| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 28210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.822757887274017e-05, | |
| "loss": 0.1767, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.645515774548033e-05, | |
| "loss": 0.1359, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.46827366182205e-05, | |
| "loss": 0.133, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 9.291031549096066e-05, | |
| "loss": 0.1302, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.113789436370082e-05, | |
| "loss": 0.1302, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 8.936547323644098e-05, | |
| "loss": 0.1298, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.759305210918115e-05, | |
| "loss": 0.1253, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 8.582063098192131e-05, | |
| "loss": 0.1288, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 8.404820985466147e-05, | |
| "loss": 0.1206, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 8.227578872740164e-05, | |
| "loss": 0.1224, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 8.05033676001418e-05, | |
| "loss": 0.1237, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.114857017993927, | |
| "eval_runtime": 83.1109, | |
| "eval_samples_per_second": 33.943, | |
| "eval_steps_per_second": 8.495, | |
| "step": 5642 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 7.873094647288196e-05, | |
| "loss": 0.1154, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.695852534562212e-05, | |
| "loss": 0.1071, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.518610421836229e-05, | |
| "loss": 0.1059, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 7.341368309110245e-05, | |
| "loss": 0.1057, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 7.164126196384261e-05, | |
| "loss": 0.1067, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 6.986884083658278e-05, | |
| "loss": 0.1116, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 6.809641970932294e-05, | |
| "loss": 0.1056, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 6.632399858206309e-05, | |
| "loss": 0.1061, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 6.455157745480327e-05, | |
| "loss": 0.104, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.277915632754343e-05, | |
| "loss": 0.1036, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.100673520028359e-05, | |
| "loss": 0.1085, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.10869502276182175, | |
| "eval_runtime": 83.1785, | |
| "eval_samples_per_second": 33.915, | |
| "eval_steps_per_second": 8.488, | |
| "step": 11284 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5.9234314073023755e-05, | |
| "loss": 0.0972, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.746189294576392e-05, | |
| "loss": 0.0905, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 5.568947181850408e-05, | |
| "loss": 0.0906, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 5.3917050691244244e-05, | |
| "loss": 0.0918, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 5.21446295639844e-05, | |
| "loss": 0.0916, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 5.0372208436724564e-05, | |
| "loss": 0.088, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.8599787309464734e-05, | |
| "loss": 0.0928, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 4.68273661822049e-05, | |
| "loss": 0.0917, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.505494505494506e-05, | |
| "loss": 0.0883, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.328252392768522e-05, | |
| "loss": 0.0946, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 4.151010280042538e-05, | |
| "loss": 0.0936, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.10643293708562851, | |
| "eval_runtime": 83.1131, | |
| "eval_samples_per_second": 33.942, | |
| "eval_steps_per_second": 8.494, | |
| "step": 16926 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.973768167316554e-05, | |
| "loss": 0.0888, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.7965260545905705e-05, | |
| "loss": 0.0783, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 3.6192839418645875e-05, | |
| "loss": 0.079, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 3.442041829138604e-05, | |
| "loss": 0.0813, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 3.26479971641262e-05, | |
| "loss": 0.0792, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.087557603686636e-05, | |
| "loss": 0.079, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.9103154909606524e-05, | |
| "loss": 0.0777, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 2.7330733782346684e-05, | |
| "loss": 0.08, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 2.5558312655086853e-05, | |
| "loss": 0.0795, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 2.3785891527827013e-05, | |
| "loss": 0.0807, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 2.2013470400567176e-05, | |
| "loss": 0.0807, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 2.024104927330734e-05, | |
| "loss": 0.0765, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.10565190017223358, | |
| "eval_runtime": 83.1461, | |
| "eval_samples_per_second": 33.928, | |
| "eval_steps_per_second": 8.491, | |
| "step": 22568 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 1.8468628146047502e-05, | |
| "loss": 0.0742, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.6696207018787665e-05, | |
| "loss": 0.0711, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.4923785891527828e-05, | |
| "loss": 0.0724, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.315136476426799e-05, | |
| "loss": 0.0689, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 1.1378943637008153e-05, | |
| "loss": 0.07, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 9.606522509748318e-06, | |
| "loss": 0.071, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 7.834101382488479e-06, | |
| "loss": 0.0713, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 6.061680255228642e-06, | |
| "loss": 0.0707, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 4.289259127968805e-06, | |
| "loss": 0.0751, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 2.5168380007089687e-06, | |
| "loss": 0.0696, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7.444168734491315e-07, | |
| "loss": 0.0698, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.1071661189198494, | |
| "eval_runtime": 83.1775, | |
| "eval_samples_per_second": 33.915, | |
| "eval_steps_per_second": 8.488, | |
| "step": 28210 | |
| } | |
| ], | |
| "max_steps": 28210, | |
| "num_train_epochs": 5, | |
| "total_flos": 5.89605470208e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |