| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 100, | |
| "global_step": 248, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04032258064516129, | |
| "grad_norm": 11.799724578857422, | |
| "learning_rate": 1.967741935483871e-05, | |
| "loss": 1.8639, | |
| "num_tokens": 584580.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08064516129032258, | |
| "grad_norm": 2.09053897857666, | |
| "learning_rate": 1.9274193548387097e-05, | |
| "loss": 1.5389, | |
| "num_tokens": 1169428.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12096774193548387, | |
| "grad_norm": 1.7200100421905518, | |
| "learning_rate": 1.8870967741935487e-05, | |
| "loss": 1.5317, | |
| "num_tokens": 1770489.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16129032258064516, | |
| "grad_norm": 1.7961115837097168, | |
| "learning_rate": 1.8467741935483873e-05, | |
| "loss": 1.5076, | |
| "num_tokens": 2367303.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "grad_norm": 1.5970382690429688, | |
| "learning_rate": 1.806451612903226e-05, | |
| "loss": 1.4555, | |
| "num_tokens": 2947847.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.24193548387096775, | |
| "grad_norm": 1.6028152704238892, | |
| "learning_rate": 1.7661290322580645e-05, | |
| "loss": 1.4162, | |
| "num_tokens": 3547811.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28225806451612906, | |
| "grad_norm": 1.4596558809280396, | |
| "learning_rate": 1.7258064516129035e-05, | |
| "loss": 1.3612, | |
| "num_tokens": 4119060.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 1.3425326347351074, | |
| "learning_rate": 1.685483870967742e-05, | |
| "loss": 1.414, | |
| "num_tokens": 4727662.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3629032258064516, | |
| "grad_norm": 1.271307110786438, | |
| "learning_rate": 1.6451612903225807e-05, | |
| "loss": 1.3948, | |
| "num_tokens": 5318819.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "grad_norm": 1.2482575178146362, | |
| "learning_rate": 1.6048387096774193e-05, | |
| "loss": 1.3413, | |
| "num_tokens": 5900929.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4435483870967742, | |
| "grad_norm": 1.2999275922775269, | |
| "learning_rate": 1.5645161290322583e-05, | |
| "loss": 1.3457, | |
| "num_tokens": 6476426.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.4838709677419355, | |
| "grad_norm": 1.4803128242492676, | |
| "learning_rate": 1.5241935483870969e-05, | |
| "loss": 1.3262, | |
| "num_tokens": 7034593.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5241935483870968, | |
| "grad_norm": 1.1995830535888672, | |
| "learning_rate": 1.4838709677419357e-05, | |
| "loss": 1.3028, | |
| "num_tokens": 7645033.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5645161290322581, | |
| "grad_norm": 2.7548296451568604, | |
| "learning_rate": 1.4435483870967743e-05, | |
| "loss": 1.314, | |
| "num_tokens": 8220494.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "grad_norm": 1.3331153392791748, | |
| "learning_rate": 1.4032258064516131e-05, | |
| "loss": 1.2997, | |
| "num_tokens": 8828117.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 1.2527021169662476, | |
| "learning_rate": 1.3629032258064517e-05, | |
| "loss": 1.331, | |
| "num_tokens": 9438365.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6854838709677419, | |
| "grad_norm": 1.3447303771972656, | |
| "learning_rate": 1.3225806451612903e-05, | |
| "loss": 1.295, | |
| "num_tokens": 10032986.0, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.7258064516129032, | |
| "grad_norm": 1.1974637508392334, | |
| "learning_rate": 1.2822580645161291e-05, | |
| "loss": 1.2876, | |
| "num_tokens": 10621259.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7661290322580645, | |
| "grad_norm": 1.3035883903503418, | |
| "learning_rate": 1.2419354838709679e-05, | |
| "loss": 1.2846, | |
| "num_tokens": 11222906.0, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 1.2527148723602295, | |
| "learning_rate": 1.2016129032258067e-05, | |
| "loss": 1.2915, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "eval_loss": 1.263860821723938, | |
| "eval_num_tokens": 11815575.0, | |
| "eval_runtime": 0.6219, | |
| "eval_samples_per_second": 321.577, | |
| "eval_steps_per_second": 6.432, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8467741935483871, | |
| "grad_norm": 1.223354458808899, | |
| "learning_rate": 1.1612903225806453e-05, | |
| "loss": 1.2529, | |
| "num_tokens": 12410628.0, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.8870967741935484, | |
| "grad_norm": 1.300518274307251, | |
| "learning_rate": 1.1209677419354839e-05, | |
| "loss": 1.3045, | |
| "num_tokens": 12992321.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9274193548387096, | |
| "grad_norm": 1.2451575994491577, | |
| "learning_rate": 1.0806451612903225e-05, | |
| "loss": 1.2933, | |
| "num_tokens": 13582229.0, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 1.42113196849823, | |
| "learning_rate": 1.0403225806451613e-05, | |
| "loss": 1.2582, | |
| "num_tokens": 14172667.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "grad_norm": 1.5717027187347412, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2315, | |
| "num_tokens": 14782591.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.0483870967741935, | |
| "grad_norm": 1.5194685459136963, | |
| "learning_rate": 9.596774193548389e-06, | |
| "loss": 1.1226, | |
| "num_tokens": 15357942.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0887096774193548, | |
| "grad_norm": 1.418308138847351, | |
| "learning_rate": 9.193548387096775e-06, | |
| "loss": 1.0846, | |
| "num_tokens": 15945153.0, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.129032258064516, | |
| "grad_norm": 1.3308682441711426, | |
| "learning_rate": 8.790322580645163e-06, | |
| "loss": 1.133, | |
| "num_tokens": 16533255.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1693548387096775, | |
| "grad_norm": 1.1994943618774414, | |
| "learning_rate": 8.387096774193549e-06, | |
| "loss": 1.0632, | |
| "num_tokens": 17123860.0, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.2096774193548387, | |
| "grad_norm": 1.1403435468673706, | |
| "learning_rate": 7.983870967741935e-06, | |
| "loss": 1.0878, | |
| "num_tokens": 17705881.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.1773253679275513, | |
| "learning_rate": 7.580645161290323e-06, | |
| "loss": 1.0848, | |
| "num_tokens": 18311747.0, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 1.1357718706130981, | |
| "learning_rate": 7.177419354838711e-06, | |
| "loss": 1.0685, | |
| "num_tokens": 18903969.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3306451612903225, | |
| "grad_norm": 1.154196858406067, | |
| "learning_rate": 6.774193548387097e-06, | |
| "loss": 1.1077, | |
| "num_tokens": 19486973.0, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.370967741935484, | |
| "grad_norm": 1.143140196800232, | |
| "learning_rate": 6.370967741935485e-06, | |
| "loss": 1.1272, | |
| "num_tokens": 20083469.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.4112903225806452, | |
| "grad_norm": 1.1998040676116943, | |
| "learning_rate": 5.967741935483872e-06, | |
| "loss": 1.1137, | |
| "num_tokens": 20660913.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.4516129032258065, | |
| "grad_norm": 1.155060887336731, | |
| "learning_rate": 5.564516129032258e-06, | |
| "loss": 1.1344, | |
| "num_tokens": 21241019.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4919354838709677, | |
| "grad_norm": 1.1472811698913574, | |
| "learning_rate": 5.161290322580646e-06, | |
| "loss": 1.081, | |
| "num_tokens": 21850671.0, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.532258064516129, | |
| "grad_norm": 1.1946035623550415, | |
| "learning_rate": 4.758064516129033e-06, | |
| "loss": 1.0698, | |
| "num_tokens": 22409894.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5725806451612905, | |
| "grad_norm": 1.121880054473877, | |
| "learning_rate": 4.35483870967742e-06, | |
| "loss": 1.1139, | |
| "num_tokens": 23009139.0, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 1.166876196861267, | |
| "learning_rate": 3.951612903225807e-06, | |
| "loss": 1.0958, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "eval_loss": 1.2381997108459473, | |
| "eval_num_tokens": 23571600.0, | |
| "eval_runtime": 0.5704, | |
| "eval_samples_per_second": 350.647, | |
| "eval_steps_per_second": 7.013, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.653225806451613, | |
| "grad_norm": 1.0650382041931152, | |
| "learning_rate": 3.548387096774194e-06, | |
| "loss": 1.1124, | |
| "num_tokens": 24153766.0, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.6935483870967742, | |
| "grad_norm": 1.1366987228393555, | |
| "learning_rate": 3.145161290322581e-06, | |
| "loss": 1.137, | |
| "num_tokens": 24761611.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.7338709677419355, | |
| "grad_norm": 1.0956673622131348, | |
| "learning_rate": 2.7419354838709676e-06, | |
| "loss": 1.1278, | |
| "num_tokens": 25370199.0, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.7741935483870968, | |
| "grad_norm": 1.0677380561828613, | |
| "learning_rate": 2.338709677419355e-06, | |
| "loss": 1.0614, | |
| "num_tokens": 26001840.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.814516129032258, | |
| "grad_norm": 1.0645986795425415, | |
| "learning_rate": 1.935483870967742e-06, | |
| "loss": 1.0969, | |
| "num_tokens": 26611726.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.8548387096774195, | |
| "grad_norm": 1.1630877256393433, | |
| "learning_rate": 1.5322580645161292e-06, | |
| "loss": 1.0922, | |
| "num_tokens": 27185670.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8951612903225805, | |
| "grad_norm": 1.1040847301483154, | |
| "learning_rate": 1.1290322580645162e-06, | |
| "loss": 1.1251, | |
| "num_tokens": 27777551.0, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.935483870967742, | |
| "grad_norm": 1.1222697496414185, | |
| "learning_rate": 7.258064516129033e-07, | |
| "loss": 1.0816, | |
| "num_tokens": 28376439.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9758064516129032, | |
| "grad_norm": 1.0824819803237915, | |
| "learning_rate": 3.2258064516129035e-07, | |
| "loss": 1.0641, | |
| "num_tokens": 28972576.0, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "num_tokens": 29322439.0, | |
| "step": 248, | |
| "total_flos": 3.616736600517509e+17, | |
| "train_loss": 1.2339729416754939, | |
| "train_runtime": 268.4154, | |
| "train_samples_per_second": 117.773, | |
| "train_steps_per_second": 0.924 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 248, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.616736600517509e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |