e-05, | |
| "loss": 1.6023, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.6934809862097788e-05, | |
| "loss": 1.6021, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.6306262016296387, | |
| "eval_runtime": 43.7183, | |
| "eval_samples_per_second": 10.408, | |
| "eval_steps_per_second": 1.304, | |
| "step": 9572 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.5193620281376237e-05, | |
| "loss": 1.5842, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.3452430700654689e-05, | |
| "loss": 1.587, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.1714723499094582e-05, | |
| "loss": 1.5566, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 9.973533918373034e-06, | |
| "loss": 1.6022, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 8.232344337651483e-06, | |
| "loss": 1.5785, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.491154756929935e-06, | |
| "loss": 1.5775, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 4.753447555369829e-06, | |
| "loss": 1.5684, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.01225797464828e-06, | |
| "loss": 1.5719, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 1.2710683939267307e-06, | |
| "loss": 1.5819, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.6228220462799072, | |
| "eval_runtime": 42.6413, | |
| "eval_samples_per_second": 10.67, | |
| "eval_steps_per_second": 1.337, | |
| "step": 14358 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 14358, | |
| "total_flos": 1.5006523981824e+16, | |
| "train_loss": 1.6758505107466766, | |
| "train_runtime": 8429.8505, | |
| "train_samples_per_second": 3.406, | |
| "train_steps_per_second": 1.703 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.987811672934949e-05, | |
| "loss": 1.5627, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.944368992895947e-05, | |
| "loss": 1.5686, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.900839253377908e-05, | |
| "loss": 1.5776, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 4.85730951385987e-05, | |
| "loss": 1.5699, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.8137797743418304e-05, | |
| "loss": 1.5476, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 4.770250034823792e-05, | |
| "loss": 1.5448, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 4.726720295305753e-05, | |
| "loss": 1.5656, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 4.683277615266751e-05, | |
| "loss": 1.5694, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.6397478757487115e-05, | |
| "loss": 1.542, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 4.596218136230673e-05, | |
| "loss": 1.5474, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.5994045734405518, | |
| "eval_runtime": 39.2637, | |
| "eval_samples_per_second": 11.588, | |
| "eval_steps_per_second": 1.452, | |
| "step": 19144 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 4.552688396712634e-05, | |
| "loss": 1.5336, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 4.5091586571945955e-05, | |
| "loss": 1.5259, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 4.465628917676557e-05, | |
| "loss": 1.495, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 4.422099178158518e-05, | |
| "loss": 1.5003, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 4.378656498119515e-05, | |
| "loss": 1.4912, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 4.3351267586014767e-05, | |
| "loss": 1.491, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 4.291597019083438e-05, | |
| "loss": 1.4888, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 4.2480672795653994e-05, | |
| "loss": 1.5098, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 4.2046245995263964e-05, | |
| "loss": 1.5011, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.5889110565185547, | |
| "eval_runtime": 42.2475, | |
| "eval_samples_per_second": 10.77, | |
| "eval_steps_per_second": 1.349, | |
| "step": 23930 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 4.161094860008358e-05, | |
| "loss": 1.4888, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 4.117565120490319e-05, | |
| "loss": 1.4664, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 4.0740353809722805e-05, | |
| "loss": 1.455, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 4.030505641454242e-05, | |
| "loss": 1.4491, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 3.986975901936203e-05, | |
| "loss": 1.4543, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 3.9435332218972e-05, | |
| "loss": 1.4588, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 3.9000034823791616e-05, | |
| "loss": 1.446, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 3.856473742861123e-05, | |
| "loss": 1.4386, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.812944003343084e-05, | |
| "loss": 1.4473, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 3.7694142638250456e-05, | |
| "loss": 1.4724, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.5893619060516357, | |
| "eval_runtime": 39.6175, | |
| "eval_samples_per_second": 11.485, | |
| "eval_steps_per_second": 1.439, | |
| "step": 28716 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 3.725884524307007e-05, | |
| "loss": 1.4439, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 3.6823547847889676e-05, | |
| "loss": 1.4021, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 3.6388250452709297e-05, | |
| "loss": 1.4147, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 3.595382365231927e-05, | |
| "loss": 1.4157, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 3.551852625713888e-05, | |
| "loss": 1.417, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 3.508322886195849e-05, | |
| "loss": 1.4094, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 3.464793146677811e-05, | |
| "loss": 1.4209, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 3.4212634071597714e-05, | |
| "loss": 1.415, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 3.377907786599805e-05, | |
| "loss": 1.4128, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.334378047081766e-05, | |
| "loss": 1.414, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.5933945178985596, | |
| "eval_runtime": 39.5291, | |
| "eval_samples_per_second": 11.51, | |
| "eval_steps_per_second": 1.442, | |
| "step": 33502 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 3.2908483075637276e-05, | |
| "loss": 1.358, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 3.247318568045689e-05, | |
| "loss": 1.3821, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 3.20378882852765e-05, | |
| "loss": 1.377, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 3.1602590890096116e-05, | |
| "loss": 1.3772, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 3.116729349491573e-05, | |
| "loss": 1.3777, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 3.0731996099735337e-05, | |
| "loss": 1.3672, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 3.0298439894135678e-05, | |
| "loss": 1.3949, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 2.9864013093745645e-05, | |
| "loss": 1.3626, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 2.9428715698565262e-05, | |
| "loss": 1.3939, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.6080235242843628, | |
| "eval_runtime": 39.1335, | |
| "eval_samples_per_second": 11.627, | |
| "eval_steps_per_second": 1.457, | |
| "step": 38288 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 2.8993418303384872e-05, | |
| "loss": 1.3667, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 2.855812090820449e-05, | |
| "loss": 1.336, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 2.81228235130241e-05, | |
| "loss": 1.3212, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 2.768752611784371e-05, | |
| "loss": 1.3235, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 2.7252228722663326e-05, | |
| "loss": 1.3649, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 2.6816931327482936e-05, | |
| "loss": 1.3423, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 2.6381633932302553e-05, | |
| "loss": 1.3396, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 2.5946336537122163e-05, | |
| "loss": 1.3528, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 2.5511909736732137e-05, | |
| "loss": 1.3488, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 2.5076612341551747e-05, | |
| "loss": 1.3448, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.626845121383667, | |
| "eval_runtime": 39.7529, | |
| "eval_samples_per_second": 11.446, | |
| "eval_steps_per_second": 1.434, | |
| "step": 43074 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 2.4641314946371364e-05, | |
| "loss": 1.3208, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 2.4206017551190974e-05, | |
| "loss": 1.3107, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 2.377159075080095e-05, | |
| "loss": 1.3029, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 2.3336293355620562e-05, | |
| "loss": 1.3034, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "learning_rate": 2.2900995960440175e-05, | |
| "loss": 1.3041, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 2.2466569160050146e-05, | |
| "loss": 1.3025, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 2.203127176486976e-05, | |
| "loss": 1.3186, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 2.1595974369689373e-05, | |
| "loss": 1.3185, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "learning_rate": 2.1160676974508987e-05, | |
| "loss": 1.3227, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.6497271060943604, | |
| "eval_runtime": 39.0443, | |
| "eval_samples_per_second": 11.653, | |
| "eval_steps_per_second": 1.46, | |
| "step": 47860 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.07253795793286e-05, | |
| "loss": 1.3038, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 2.029008218414821e-05, | |
| "loss": 1.2825, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 1.9854784788967824e-05, | |
| "loss": 1.2657, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 1.9419487393787437e-05, | |
| "loss": 1.2851, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 10.45, | |
| "learning_rate": 1.898418999860705e-05, | |
| "loss": 1.294, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 1.8549763198217025e-05, | |
| "loss": 1.2857, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 10.66, | |
| "learning_rate": 1.8114465803036635e-05, | |
| "loss": 1.2665, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 1.7679168407856248e-05, | |
| "loss": 1.273, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 1.724387101267586e-05, | |
| "loss": 1.2838, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 1.6809444212285836e-05, | |
| "loss": 1.2873, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.673952341079712, | |
| "eval_runtime": 38.6787, | |
| "eval_samples_per_second": 11.764, | |
| "eval_steps_per_second": 1.474, | |
| "step": 52646 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 1.637414681710545e-05, | |
| "loss": 1.2753, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 1.593884942192506e-05, | |
| "loss": 1.2625, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "learning_rate": 1.5503552026744673e-05, | |
| "loss": 1.2471, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 1.5068254631564285e-05, | |
| "loss": 1.2523, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 1.463382783117426e-05, | |
| "loss": 1.2534, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 1.4199401030784231e-05, | |
| "loss": 1.2606, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 1.3764103635603845e-05, | |
| "loss": 1.244, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 11.81, | |
| "learning_rate": 1.332967683521382e-05, | |
| "loss": 1.2577, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 11.91, | |
| "learning_rate": 1.2894379440033432e-05, | |
| "loss": 1.2671, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.7010221481323242, | |
| "eval_runtime": 38.3243, | |
| "eval_samples_per_second": 11.872, | |
| "eval_steps_per_second": 1.487, | |
| "step": 57432 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 57432, | |
| "total_flos": 6.0026095927296e+16, | |
| "train_loss": 1.0414019944791026, | |
| "train_runtime": 23906.9056, | |
| "train_samples_per_second": 4.805, | |
| "train_steps_per_second": 2.402 | |
| } | |
| ], | |
| "max_steps": 57432, | |
| "num_train_epochs": 12, | |
| "total_flos": 6.0026095927296e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |