| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.238095238095237, | |
| "eval_steps": 500, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 0.9942206144332886, | |
| "learning_rate": 0.00019750000000000003, | |
| "loss": 9.5613, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 1.2790788412094116, | |
| "learning_rate": 0.000195, | |
| "loss": 9.2339, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 1.9939367771148682, | |
| "learning_rate": 0.00019250000000000002, | |
| "loss": 8.7953, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 2.010485887527466, | |
| "learning_rate": 0.00019, | |
| "loss": 8.2168, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 1.352328896522522, | |
| "learning_rate": 0.0001875, | |
| "loss": 7.8941, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.9626594185829163, | |
| "learning_rate": 0.00018500000000000002, | |
| "loss": 7.6817, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 1.1568268537521362, | |
| "learning_rate": 0.0001825, | |
| "loss": 7.5131, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 1.0264520645141602, | |
| "learning_rate": 0.00018, | |
| "loss": 7.4247, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.9865540862083435, | |
| "learning_rate": 0.0001775, | |
| "loss": 7.4369, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 1.0182702541351318, | |
| "learning_rate": 0.000175, | |
| "loss": 7.3787, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.7922359108924866, | |
| "learning_rate": 0.00017250000000000002, | |
| "loss": 7.373, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.7033187747001648, | |
| "learning_rate": 0.00017, | |
| "loss": 7.3096, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 2.9758119583129883, | |
| "learning_rate": 0.0001675, | |
| "loss": 7.1991, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.7531760931015015, | |
| "learning_rate": 0.000165, | |
| "loss": 7.2661, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.3790533542633057, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 7.2782, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0476190476190474, | |
| "grad_norm": 0.6538093686103821, | |
| "learning_rate": 0.00016, | |
| "loss": 7.2109, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.238095238095238, | |
| "grad_norm": 0.6145215630531311, | |
| "learning_rate": 0.0001575, | |
| "loss": 7.2192, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.4128475785255432, | |
| "learning_rate": 0.000155, | |
| "loss": 7.2892, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.619047619047619, | |
| "grad_norm": 1.0160013437271118, | |
| "learning_rate": 0.0001525, | |
| "loss": 7.2049, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.5834835171699524, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 7.1672, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.4894554615020752, | |
| "learning_rate": 0.0001475, | |
| "loss": 7.1269, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 0.593618631362915, | |
| "learning_rate": 0.000145, | |
| "loss": 7.0175, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 1.6190487146377563, | |
| "learning_rate": 0.00014250000000000002, | |
| "loss": 7.2919, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.755859911441803, | |
| "learning_rate": 0.00014, | |
| "loss": 7.1624, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.46613645553588867, | |
| "learning_rate": 0.0001375, | |
| "loss": 7.2233, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.9523809523809526, | |
| "grad_norm": 0.5973020792007446, | |
| "learning_rate": 0.00013500000000000003, | |
| "loss": 7.1642, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.97837233543396, | |
| "learning_rate": 0.0001325, | |
| "loss": 7.1172, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.9348046183586121, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 7.1564, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.523809523809524, | |
| "grad_norm": 0.6632198691368103, | |
| "learning_rate": 0.0001275, | |
| "loss": 7.0821, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.7776179909706116, | |
| "learning_rate": 0.000125, | |
| "loss": 7.2272, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.904761904761905, | |
| "grad_norm": 0.6282438039779663, | |
| "learning_rate": 0.00012250000000000002, | |
| "loss": 7.0926, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.095238095238095, | |
| "grad_norm": 0.6008353233337402, | |
| "learning_rate": 0.00012, | |
| "loss": 7.1073, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.8796420097351074, | |
| "learning_rate": 0.00011750000000000001, | |
| "loss": 7.1737, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.476190476190476, | |
| "grad_norm": 0.6400454640388489, | |
| "learning_rate": 0.00011499999999999999, | |
| "loss": 7.0924, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.5479526519775391, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 7.1275, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 0.5992618203163147, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 7.0599, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 7.0476190476190474, | |
| "grad_norm": 0.5336684584617615, | |
| "learning_rate": 0.0001075, | |
| "loss": 7.0206, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 7.238095238095238, | |
| "grad_norm": 0.3991040289402008, | |
| "learning_rate": 0.000105, | |
| "loss": 7.0123, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.428571428571429, | |
| "grad_norm": 1.032917857170105, | |
| "learning_rate": 0.0001025, | |
| "loss": 7.0267, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.619047619047619, | |
| "grad_norm": 0.5554404854774475, | |
| "learning_rate": 0.0001, | |
| "loss": 7.0203, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.809523809523809, | |
| "grad_norm": 0.7755109667778015, | |
| "learning_rate": 9.75e-05, | |
| "loss": 7.1445, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.8295842409133911, | |
| "learning_rate": 9.5e-05, | |
| "loss": 7.0002, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 8.19047619047619, | |
| "grad_norm": 1.4985620975494385, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 7.0613, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.380952380952381, | |
| "grad_norm": 1.0733778476715088, | |
| "learning_rate": 9e-05, | |
| "loss": 7.0594, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.7009026408195496, | |
| "learning_rate": 8.75e-05, | |
| "loss": 6.9432, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.761904761904763, | |
| "grad_norm": 1.195196509361267, | |
| "learning_rate": 8.5e-05, | |
| "loss": 6.9266, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.952380952380953, | |
| "grad_norm": 2.6835684776306152, | |
| "learning_rate": 8.25e-05, | |
| "loss": 6.9855, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 9.142857142857142, | |
| "grad_norm": 0.7434377670288086, | |
| "learning_rate": 8e-05, | |
| "loss": 6.7975, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 9.333333333333334, | |
| "grad_norm": 0.5993837118148804, | |
| "learning_rate": 7.75e-05, | |
| "loss": 7.0476, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.523809523809524, | |
| "grad_norm": 0.4656153619289398, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 6.9894, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "grad_norm": 0.7926774621009827, | |
| "learning_rate": 7.25e-05, | |
| "loss": 6.9854, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.904761904761905, | |
| "grad_norm": 1.0828678607940674, | |
| "learning_rate": 7e-05, | |
| "loss": 6.9185, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 10.095238095238095, | |
| "grad_norm": 0.6923830509185791, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 6.9804, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 10.285714285714286, | |
| "grad_norm": 0.5546735525131226, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 6.9273, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.476190476190476, | |
| "grad_norm": 0.8265076875686646, | |
| "learning_rate": 6.25e-05, | |
| "loss": 6.9087, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.666666666666666, | |
| "grad_norm": 0.3945198655128479, | |
| "learning_rate": 6e-05, | |
| "loss": 6.9375, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "grad_norm": 0.5948878526687622, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 6.8764, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 11.047619047619047, | |
| "grad_norm": 0.7741471529006958, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 6.7551, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 11.238095238095237, | |
| "grad_norm": 0.32554784417152405, | |
| "learning_rate": 5.25e-05, | |
| "loss": 6.8862, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 0.5033702850341797, | |
| "learning_rate": 5e-05, | |
| "loss": 6.7297, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.619047619047619, | |
| "grad_norm": 0.5291158556938171, | |
| "learning_rate": 4.75e-05, | |
| "loss": 6.9826, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 11.80952380952381, | |
| "grad_norm": 0.39498385787010193, | |
| "learning_rate": 4.5e-05, | |
| "loss": 6.837, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.4402136206626892, | |
| "learning_rate": 4.25e-05, | |
| "loss": 7.0434, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 12.19047619047619, | |
| "grad_norm": 0.6476764678955078, | |
| "learning_rate": 4e-05, | |
| "loss": 6.8524, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 12.380952380952381, | |
| "grad_norm": 0.330609530210495, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 6.8742, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "grad_norm": 0.5420040488243103, | |
| "learning_rate": 3.5e-05, | |
| "loss": 6.7931, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.761904761904763, | |
| "grad_norm": 0.3482373356819153, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 6.883, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 12.952380952380953, | |
| "grad_norm": 0.3476051092147827, | |
| "learning_rate": 3e-05, | |
| "loss": 6.9857, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 13.142857142857142, | |
| "grad_norm": 0.43590274453163147, | |
| "learning_rate": 2.8749999999999997e-05, | |
| "loss": 8.116, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.2993098497390747, | |
| "learning_rate": 2.625e-05, | |
| "loss": 6.657, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 13.523809523809524, | |
| "grad_norm": 0.3477262556552887, | |
| "learning_rate": 2.375e-05, | |
| "loss": 6.9781, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "grad_norm": 0.47370073199272156, | |
| "learning_rate": 2.125e-05, | |
| "loss": 6.9277, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 13.904761904761905, | |
| "grad_norm": 0.3924289345741272, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 6.8967, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 14.095238095238095, | |
| "grad_norm": 0.5621922612190247, | |
| "learning_rate": 1.6250000000000002e-05, | |
| "loss": 6.7197, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 0.3454875349998474, | |
| "learning_rate": 1.3750000000000002e-05, | |
| "loss": 6.9314, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 14.476190476190476, | |
| "grad_norm": 0.3146642744541168, | |
| "learning_rate": 1.125e-05, | |
| "loss": 6.9142, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 14.666666666666666, | |
| "grad_norm": 0.3762160837650299, | |
| "learning_rate": 8.75e-06, | |
| "loss": 6.8759, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "grad_norm": 0.33906954526901245, | |
| "learning_rate": 6.25e-06, | |
| "loss": 6.8712, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 15.047619047619047, | |
| "grad_norm": 0.3414846360683441, | |
| "learning_rate": 3.75e-06, | |
| "loss": 6.737, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 15.238095238095237, | |
| "grad_norm": 0.4463809132575989, | |
| "learning_rate": 1.25e-06, | |
| "loss": 6.9144, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 15.238095238095237, | |
| "step": 160, | |
| "total_flos": 800861569170024.0, | |
| "train_loss": 7.170098584890366, | |
| "train_runtime": 677.1666, | |
| "train_samples_per_second": 3.969, | |
| "train_steps_per_second": 0.236 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 16, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 800861569170024.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |