| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5555555555555556, | |
| "eval_steps": 500, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013888888888888888, | |
| "grad_norm": 33.87594223022461, | |
| "learning_rate": 4.930555555555556e-05, | |
| "loss": 2.5607, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.027777777777777776, | |
| "grad_norm": 19.041858673095703, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 2.3205, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.041666666666666664, | |
| "grad_norm": 22.798561096191406, | |
| "learning_rate": 4.791666666666667e-05, | |
| "loss": 2.2861, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 26.49089241027832, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 2.256, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06944444444444445, | |
| "grad_norm": 19.455162048339844, | |
| "learning_rate": 4.652777777777778e-05, | |
| "loss": 2.2464, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "grad_norm": 19.90087890625, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 2.2445, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.09722222222222222, | |
| "grad_norm": 15.378240585327148, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 2.2345, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 17.459095001220703, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 2.2252, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 23.182748794555664, | |
| "learning_rate": 4.375e-05, | |
| "loss": 2.2134, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.1388888888888889, | |
| "grad_norm": 19.53196144104004, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 2.2168, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1527777777777778, | |
| "grad_norm": 15.988591194152832, | |
| "learning_rate": 4.236111111111111e-05, | |
| "loss": 2.2232, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 16.655384063720703, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 2.2147, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.18055555555555555, | |
| "grad_norm": 14.632539749145508, | |
| "learning_rate": 4.0972222222222225e-05, | |
| "loss": 2.1922, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.19444444444444445, | |
| "grad_norm": 13.886147499084473, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 2.2036, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.20833333333333334, | |
| "grad_norm": 13.439935684204102, | |
| "learning_rate": 3.958333333333333e-05, | |
| "loss": 2.1865, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 10.455207824707031, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 2.1794, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.2361111111111111, | |
| "grad_norm": 17.11553382873535, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 2.2025, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 14.08956527709961, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 2.1867, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.2638888888888889, | |
| "grad_norm": 13.953351020812988, | |
| "learning_rate": 3.6805555555555556e-05, | |
| "loss": 2.1651, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 10.86223030090332, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 2.1835, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2916666666666667, | |
| "grad_norm": 10.25207805633545, | |
| "learning_rate": 3.541666666666667e-05, | |
| "loss": 2.1831, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.3055555555555556, | |
| "grad_norm": 10.212087631225586, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 2.1699, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.3194444444444444, | |
| "grad_norm": 8.576708793640137, | |
| "learning_rate": 3.402777777777778e-05, | |
| "loss": 2.1689, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 7.758203506469727, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.1726, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.3472222222222222, | |
| "grad_norm": 10.195472717285156, | |
| "learning_rate": 3.263888888888889e-05, | |
| "loss": 2.1753, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.3611111111111111, | |
| "grad_norm": 7.407456398010254, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 2.1825, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 8.167924880981445, | |
| "learning_rate": 3.125e-05, | |
| "loss": 2.1762, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 7.759510517120361, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 2.1687, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.4027777777777778, | |
| "grad_norm": 8.27856731414795, | |
| "learning_rate": 2.9861111111111113e-05, | |
| "loss": 2.18, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 8.577377319335938, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 2.1718, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.4305555555555556, | |
| "grad_norm": 7.465952396392822, | |
| "learning_rate": 2.8472222222222223e-05, | |
| "loss": 2.1778, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 6.5437331199646, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 2.1756, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.4583333333333333, | |
| "grad_norm": 9.449009895324707, | |
| "learning_rate": 2.7083333333333332e-05, | |
| "loss": 2.1608, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.4722222222222222, | |
| "grad_norm": 5.527251243591309, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 2.1742, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.4861111111111111, | |
| "grad_norm": 6.589778423309326, | |
| "learning_rate": 2.5694444444444445e-05, | |
| "loss": 2.1754, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 7.19509744644165, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.1715, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.5138888888888888, | |
| "grad_norm": 6.292264938354492, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 2.1555, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.5277777777777778, | |
| "grad_norm": 7.13681697845459, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 2.1603, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.5416666666666666, | |
| "grad_norm": 6.618354320526123, | |
| "learning_rate": 2.2916666666666667e-05, | |
| "loss": 2.1636, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 6.527144908905029, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 2.1617, | |
| "step": 20000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 36000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10000, | |
| "total_flos": 1.857401389056e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |