| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 954, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.208333333333333e-08, | |
| "loss": 5.0567, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.0416666666666667e-07, | |
| "loss": 4.3042, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.5624999999999999e-07, | |
| "loss": 2.0133, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "loss": 0.7885, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 2.604166666666667e-07, | |
| "loss": 0.4101, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "loss": 0.2846, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.645833333333333e-07, | |
| "loss": 0.2442, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "loss": 0.2025, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.6874999999999996e-07, | |
| "loss": 0.1316, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.999731868769026e-07, | |
| "loss": 0.1563, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.996716052911017e-07, | |
| "loss": 0.1045, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.990353313429303e-07, | |
| "loss": 0.0661, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.980652179769217e-07, | |
| "loss": 0.0561, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.967625656594781e-07, | |
| "loss": 0.0518, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.951291206355559e-07, | |
| "loss": 0.0491, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.050289858132600784, | |
| "eval_runtime": 22.1919, | |
| "eval_samples_per_second": 9.283, | |
| "eval_steps_per_second": 0.406, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.93167072587771e-07, | |
| "loss": 0.0469, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.908790517010636e-07, | |
| "loss": 0.0451, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 4.882681251368548e-07, | |
| "loss": 0.0437, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.853377929214243e-07, | |
| "loss": 0.0481, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 4.820919832540181e-07, | |
| "loss": 0.0493, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.785350472409791e-07, | |
| "loss": 0.0418, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.7467175306295647e-07, | |
| "loss": 0.0432, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.70507279583015e-07, | |
| "loss": 0.0426, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.6604720940421207e-07, | |
| "loss": 0.0414, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 4.612975213859487e-07, | |
| "loss": 0.0392, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 4.5626458262912735e-07, | |
| "loss": 0.0426, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.5095513994085974e-07, | |
| "loss": 0.0376, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 4.453763107901675e-07, | |
| "loss": 0.0413, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.395355737667985e-07, | |
| "loss": 0.0382, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 4.3344075855595097e-07, | |
| "loss": 0.0391, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.271000354423425e-07, | |
| "loss": 0.0407, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.04133007302880287, | |
| "eval_runtime": 22.0322, | |
| "eval_samples_per_second": 9.35, | |
| "eval_steps_per_second": 0.408, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.2052190435769554e-07, | |
| "loss": 0.0367, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 4.137151834863213e-07, | |
| "loss": 0.0345, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.0668899744407567e-07, | |
| "loss": 0.03, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 3.994527650465352e-07, | |
| "loss": 0.0335, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 3.920161866827889e-07, | |
| "loss": 0.0277, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 3.8438923131177237e-07, | |
| "loss": 0.0371, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 3.765821230985757e-07, | |
| "loss": 0.0316, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.6860532770864005e-07, | |
| "loss": 0.0359, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 3.604695382782159e-07, | |
| "loss": 0.0345, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.5218566107988867e-07, | |
| "loss": 0.0307, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 3.4376480090239047e-07, | |
| "loss": 0.0366, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 3.3521824616429284e-07, | |
| "loss": 0.0276, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.265574537815398e-07, | |
| "loss": 0.0365, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 3.1779403380910425e-07, | |
| "loss": 0.0321, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.0893973387735683e-07, | |
| "loss": 0.029, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 3.000064234440111e-07, | |
| "loss": 0.0307, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.04235048219561577, | |
| "eval_runtime": 22.107, | |
| "eval_samples_per_second": 9.318, | |
| "eval_steps_per_second": 0.407, | |
| "step": 477 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 2.910060778827554e-07, | |
| "loss": 0.0241, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 2.8195076242990116e-07, | |
| "loss": 0.0178, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 2.7285261601056697e-07, | |
| "loss": 0.025, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 2.6372383496608186e-07, | |
| "loss": 0.0225, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 2.5457665670441937e-07, | |
| "loss": 0.0237, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 2.454233432955807e-07, | |
| "loss": 0.028, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 2.3627616503391812e-07, | |
| "loss": 0.0246, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 2.2714738398943308e-07, | |
| "loss": 0.0167, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 2.1804923757009882e-07, | |
| "loss": 0.0245, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 2.089939221172446e-07, | |
| "loss": 0.0217, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.9999357655598891e-07, | |
| "loss": 0.0188, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 1.9106026612264315e-07, | |
| "loss": 0.0299, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 1.8220596619089573e-07, | |
| "loss": 0.0201, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.7344254621846017e-07, | |
| "loss": 0.0254, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 1.647817538357072e-07, | |
| "loss": 0.0208, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.562351990976095e-07, | |
| "loss": 0.0226, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.04757673665881157, | |
| "eval_runtime": 22.0105, | |
| "eval_samples_per_second": 9.359, | |
| "eval_steps_per_second": 0.409, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 1.478143389201113e-07, | |
| "loss": 0.0206, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 1.3953046172178413e-07, | |
| "loss": 0.0148, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 1.3139467229135998e-07, | |
| "loss": 0.0168, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.2341787690142435e-07, | |
| "loss": 0.0143, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.1561076868822755e-07, | |
| "loss": 0.0102, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.0798381331721107e-07, | |
| "loss": 0.0113, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 1.0054723495346482e-07, | |
| "loss": 0.0105, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 9.331100255592436e-08, | |
| "loss": 0.0093, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 8.628481651367875e-08, | |
| "loss": 0.0135, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 7.947809564230445e-08, | |
| "loss": 0.0151, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 7.289996455765748e-08, | |
| "loss": 0.0181, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 6.655924144404906e-08, | |
| "loss": 0.0121, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 6.046442623320145e-08, | |
| "loss": 0.0118, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 5.4623689209832484e-08, | |
| "loss": 0.0164, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 4.904486005914027e-08, | |
| "loss": 0.0112, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 4.373541737087263e-08, | |
| "loss": 0.0115, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.05715640261769295, | |
| "eval_runtime": 21.9917, | |
| "eval_samples_per_second": 9.367, | |
| "eval_steps_per_second": 0.409, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 3.8702478614051345e-08, | |
| "loss": 0.012, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 3.3952790595787986e-08, | |
| "loss": 0.0049, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 2.9492720416985e-08, | |
| "loss": 0.0083, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 2.5328246937043525e-08, | |
| "loss": 0.0085, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 2.1464952759020856e-08, | |
| "loss": 0.009, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 1.7908016745981856e-08, | |
| "loss": 0.0095, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 1.4662207078575684e-08, | |
| "loss": 0.0071, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 1.1731874863145142e-08, | |
| "loss": 0.0067, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 9.12094829893642e-09, | |
| "loss": 0.0073, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 6.832927412229017e-09, | |
| "loss": 0.0103, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 4.8708793644441086e-09, | |
| "loss": 0.0056, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 3.2374343405217884e-09, | |
| "loss": 0.0067, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 1.9347820230782295e-09, | |
| "loss": 0.0076, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 9.64668657069706e-10, | |
| "loss": 0.0052, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 3.2839470889836627e-10, | |
| "loss": 0.0067, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 2.6813123097352287e-11, | |
| "loss": 0.0052, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.06608477234840393, | |
| "eval_runtime": 22.1929, | |
| "eval_samples_per_second": 9.282, | |
| "eval_steps_per_second": 0.406, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 954, | |
| "total_flos": 126473593159680.0, | |
| "train_loss": 0.16594175518777385, | |
| "train_runtime": 9321.3915, | |
| "train_samples_per_second": 2.443, | |
| "train_steps_per_second": 0.102 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 954, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 126473593159680.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |