| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 26.29440168284171, | |
| "eval_steps": 50000, | |
| "global_step": 1650000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 1.1412, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.011365294456482, | |
| "eval_runtime": 2006.5534, | |
| "eval_samples_per_second": 111.193, | |
| "eval_steps_per_second": 1.738, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.992035346050262e-05, | |
| "loss": 1.0123, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_loss": 0.9693423509597778, | |
| "eval_runtime": 1995.4615, | |
| "eval_samples_per_second": 111.811, | |
| "eval_steps_per_second": 1.747, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 4.992034708610221e-05, | |
| "loss": 0.9754, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_loss": 0.9472731351852417, | |
| "eval_runtime": 1997.2157, | |
| "eval_samples_per_second": 111.713, | |
| "eval_steps_per_second": 1.746, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.992034549250211e-05, | |
| "loss": 0.9539, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_loss": 0.9325647354125977, | |
| "eval_runtime": 2015.1208, | |
| "eval_samples_per_second": 110.72, | |
| "eval_steps_per_second": 1.73, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 4.992035027330242e-05, | |
| "loss": 0.9387, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "eval_loss": 0.9212433099746704, | |
| "eval_runtime": 2011.4269, | |
| "eval_samples_per_second": 110.924, | |
| "eval_steps_per_second": 1.734, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 4.992035027330242e-05, | |
| "loss": 0.9243, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_loss": 0.9138051271438599, | |
| "eval_runtime": 2011.3973, | |
| "eval_samples_per_second": 110.925, | |
| "eval_steps_per_second": 1.734, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.9144, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "eval_loss": 0.9093130826950073, | |
| "eval_runtime": 1998.4573, | |
| "eval_samples_per_second": 111.644, | |
| "eval_steps_per_second": 1.745, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 0.906, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "eval_loss": 0.9041373133659363, | |
| "eval_runtime": 1998.7351, | |
| "eval_samples_per_second": 111.628, | |
| "eval_steps_per_second": 1.745, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 0.8994, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "eval_loss": 0.9003444910049438, | |
| "eval_runtime": 1982.6092, | |
| "eval_samples_per_second": 112.536, | |
| "eval_steps_per_second": 1.759, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.8933, | |
| "step": 500000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "eval_loss": 0.8956149220466614, | |
| "eval_runtime": 2002.7479, | |
| "eval_samples_per_second": 111.404, | |
| "eval_steps_per_second": 1.741, | |
| "step": 500000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 0.8856, | |
| "step": 550000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "eval_loss": 0.8930546045303345, | |
| "eval_runtime": 1996.5839, | |
| "eval_samples_per_second": 111.748, | |
| "eval_steps_per_second": 1.746, | |
| "step": 550000 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 4.992035346050262e-05, | |
| "loss": 0.8802, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "eval_loss": 0.89084392786026, | |
| "eval_runtime": 1991.4984, | |
| "eval_samples_per_second": 112.034, | |
| "eval_steps_per_second": 1.751, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "learning_rate": 4.992035346050262e-05, | |
| "loss": 0.8763, | |
| "step": 650000 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "eval_loss": 0.8895950317382812, | |
| "eval_runtime": 2006.0064, | |
| "eval_samples_per_second": 111.223, | |
| "eval_steps_per_second": 1.738, | |
| "step": 650000 | |
| }, | |
| { | |
| "epoch": 11.16, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.8725, | |
| "step": 700000 | |
| }, | |
| { | |
| "epoch": 11.16, | |
| "eval_loss": 0.8886296153068542, | |
| "eval_runtime": 2027.9335, | |
| "eval_samples_per_second": 110.021, | |
| "eval_steps_per_second": 1.719, | |
| "step": 700000 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.8688, | |
| "step": 750000 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "eval_loss": 0.885003924369812, | |
| "eval_runtime": 1989.8507, | |
| "eval_samples_per_second": 112.127, | |
| "eval_steps_per_second": 1.752, | |
| "step": 750000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.8628, | |
| "step": 800000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "eval_loss": 0.8833887577056885, | |
| "eval_runtime": 2010.1701, | |
| "eval_samples_per_second": 110.993, | |
| "eval_steps_per_second": 1.735, | |
| "step": 800000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8599, | |
| "step": 850000 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "eval_loss": 0.883805513381958, | |
| "eval_runtime": 2005.9725, | |
| "eval_samples_per_second": 111.225, | |
| "eval_steps_per_second": 1.738, | |
| "step": 850000 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8572, | |
| "step": 900000 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "eval_loss": 0.8837567567825317, | |
| "eval_runtime": 2005.535, | |
| "eval_samples_per_second": 111.25, | |
| "eval_steps_per_second": 1.739, | |
| "step": 900000 | |
| }, | |
| { | |
| "epoch": 15.14, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8548, | |
| "step": 950000 | |
| }, | |
| { | |
| "epoch": 15.14, | |
| "eval_loss": 0.8825677037239075, | |
| "eval_runtime": 1984.894, | |
| "eval_samples_per_second": 112.407, | |
| "eval_steps_per_second": 1.757, | |
| "step": 950000 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "learning_rate": 4.992034549250211e-05, | |
| "loss": 0.8502, | |
| "step": 1000000 | |
| }, | |
| { | |
| "epoch": 15.94, | |
| "eval_loss": 0.8808427453041077, | |
| "eval_runtime": 2006.4913, | |
| "eval_samples_per_second": 111.197, | |
| "eval_steps_per_second": 1.738, | |
| "step": 1000000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 4.992034708610221e-05, | |
| "loss": 0.8471, | |
| "step": 1050000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "eval_loss": 0.8812766075134277, | |
| "eval_runtime": 1998.6292, | |
| "eval_samples_per_second": 111.634, | |
| "eval_steps_per_second": 1.745, | |
| "step": 1050000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 4.992034708610221e-05, | |
| "loss": 0.8427, | |
| "step": 1100000 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "eval_loss": 0.8817498683929443, | |
| "eval_runtime": 1994.6872, | |
| "eval_samples_per_second": 111.855, | |
| "eval_steps_per_second": 1.748, | |
| "step": 1100000 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 4.992034549250211e-05, | |
| "loss": 0.841, | |
| "step": 1150000 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "eval_loss": 0.8802331686019897, | |
| "eval_runtime": 1993.2844, | |
| "eval_samples_per_second": 111.933, | |
| "eval_steps_per_second": 1.749, | |
| "step": 1150000 | |
| }, | |
| { | |
| "epoch": 19.12, | |
| "learning_rate": 4.992034549250211e-05, | |
| "loss": 0.8399, | |
| "step": 1200000 | |
| }, | |
| { | |
| "epoch": 19.12, | |
| "eval_loss": 0.8813353180885315, | |
| "eval_runtime": 2003.661, | |
| "eval_samples_per_second": 111.354, | |
| "eval_steps_per_second": 1.74, | |
| "step": 1200000 | |
| }, | |
| { | |
| "epoch": 19.92, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8382, | |
| "step": 1250000 | |
| }, | |
| { | |
| "epoch": 19.92, | |
| "eval_loss": 0.8779821991920471, | |
| "eval_runtime": 1999.9414, | |
| "eval_samples_per_second": 111.561, | |
| "eval_steps_per_second": 1.744, | |
| "step": 1250000 | |
| }, | |
| { | |
| "epoch": 20.72, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.8356, | |
| "step": 1300000 | |
| }, | |
| { | |
| "epoch": 20.72, | |
| "eval_loss": 0.878333330154419, | |
| "eval_runtime": 1991.7989, | |
| "eval_samples_per_second": 112.017, | |
| "eval_steps_per_second": 1.751, | |
| "step": 1300000 | |
| }, | |
| { | |
| "epoch": 21.51, | |
| "learning_rate": 4.992034708610221e-05, | |
| "loss": 0.8311, | |
| "step": 1350000 | |
| }, | |
| { | |
| "epoch": 21.51, | |
| "eval_loss": 0.8799993991851807, | |
| "eval_runtime": 2009.116, | |
| "eval_samples_per_second": 111.051, | |
| "eval_steps_per_second": 1.736, | |
| "step": 1350000 | |
| }, | |
| { | |
| "epoch": 22.31, | |
| "learning_rate": 4.992035027330242e-05, | |
| "loss": 0.8297, | |
| "step": 1400000 | |
| }, | |
| { | |
| "epoch": 22.31, | |
| "eval_loss": 0.8792157769203186, | |
| "eval_runtime": 1985.6931, | |
| "eval_samples_per_second": 112.361, | |
| "eval_steps_per_second": 1.756, | |
| "step": 1400000 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8294, | |
| "step": 1450000 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "eval_loss": 0.8799900412559509, | |
| "eval_runtime": 2021.1167, | |
| "eval_samples_per_second": 110.392, | |
| "eval_steps_per_second": 1.725, | |
| "step": 1450000 | |
| }, | |
| { | |
| "epoch": 23.9, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8277, | |
| "step": 1500000 | |
| }, | |
| { | |
| "epoch": 23.9, | |
| "eval_loss": 0.8771235346794128, | |
| "eval_runtime": 1995.2079, | |
| "eval_samples_per_second": 111.825, | |
| "eval_steps_per_second": 1.748, | |
| "step": 1500000 | |
| }, | |
| { | |
| "epoch": 24.7, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8254, | |
| "step": 1550000 | |
| }, | |
| { | |
| "epoch": 24.7, | |
| "eval_loss": 0.8784825205802917, | |
| "eval_runtime": 1994.3392, | |
| "eval_samples_per_second": 111.874, | |
| "eval_steps_per_second": 1.748, | |
| "step": 1550000 | |
| }, | |
| { | |
| "epoch": 25.5, | |
| "learning_rate": 4.992034708610221e-05, | |
| "loss": 0.821, | |
| "step": 1600000 | |
| }, | |
| { | |
| "epoch": 25.5, | |
| "eval_loss": 0.8786540627479553, | |
| "eval_runtime": 1998.0285, | |
| "eval_samples_per_second": 111.668, | |
| "eval_steps_per_second": 1.745, | |
| "step": 1600000 | |
| }, | |
| { | |
| "epoch": 26.29, | |
| "learning_rate": 4.9920348679702315e-05, | |
| "loss": 0.8201, | |
| "step": 1650000 | |
| }, | |
| { | |
| "epoch": 26.29, | |
| "eval_loss": 0.8791316151618958, | |
| "eval_runtime": 2008.615, | |
| "eval_samples_per_second": 111.079, | |
| "eval_steps_per_second": 1.736, | |
| "step": 1650000 | |
| } | |
| ], | |
| "logging_steps": 50000, | |
| "max_steps": 31375500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 500, | |
| "save_steps": 50000, | |
| "total_flos": 3.44904633040896e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |