| { | |
| "best_metric": 1.0877293348312378, | |
| "best_model_checkpoint": "./outputs/checkpoint-4100", | |
| "epoch": 2.9879781420765026, | |
| "eval_steps": 100, | |
| "global_step": 4100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7655, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 1.6318098306655884, | |
| "eval_runtime": 431.2124, | |
| "eval_samples_per_second": 14.55, | |
| "eval_steps_per_second": 1.82, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6075, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.5865398645401, | |
| "eval_runtime": 418.8687, | |
| "eval_samples_per_second": 14.978, | |
| "eval_steps_per_second": 1.874, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.575, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.555732250213623, | |
| "eval_runtime": 419.0983, | |
| "eval_samples_per_second": 14.97, | |
| "eval_steps_per_second": 1.873, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5438, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.5291692018508911, | |
| "eval_runtime": 418.8898, | |
| "eval_samples_per_second": 14.978, | |
| "eval_steps_per_second": 1.874, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5106, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.5064970254898071, | |
| "eval_runtime": 419.133, | |
| "eval_samples_per_second": 14.969, | |
| "eval_steps_per_second": 1.873, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4939, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.4864610433578491, | |
| "eval_runtime": 418.2311, | |
| "eval_samples_per_second": 15.001, | |
| "eval_steps_per_second": 1.877, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4713, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.4674705266952515, | |
| "eval_runtime": 418.2014, | |
| "eval_samples_per_second": 15.002, | |
| "eval_steps_per_second": 1.877, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4616, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.4496861696243286, | |
| "eval_runtime": 418.2918, | |
| "eval_samples_per_second": 14.999, | |
| "eval_steps_per_second": 1.877, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4323, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.4320642948150635, | |
| "eval_runtime": 418.2292, | |
| "eval_samples_per_second": 15.001, | |
| "eval_steps_per_second": 1.877, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4266, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.4162260293960571, | |
| "eval_runtime": 417.8585, | |
| "eval_samples_per_second": 15.015, | |
| "eval_steps_per_second": 1.879, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4124, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.4014889001846313, | |
| "eval_runtime": 417.75, | |
| "eval_samples_per_second": 15.019, | |
| "eval_steps_per_second": 1.879, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3846, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.3861923217773438, | |
| "eval_runtime": 417.77, | |
| "eval_samples_per_second": 15.018, | |
| "eval_steps_per_second": 1.879, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3934, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.371946096420288, | |
| "eval_runtime": 417.7963, | |
| "eval_samples_per_second": 15.017, | |
| "eval_steps_per_second": 1.879, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3557, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.3575325012207031, | |
| "eval_runtime": 417.6653, | |
| "eval_samples_per_second": 15.022, | |
| "eval_steps_per_second": 1.879, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3316, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.3450016975402832, | |
| "eval_runtime": 417.6497, | |
| "eval_samples_per_second": 15.022, | |
| "eval_steps_per_second": 1.88, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3211, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.3318936824798584, | |
| "eval_runtime": 418.2286, | |
| "eval_samples_per_second": 15.001, | |
| "eval_steps_per_second": 1.877, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3213, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.3180677890777588, | |
| "eval_runtime": 418.7732, | |
| "eval_samples_per_second": 14.982, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2972, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.3067623376846313, | |
| "eval_runtime": 417.8183, | |
| "eval_samples_per_second": 15.016, | |
| "eval_steps_per_second": 1.879, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.289, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.294016718864441, | |
| "eval_runtime": 418.0677, | |
| "eval_samples_per_second": 15.007, | |
| "eval_steps_per_second": 1.878, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2764, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.2830415964126587, | |
| "eval_runtime": 448.8239, | |
| "eval_samples_per_second": 13.979, | |
| "eval_steps_per_second": 1.749, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2666, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.271437406539917, | |
| "eval_runtime": 417.7696, | |
| "eval_samples_per_second": 15.018, | |
| "eval_steps_per_second": 1.879, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2456, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.2596800327301025, | |
| "eval_runtime": 418.2044, | |
| "eval_samples_per_second": 15.002, | |
| "eval_steps_per_second": 1.877, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2442, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.2501691579818726, | |
| "eval_runtime": 418.0356, | |
| "eval_samples_per_second": 15.008, | |
| "eval_steps_per_second": 1.878, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2415, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.2380064725875854, | |
| "eval_runtime": 417.8498, | |
| "eval_samples_per_second": 15.015, | |
| "eval_steps_per_second": 1.879, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2229, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.227720856666565, | |
| "eval_runtime": 417.9232, | |
| "eval_samples_per_second": 15.012, | |
| "eval_steps_per_second": 1.878, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2112, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.2167377471923828, | |
| "eval_runtime": 418.696, | |
| "eval_samples_per_second": 14.985, | |
| "eval_steps_per_second": 1.875, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2038, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.2063579559326172, | |
| "eval_runtime": 417.9549, | |
| "eval_samples_per_second": 15.011, | |
| "eval_steps_per_second": 1.878, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1797, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.1974104642868042, | |
| "eval_runtime": 418.4692, | |
| "eval_samples_per_second": 14.993, | |
| "eval_steps_per_second": 1.876, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1547, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.1865276098251343, | |
| "eval_runtime": 417.7842, | |
| "eval_samples_per_second": 15.017, | |
| "eval_steps_per_second": 1.879, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1572, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.1763572692871094, | |
| "eval_runtime": 417.556, | |
| "eval_samples_per_second": 15.026, | |
| "eval_steps_per_second": 1.88, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1399, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.1691069602966309, | |
| "eval_runtime": 417.8778, | |
| "eval_samples_per_second": 15.014, | |
| "eval_steps_per_second": 1.879, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1534, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.1825001239776611, | |
| "eval_runtime": 339.9447, | |
| "eval_samples_per_second": 18.456, | |
| "eval_steps_per_second": 2.309, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1428, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 1.169150948524475, | |
| "eval_runtime": 346.6169, | |
| "eval_samples_per_second": 18.101, | |
| "eval_steps_per_second": 2.265, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1263, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.1573188304901123, | |
| "eval_runtime": 340.3731, | |
| "eval_samples_per_second": 18.433, | |
| "eval_steps_per_second": 2.306, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1238, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.1467550992965698, | |
| "eval_runtime": 340.2433, | |
| "eval_samples_per_second": 18.44, | |
| "eval_steps_per_second": 2.307, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1167, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.136830449104309, | |
| "eval_runtime": 340.4238, | |
| "eval_samples_per_second": 18.43, | |
| "eval_steps_per_second": 2.306, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1115, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.1273517608642578, | |
| "eval_runtime": 340.9753, | |
| "eval_samples_per_second": 18.4, | |
| "eval_steps_per_second": 2.302, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0948, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.1172122955322266, | |
| "eval_runtime": 340.8373, | |
| "eval_samples_per_second": 18.408, | |
| "eval_steps_per_second": 2.303, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0913, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.1067975759506226, | |
| "eval_runtime": 341.0366, | |
| "eval_samples_per_second": 18.397, | |
| "eval_steps_per_second": 2.302, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0811, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 1.0958919525146484, | |
| "eval_runtime": 340.8306, | |
| "eval_samples_per_second": 18.408, | |
| "eval_steps_per_second": 2.303, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0002, | |
| "loss": 1.07, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 1.0877293348312378, | |
| "eval_runtime": 357.5953, | |
| "eval_samples_per_second": 17.545, | |
| "eval_steps_per_second": 2.195, | |
| "step": 4100 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.3292894134103962e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |