| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.997867803837953, | |
| "eval_steps": 10, | |
| "global_step": 351, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 2.3058, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.0794, | |
| "eval_loss": 2.310584306716919, | |
| "eval_runtime": 23.1302, | |
| "eval_samples_per_second": 216.168, | |
| "eval_steps_per_second": 6.788, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 2.3033, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.0892, | |
| "eval_loss": 2.302612066268921, | |
| "eval_runtime": 20.5744, | |
| "eval_samples_per_second": 243.02, | |
| "eval_steps_per_second": 7.631, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 2.3012, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.1042, | |
| "eval_loss": 2.2971391677856445, | |
| "eval_runtime": 19.7054, | |
| "eval_samples_per_second": 253.737, | |
| "eval_steps_per_second": 7.967, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.936507936507937e-05, | |
| "loss": 2.2914, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.1254, | |
| "eval_loss": 2.2889511585235596, | |
| "eval_runtime": 20.5875, | |
| "eval_samples_per_second": 242.866, | |
| "eval_steps_per_second": 7.626, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 2.2869, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.16, | |
| "eval_loss": 2.281619071960449, | |
| "eval_runtime": 20.7333, | |
| "eval_samples_per_second": 241.158, | |
| "eval_steps_per_second": 7.572, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.6190476190476194e-05, | |
| "loss": 2.2785, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.1902, | |
| "eval_loss": 2.269974708557129, | |
| "eval_runtime": 20.4734, | |
| "eval_samples_per_second": 244.22, | |
| "eval_steps_per_second": 7.669, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.460317460317461e-05, | |
| "loss": 2.2712, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.2354, | |
| "eval_loss": 2.260219097137451, | |
| "eval_runtime": 19.9116, | |
| "eval_samples_per_second": 251.11, | |
| "eval_steps_per_second": 7.885, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.301587301587302e-05, | |
| "loss": 2.2619, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.2688, | |
| "eval_loss": 2.250117063522339, | |
| "eval_runtime": 21.2411, | |
| "eval_samples_per_second": 235.392, | |
| "eval_steps_per_second": 7.391, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 2.2509, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_accuracy": 0.3022, | |
| "eval_loss": 2.2383360862731934, | |
| "eval_runtime": 20.8445, | |
| "eval_samples_per_second": 239.872, | |
| "eval_steps_per_second": 7.532, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.984126984126984e-05, | |
| "loss": 2.2382, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.3268, | |
| "eval_loss": 2.222919225692749, | |
| "eval_runtime": 19.2819, | |
| "eval_samples_per_second": 259.311, | |
| "eval_steps_per_second": 8.142, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.8253968253968256e-05, | |
| "loss": 2.2255, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.353, | |
| "eval_loss": 2.2083821296691895, | |
| "eval_runtime": 23.1248, | |
| "eval_samples_per_second": 216.218, | |
| "eval_steps_per_second": 6.789, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 2.2164, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.3608, | |
| "eval_loss": 2.1939358711242676, | |
| "eval_runtime": 25.8088, | |
| "eval_samples_per_second": 193.732, | |
| "eval_steps_per_second": 6.083, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.5079365079365075e-05, | |
| "loss": 2.2028, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.3668, | |
| "eval_loss": 2.182861804962158, | |
| "eval_runtime": 25.579, | |
| "eval_samples_per_second": 195.473, | |
| "eval_steps_per_second": 6.138, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.349206349206349e-05, | |
| "loss": 2.1977, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.401, | |
| "eval_loss": 2.164577007293701, | |
| "eval_runtime": 25.165, | |
| "eval_samples_per_second": 198.688, | |
| "eval_steps_per_second": 6.239, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.19047619047619e-05, | |
| "loss": 2.1844, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.4244, | |
| "eval_loss": 2.144054651260376, | |
| "eval_runtime": 24.2393, | |
| "eval_samples_per_second": 206.277, | |
| "eval_steps_per_second": 6.477, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.0317460317460318e-05, | |
| "loss": 2.1689, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.437, | |
| "eval_loss": 2.1322500705718994, | |
| "eval_runtime": 25.4332, | |
| "eval_samples_per_second": 196.593, | |
| "eval_steps_per_second": 6.173, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.8730158730158728e-05, | |
| "loss": 2.1555, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.4462, | |
| "eval_loss": 2.1159207820892334, | |
| "eval_runtime": 25.3372, | |
| "eval_samples_per_second": 197.338, | |
| "eval_steps_per_second": 6.196, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 2.1448, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 2.0992112159729004, | |
| "eval_runtime": 25.0688, | |
| "eval_samples_per_second": 199.451, | |
| "eval_steps_per_second": 6.263, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 2.1313, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.4642, | |
| "eval_loss": 2.080961227416992, | |
| "eval_runtime": 25.2129, | |
| "eval_samples_per_second": 198.311, | |
| "eval_steps_per_second": 6.227, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.396825396825397e-05, | |
| "loss": 2.1189, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.4708, | |
| "eval_loss": 2.0589163303375244, | |
| "eval_runtime": 25.6278, | |
| "eval_samples_per_second": 195.101, | |
| "eval_steps_per_second": 6.126, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.2380952380952384e-05, | |
| "loss": 2.1111, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.4828, | |
| "eval_loss": 2.0430362224578857, | |
| "eval_runtime": 19.2415, | |
| "eval_samples_per_second": 259.854, | |
| "eval_steps_per_second": 8.159, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 2.0793650793650797e-05, | |
| "loss": 2.0905, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_accuracy": 0.4938, | |
| "eval_loss": 2.028820753097534, | |
| "eval_runtime": 20.6864, | |
| "eval_samples_per_second": 241.705, | |
| "eval_steps_per_second": 7.59, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.920634920634921e-05, | |
| "loss": 2.082, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.4938, | |
| "eval_loss": 2.008862257003784, | |
| "eval_runtime": 20.028, | |
| "eval_samples_per_second": 249.65, | |
| "eval_steps_per_second": 7.839, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.761904761904762e-05, | |
| "loss": 2.0646, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.5014, | |
| "eval_loss": 1.9969898462295532, | |
| "eval_runtime": 20.7247, | |
| "eval_samples_per_second": 241.258, | |
| "eval_steps_per_second": 7.576, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.6031746031746033e-05, | |
| "loss": 2.0636, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.4946, | |
| "eval_loss": 1.9777544736862183, | |
| "eval_runtime": 18.9902, | |
| "eval_samples_per_second": 263.294, | |
| "eval_steps_per_second": 8.267, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 2.0579, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.49, | |
| "eval_loss": 1.9608845710754395, | |
| "eval_runtime": 18.9333, | |
| "eval_samples_per_second": 264.084, | |
| "eval_steps_per_second": 8.292, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 2.028, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.4862, | |
| "eval_loss": 1.960185170173645, | |
| "eval_runtime": 18.9535, | |
| "eval_samples_per_second": 263.804, | |
| "eval_steps_per_second": 8.283, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.126984126984127e-05, | |
| "loss": 2.0447, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.4934, | |
| "eval_loss": 1.9459648132324219, | |
| "eval_runtime": 18.8589, | |
| "eval_samples_per_second": 265.126, | |
| "eval_steps_per_second": 8.325, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 9.682539682539683e-06, | |
| "loss": 2.0168, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.505, | |
| "eval_loss": 1.9368737936019897, | |
| "eval_runtime": 19.0033, | |
| "eval_samples_per_second": 263.112, | |
| "eval_steps_per_second": 8.262, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 8.095238095238097e-06, | |
| "loss": 2.0126, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.4926, | |
| "eval_loss": 1.931652545928955, | |
| "eval_runtime": 19.1274, | |
| "eval_samples_per_second": 261.406, | |
| "eval_steps_per_second": 8.208, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6.507936507936509e-06, | |
| "loss": 2.0099, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.4952, | |
| "eval_loss": 1.9234933853149414, | |
| "eval_runtime": 19.032, | |
| "eval_samples_per_second": 262.715, | |
| "eval_steps_per_second": 8.249, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.920634920634921e-06, | |
| "loss": 1.9978, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.4972, | |
| "eval_loss": 1.9174150228500366, | |
| "eval_runtime": 18.9404, | |
| "eval_samples_per_second": 263.986, | |
| "eval_steps_per_second": 8.289, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.9951, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.507, | |
| "eval_loss": 1.9119243621826172, | |
| "eval_runtime": 18.9559, | |
| "eval_samples_per_second": 263.769, | |
| "eval_steps_per_second": 8.282, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.7460317460317462e-06, | |
| "loss": 1.9823, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.4992, | |
| "eval_loss": 1.9119775295257568, | |
| "eval_runtime": 19.001, | |
| "eval_samples_per_second": 263.143, | |
| "eval_steps_per_second": 8.263, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.5873015873015874e-07, | |
| "loss": 1.985, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5022, | |
| "eval_loss": 1.9064103364944458, | |
| "eval_runtime": 19.1234, | |
| "eval_samples_per_second": 261.459, | |
| "eval_steps_per_second": 8.21, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 351, | |
| "total_flos": 9.547293521089659e+17, | |
| "train_loss": 2.144495400947723, | |
| "train_runtime": 1271.552, | |
| "train_samples_per_second": 35.39, | |
| "train_steps_per_second": 0.276 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 351, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 9.547293521089659e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |