| { | |
| "best_metric": 0.8179661130234268, | |
| "best_model_checkpoint": "outputs/t5-mini/kw/qqp_21/checkpoint-45484", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 56855, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 78.46215057373047, | |
| "learning_rate": 4.956028493536189e-05, | |
| "loss": 20.096, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 69.02784729003906, | |
| "learning_rate": 4.9120569870723775e-05, | |
| "loss": 17.1696, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 73.71961212158203, | |
| "learning_rate": 4.868085480608566e-05, | |
| "loss": 15.9207, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 56.62882995605469, | |
| "learning_rate": 4.824113974144755e-05, | |
| "loss": 15.5129, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 109.83588409423828, | |
| "learning_rate": 4.780142467680943e-05, | |
| "loss": 14.5543, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 63.06591796875, | |
| "learning_rate": 4.736170961217132e-05, | |
| "loss": 14.1194, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 97.26592254638672, | |
| "learning_rate": 4.6921994547533196e-05, | |
| "loss": 13.8172, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 59.0395393371582, | |
| "learning_rate": 4.6482279482895086e-05, | |
| "loss": 13.9323, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 50.412662506103516, | |
| "learning_rate": 4.6042564418256975e-05, | |
| "loss": 13.5055, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 61.14406967163086, | |
| "learning_rate": 4.560284935361886e-05, | |
| "loss": 13.5669, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 87.61188507080078, | |
| "learning_rate": 4.516313428898074e-05, | |
| "loss": 13.0135, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 47.9595832824707, | |
| "learning_rate": 4.472341922434263e-05, | |
| "loss": 13.009, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 132.860107421875, | |
| "learning_rate": 4.428370415970451e-05, | |
| "loss": 12.725, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 107.28248596191406, | |
| "learning_rate": 4.38439890950664e-05, | |
| "loss": 12.5346, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 124.45763397216797, | |
| "learning_rate": 4.3404274030428286e-05, | |
| "loss": 12.581, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 70.87162780761719, | |
| "learning_rate": 4.296455896579017e-05, | |
| "loss": 12.4756, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 79.66390228271484, | |
| "learning_rate": 4.252484390115206e-05, | |
| "loss": 12.2702, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 64.87586212158203, | |
| "learning_rate": 4.208512883651394e-05, | |
| "loss": 11.9856, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 90.74152374267578, | |
| "learning_rate": 4.1645413771875824e-05, | |
| "loss": 12.1635, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 84.467041015625, | |
| "learning_rate": 4.1205698707237714e-05, | |
| "loss": 12.0625, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 76.96673583984375, | |
| "learning_rate": 4.0765983642599596e-05, | |
| "loss": 11.9467, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 82.92093658447266, | |
| "learning_rate": 4.0326268577961486e-05, | |
| "loss": 11.7273, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.7798604912559093, | |
| "eval_f1": 0.7798604912559093, | |
| "eval_loss": 0.3752712905406952, | |
| "eval_runtime": 64.3854, | |
| "eval_samples_per_second": 627.937, | |
| "eval_steps_per_second": 2.454, | |
| "step": 11371 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 92.10366821289062, | |
| "learning_rate": 3.988655351332337e-05, | |
| "loss": 11.3212, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 47.02320861816406, | |
| "learning_rate": 3.944683844868525e-05, | |
| "loss": 11.4345, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 42.389854431152344, | |
| "learning_rate": 3.900712338404714e-05, | |
| "loss": 11.4321, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 70.61559295654297, | |
| "learning_rate": 3.8567408319409024e-05, | |
| "loss": 11.2646, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 102.52373504638672, | |
| "learning_rate": 3.812769325477091e-05, | |
| "loss": 11.3316, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 53.84244918823242, | |
| "learning_rate": 3.76879781901328e-05, | |
| "loss": 11.3191, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 62.559635162353516, | |
| "learning_rate": 3.724826312549468e-05, | |
| "loss": 11.1774, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 77.8012924194336, | |
| "learning_rate": 3.680854806085657e-05, | |
| "loss": 11.3886, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 59.538143157958984, | |
| "learning_rate": 3.636883299621845e-05, | |
| "loss": 10.97, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 72.5059814453125, | |
| "learning_rate": 3.5929117931580335e-05, | |
| "loss": 10.9806, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 50.77811050415039, | |
| "learning_rate": 3.5489402866942224e-05, | |
| "loss": 11.145, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 73.23597717285156, | |
| "learning_rate": 3.504968780230411e-05, | |
| "loss": 10.9792, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 60.301700592041016, | |
| "learning_rate": 3.4609972737666e-05, | |
| "loss": 10.7031, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 63.75108337402344, | |
| "learning_rate": 3.417025767302788e-05, | |
| "loss": 11.1893, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 67.40409088134766, | |
| "learning_rate": 3.373054260838976e-05, | |
| "loss": 10.6892, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 42.0967903137207, | |
| "learning_rate": 3.329082754375165e-05, | |
| "loss": 10.926, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 97.28190612792969, | |
| "learning_rate": 3.2851112479113535e-05, | |
| "loss": 10.6576, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 71.09336853027344, | |
| "learning_rate": 3.241139741447542e-05, | |
| "loss": 10.5294, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 52.61311721801758, | |
| "learning_rate": 3.197168234983731e-05, | |
| "loss": 10.7066, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 70.7719497680664, | |
| "learning_rate": 3.153196728519919e-05, | |
| "loss": 10.6082, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 57.35883331298828, | |
| "learning_rate": 3.109225222056108e-05, | |
| "loss": 10.8771, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 53.73151397705078, | |
| "learning_rate": 3.065253715592296e-05, | |
| "loss": 10.7454, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 75.67459106445312, | |
| "learning_rate": 3.021282209128485e-05, | |
| "loss": 10.8851, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.8087286404461593, | |
| "eval_f1": 0.8087286404461593, | |
| "eval_loss": 0.3425763249397278, | |
| "eval_runtime": 33.7764, | |
| "eval_samples_per_second": 1196.991, | |
| "eval_steps_per_second": 4.678, | |
| "step": 22742 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 68.96992492675781, | |
| "learning_rate": 2.9773107026646735e-05, | |
| "loss": 10.4706, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 80.8647232055664, | |
| "learning_rate": 2.933339196200862e-05, | |
| "loss": 10.2234, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 96.2059326171875, | |
| "learning_rate": 2.8893676897370504e-05, | |
| "loss": 10.3815, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 86.25205993652344, | |
| "learning_rate": 2.845396183273239e-05, | |
| "loss": 10.1995, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 64.76643371582031, | |
| "learning_rate": 2.8014246768094277e-05, | |
| "loss": 10.4213, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 59.44820022583008, | |
| "learning_rate": 2.7574531703456163e-05, | |
| "loss": 9.9771, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 64.23847198486328, | |
| "learning_rate": 2.7134816638818046e-05, | |
| "loss": 10.1808, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 74.40206146240234, | |
| "learning_rate": 2.6695101574179932e-05, | |
| "loss": 10.091, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 47.61061096191406, | |
| "learning_rate": 2.625538650954182e-05, | |
| "loss": 10.142, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 74.26252746582031, | |
| "learning_rate": 2.5815671444903705e-05, | |
| "loss": 10.104, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 41.37887191772461, | |
| "learning_rate": 2.5375956380265588e-05, | |
| "loss": 10.4493, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 84.93189239501953, | |
| "learning_rate": 2.4936241315627474e-05, | |
| "loss": 10.0927, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 50.18044662475586, | |
| "learning_rate": 2.449652625098936e-05, | |
| "loss": 10.1145, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 62.862892150878906, | |
| "learning_rate": 2.4056811186351246e-05, | |
| "loss": 10.0547, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 68.27816772460938, | |
| "learning_rate": 2.3617096121713133e-05, | |
| "loss": 10.0757, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 62.146175384521484, | |
| "learning_rate": 2.317738105707502e-05, | |
| "loss": 10.0992, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 85.38493347167969, | |
| "learning_rate": 2.27376659924369e-05, | |
| "loss": 9.7253, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 70.77899932861328, | |
| "learning_rate": 2.2297950927798788e-05, | |
| "loss": 10.0789, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 76.08915710449219, | |
| "learning_rate": 2.1858235863160674e-05, | |
| "loss": 9.9853, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 80.961669921875, | |
| "learning_rate": 2.141852079852256e-05, | |
| "loss": 9.8662, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 56.63176727294922, | |
| "learning_rate": 2.0978805733884443e-05, | |
| "loss": 9.9995, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 78.24808502197266, | |
| "learning_rate": 2.053909066924633e-05, | |
| "loss": 9.9561, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 82.20337677001953, | |
| "learning_rate": 2.0099375604608216e-05, | |
| "loss": 9.8649, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.8170599269424424, | |
| "eval_f1": 0.8170599269424424, | |
| "eval_loss": 0.32534557580947876, | |
| "eval_runtime": 33.5119, | |
| "eval_samples_per_second": 1206.439, | |
| "eval_steps_per_second": 4.715, | |
| "step": 34113 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 98.00669860839844, | |
| "learning_rate": 1.9659660539970102e-05, | |
| "loss": 9.7264, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 80.73302459716797, | |
| "learning_rate": 1.9219945475331985e-05, | |
| "loss": 9.6493, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 61.9378776550293, | |
| "learning_rate": 1.878023041069387e-05, | |
| "loss": 9.6228, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 45.61129379272461, | |
| "learning_rate": 1.8340515346055757e-05, | |
| "loss": 9.5834, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 68.07831573486328, | |
| "learning_rate": 1.7900800281417643e-05, | |
| "loss": 9.7149, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 107.45940399169922, | |
| "learning_rate": 1.7461085216779526e-05, | |
| "loss": 9.7818, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 57.60527801513672, | |
| "learning_rate": 1.7021370152141413e-05, | |
| "loss": 9.476, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "grad_norm": 93.6783447265625, | |
| "learning_rate": 1.65816550875033e-05, | |
| "loss": 9.7492, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 65.9582748413086, | |
| "learning_rate": 1.6141940022865185e-05, | |
| "loss": 9.7112, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "grad_norm": 76.75605010986328, | |
| "learning_rate": 1.5702224958227068e-05, | |
| "loss": 9.6361, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "grad_norm": 112.33428955078125, | |
| "learning_rate": 1.5262509893588954e-05, | |
| "loss": 9.6922, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 48.53580093383789, | |
| "learning_rate": 1.482279482895084e-05, | |
| "loss": 9.4253, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 75.4821548461914, | |
| "learning_rate": 1.4383079764312727e-05, | |
| "loss": 9.6344, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 50.351139068603516, | |
| "learning_rate": 1.3943364699674611e-05, | |
| "loss": 9.6358, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 161.5642547607422, | |
| "learning_rate": 1.3503649635036497e-05, | |
| "loss": 9.6302, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 42.763099670410156, | |
| "learning_rate": 1.3063934570398382e-05, | |
| "loss": 9.2827, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 84.87345123291016, | |
| "learning_rate": 1.2624219505760268e-05, | |
| "loss": 9.6216, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "grad_norm": 88.34630584716797, | |
| "learning_rate": 1.2184504441122153e-05, | |
| "loss": 9.4236, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 51.73760223388672, | |
| "learning_rate": 1.1744789376484039e-05, | |
| "loss": 9.5094, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 107.87085723876953, | |
| "learning_rate": 1.1305074311845925e-05, | |
| "loss": 9.5279, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "grad_norm": 121.04542541503906, | |
| "learning_rate": 1.086535924720781e-05, | |
| "loss": 9.5586, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 59.98193359375, | |
| "learning_rate": 1.0425644182569696e-05, | |
| "loss": 9.4168, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.8179661130234268, | |
| "eval_f1": 0.8179661130234268, | |
| "eval_loss": 0.3207918703556061, | |
| "eval_runtime": 33.3558, | |
| "eval_samples_per_second": 1212.083, | |
| "eval_steps_per_second": 4.737, | |
| "step": 45484 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 66.1643295288086, | |
| "learning_rate": 9.98592911793158e-06, | |
| "loss": 9.5956, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 89.93461608886719, | |
| "learning_rate": 9.546214053293467e-06, | |
| "loss": 9.5001, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 50.218685150146484, | |
| "learning_rate": 9.106498988655351e-06, | |
| "loss": 9.3642, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 91.35980224609375, | |
| "learning_rate": 8.666783924017237e-06, | |
| "loss": 9.1908, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "grad_norm": 64.65238952636719, | |
| "learning_rate": 8.227068859379122e-06, | |
| "loss": 9.5462, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 102.76400756835938, | |
| "learning_rate": 7.787353794741008e-06, | |
| "loss": 9.134, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "grad_norm": 72.09874725341797, | |
| "learning_rate": 7.347638730102893e-06, | |
| "loss": 9.1759, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 77.65538024902344, | |
| "learning_rate": 6.907923665464779e-06, | |
| "loss": 9.3328, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "grad_norm": 62.97948455810547, | |
| "learning_rate": 6.468208600826664e-06, | |
| "loss": 9.207, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 91.46334838867188, | |
| "learning_rate": 6.02849353618855e-06, | |
| "loss": 9.0635, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "grad_norm": 70.92437744140625, | |
| "learning_rate": 5.588778471550435e-06, | |
| "loss": 9.3683, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 73.28337097167969, | |
| "learning_rate": 5.149063406912321e-06, | |
| "loss": 9.4556, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "grad_norm": 55.88742446899414, | |
| "learning_rate": 4.709348342274207e-06, | |
| "loss": 9.1352, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 66.97167205810547, | |
| "learning_rate": 4.269633277636092e-06, | |
| "loss": 8.9607, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "grad_norm": 62.52705001831055, | |
| "learning_rate": 3.8299182129979776e-06, | |
| "loss": 9.1861, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "grad_norm": 47.30684280395508, | |
| "learning_rate": 3.3902031483598625e-06, | |
| "loss": 9.1937, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 62.73200988769531, | |
| "learning_rate": 2.9504880837217483e-06, | |
| "loss": 9.1071, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 124.52106475830078, | |
| "learning_rate": 2.5107730190836337e-06, | |
| "loss": 9.174, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 93.40811157226562, | |
| "learning_rate": 2.0710579544455195e-06, | |
| "loss": 9.4766, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "grad_norm": 59.61379623413086, | |
| "learning_rate": 1.6313428898074047e-06, | |
| "loss": 9.3928, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 104.62787628173828, | |
| "learning_rate": 1.1916278251692903e-06, | |
| "loss": 9.2219, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 79.24739074707031, | |
| "learning_rate": 7.519127605311758e-07, | |
| "loss": 9.2934, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "grad_norm": 51.52265548706055, | |
| "learning_rate": 3.121976958930613e-07, | |
| "loss": 9.2123, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.8173628720138796, | |
| "eval_f1": 0.8173628720138796, | |
| "eval_loss": 0.32361531257629395, | |
| "eval_runtime": 70.6423, | |
| "eval_samples_per_second": 572.32, | |
| "eval_steps_per_second": 2.237, | |
| "step": 56855 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 56855, | |
| "total_flos": 0.0, | |
| "train_loss": 7.986851407397381, | |
| "train_runtime": 51406.9762, | |
| "train_samples_per_second": 35.389, | |
| "train_steps_per_second": 1.106 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 56855, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |