| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 75.1879934828926, |
| "eval_steps": 2000, |
| "global_step": 150000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0, |
| "eval_accuracy": 1.3660035857594126e-07, |
| "eval_loss": 132.875, |
| "eval_runtime": 254.898, |
| "eval_samples_per_second": 6486.396, |
| "eval_steps_per_second": 12.672, |
| "step": 0 |
| }, |
| { |
| "epoch": 0.13762730525736305, |
| "grad_norm": 49.58098602294922, |
| "learning_rate": 2.967e-05, |
| "loss": 178.872046875, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.2752546105147261, |
| "grad_norm": 40.552101135253906, |
| "learning_rate": 5.966999999999999e-05, |
| "loss": 82.2545625, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.2752546105147261, |
| "eval_accuracy": 0.6086885594122716, |
| "eval_loss": 16.890625, |
| "eval_runtime": 245.8339, |
| "eval_samples_per_second": 6725.555, |
| "eval_steps_per_second": 13.139, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.41288191577208916, |
| "grad_norm": 37.14718246459961, |
| "learning_rate": 8.966999999999999e-05, |
| "loss": 66.408390625, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.5505092210294522, |
| "grad_norm": 32.7165641784668, |
| "learning_rate": 0.00011960999999999999, |
| "loss": 59.63419921875, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.5505092210294522, |
| "eval_accuracy": 0.6626575571303744, |
| "eval_loss": 13.6171875, |
| "eval_runtime": 239.1422, |
| "eval_samples_per_second": 6913.748, |
| "eval_steps_per_second": 13.507, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.6881365262868153, |
| "grad_norm": 26.651721954345703, |
| "learning_rate": 0.00014960999999999997, |
| "loss": 55.59596484375, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.8257638315441783, |
| "grad_norm": 25.13609504699707, |
| "learning_rate": 0.00017961, |
| "loss": 52.9440390625, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.8257638315441783, |
| "eval_accuracy": 0.6820435002667244, |
| "eval_loss": 12.4453125, |
| "eval_runtime": 241.0405, |
| "eval_samples_per_second": 6859.298, |
| "eval_steps_per_second": 13.4, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.9633911368015414, |
| "grad_norm": 25.540454864501953, |
| "learning_rate": 0.00020960999999999997, |
| "loss": 51.0653359375, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.1010184420589044, |
| "grad_norm": 22.598819732666016, |
| "learning_rate": 0.00023960999999999996, |
| "loss": 49.669171875, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.1010184420589044, |
| "eval_accuracy": 0.6911625811516694, |
| "eval_loss": 11.875, |
| "eval_runtime": 238.8748, |
| "eval_samples_per_second": 6921.488, |
| "eval_steps_per_second": 13.522, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.2386457473162675, |
| "grad_norm": 21.2167911529541, |
| "learning_rate": 0.00026957999999999995, |
| "loss": 48.713546875, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.3762730525736306, |
| "grad_norm": 20.751371383666992, |
| "learning_rate": 0.00029955, |
| "loss": 47.99215625, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.3762730525736306, |
| "eval_accuracy": 0.6964798957051618, |
| "eval_loss": 11.5546875, |
| "eval_runtime": 240.5455, |
| "eval_samples_per_second": 6873.416, |
| "eval_steps_per_second": 13.428, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.5139003578309937, |
| "grad_norm": 21.686861038208008, |
| "learning_rate": 0.0003, |
| "loss": 47.29948046875, |
| "step": 11000 |
| }, |
| { |
| "epoch": 1.6515276630883569, |
| "grad_norm": 18.800752639770508, |
| "learning_rate": 0.0003, |
| "loss": 46.53960546875, |
| "step": 12000 |
| }, |
| { |
| "epoch": 1.6515276630883569, |
| "eval_accuracy": 0.701906575930677, |
| "eval_loss": 11.2265625, |
| "eval_runtime": 239.4873, |
| "eval_samples_per_second": 6903.785, |
| "eval_steps_per_second": 13.487, |
| "step": 12000 |
| }, |
| { |
| "epoch": 1.7891549683457197, |
| "grad_norm": 19.42099952697754, |
| "learning_rate": 0.0003, |
| "loss": 45.9191953125, |
| "step": 13000 |
| }, |
| { |
| "epoch": 1.9267822736030829, |
| "grad_norm": 19.15869140625, |
| "learning_rate": 0.0003, |
| "loss": 45.381796875, |
| "step": 14000 |
| }, |
| { |
| "epoch": 1.9267822736030829, |
| "eval_accuracy": 0.7061886218301343, |
| "eval_loss": 10.9765625, |
| "eval_runtime": 239.2568, |
| "eval_samples_per_second": 6910.438, |
| "eval_steps_per_second": 13.5, |
| "step": 14000 |
| }, |
| { |
| "epoch": 2.0644095788604457, |
| "grad_norm": 16.94078826904297, |
| "learning_rate": 0.0003, |
| "loss": 44.90976953125, |
| "step": 15000 |
| }, |
| { |
| "epoch": 2.202036884117809, |
| "grad_norm": 17.655250549316406, |
| "learning_rate": 0.0003, |
| "loss": 44.45808203125, |
| "step": 16000 |
| }, |
| { |
| "epoch": 2.202036884117809, |
| "eval_accuracy": 0.7093567074531988, |
| "eval_loss": 10.8203125, |
| "eval_runtime": 241.1911, |
| "eval_samples_per_second": 6855.016, |
| "eval_steps_per_second": 13.392, |
| "step": 16000 |
| }, |
| { |
| "epoch": 2.339664189375172, |
| "grad_norm": 17.694721221923828, |
| "learning_rate": 0.0003, |
| "loss": 44.16196484375, |
| "step": 17000 |
| }, |
| { |
| "epoch": 2.477291494632535, |
| "grad_norm": 17.49053955078125, |
| "learning_rate": 0.0003, |
| "loss": 43.84825, |
| "step": 18000 |
| }, |
| { |
| "epoch": 2.477291494632535, |
| "eval_accuracy": 0.7115680703757034, |
| "eval_loss": 10.6640625, |
| "eval_runtime": 239.4688, |
| "eval_samples_per_second": 6904.32, |
| "eval_steps_per_second": 13.488, |
| "step": 18000 |
| }, |
| { |
| "epoch": 2.614918799889898, |
| "grad_norm": 19.09914207458496, |
| "learning_rate": 0.0003, |
| "loss": 43.59271875, |
| "step": 19000 |
| }, |
| { |
| "epoch": 2.7525461051472613, |
| "grad_norm": 16.3907527923584, |
| "learning_rate": 0.0003, |
| "loss": 43.352640625, |
| "step": 20000 |
| }, |
| { |
| "epoch": 2.7525461051472613, |
| "eval_accuracy": 0.7139675040013439, |
| "eval_loss": 10.5546875, |
| "eval_runtime": 238.1647, |
| "eval_samples_per_second": 6942.126, |
| "eval_steps_per_second": 13.562, |
| "step": 20000 |
| }, |
| { |
| "epoch": 2.8901734104046244, |
| "grad_norm": 15.896549224853516, |
| "learning_rate": 0.0003, |
| "loss": 43.17196875, |
| "step": 21000 |
| }, |
| { |
| "epoch": 3.0278007156619875, |
| "grad_norm": 29.67310905456543, |
| "learning_rate": 0.0003, |
| "loss": 42.92155859375, |
| "step": 22000 |
| }, |
| { |
| "epoch": 3.0278007156619875, |
| "eval_accuracy": 0.7146398068421496, |
| "eval_loss": 10.484375, |
| "eval_runtime": 238.4701, |
| "eval_samples_per_second": 6933.233, |
| "eval_steps_per_second": 13.545, |
| "step": 22000 |
| }, |
| { |
| "epoch": 3.1654280209193506, |
| "grad_norm": 16.424579620361328, |
| "learning_rate": 0.0003, |
| "loss": 42.65390625, |
| "step": 23000 |
| }, |
| { |
| "epoch": 3.3030553261767133, |
| "grad_norm": 16.19496726989746, |
| "learning_rate": 0.0003, |
| "loss": 42.48802734375, |
| "step": 24000 |
| }, |
| { |
| "epoch": 3.3030553261767133, |
| "eval_accuracy": 0.7170818189501579, |
| "eval_loss": 10.3671875, |
| "eval_runtime": 240.7935, |
| "eval_samples_per_second": 6866.337, |
| "eval_steps_per_second": 13.414, |
| "step": 24000 |
| }, |
| { |
| "epoch": 3.4406826314340764, |
| "grad_norm": 15.53753662109375, |
| "learning_rate": 0.0003, |
| "loss": 42.357984375, |
| "step": 25000 |
| }, |
| { |
| "epoch": 3.5783099366914395, |
| "grad_norm": 16.701377868652344, |
| "learning_rate": 0.0003, |
| "loss": 42.1965703125, |
| "step": 26000 |
| }, |
| { |
| "epoch": 3.5783099366914395, |
| "eval_accuracy": 0.7182790131411184, |
| "eval_loss": 10.2890625, |
| "eval_runtime": 241.2004, |
| "eval_samples_per_second": 6854.753, |
| "eval_steps_per_second": 13.391, |
| "step": 26000 |
| }, |
| { |
| "epoch": 3.7159372419488026, |
| "grad_norm": 15.334391593933105, |
| "learning_rate": 0.0003, |
| "loss": 42.05885546875, |
| "step": 27000 |
| }, |
| { |
| "epoch": 3.8535645472061657, |
| "grad_norm": 15.341226577758789, |
| "learning_rate": 0.0003, |
| "loss": 41.9392421875, |
| "step": 28000 |
| }, |
| { |
| "epoch": 3.8535645472061657, |
| "eval_accuracy": 0.719322972712139, |
| "eval_loss": 10.2421875, |
| "eval_runtime": 241.186, |
| "eval_samples_per_second": 6855.162, |
| "eval_steps_per_second": 13.392, |
| "step": 28000 |
| }, |
| { |
| "epoch": 3.991191852463529, |
| "grad_norm": 16.253334045410156, |
| "learning_rate": 0.0003, |
| "loss": 41.81163671875, |
| "step": 29000 |
| }, |
| { |
| "epoch": 4.1288191577208915, |
| "grad_norm": 15.035149574279785, |
| "learning_rate": 0.0003, |
| "loss": 41.617953125, |
| "step": 30000 |
| }, |
| { |
| "epoch": 4.1288191577208915, |
| "eval_accuracy": 0.720451396648655, |
| "eval_loss": 10.171875, |
| "eval_runtime": 240.2855, |
| "eval_samples_per_second": 6880.853, |
| "eval_steps_per_second": 13.442, |
| "step": 30000 |
| }, |
| { |
| "epoch": 4.266446462978255, |
| "grad_norm": 14.762296676635742, |
| "learning_rate": 0.0003, |
| "loss": 41.5138203125, |
| "step": 31000 |
| }, |
| { |
| "epoch": 4.404073768235618, |
| "grad_norm": 14.627701759338379, |
| "learning_rate": 0.0003, |
| "loss": 41.4306015625, |
| "step": 32000 |
| }, |
| { |
| "epoch": 4.404073768235618, |
| "eval_accuracy": 0.7213651673804347, |
| "eval_loss": 10.1328125, |
| "eval_runtime": 242.2962, |
| "eval_samples_per_second": 6823.752, |
| "eval_steps_per_second": 13.331, |
| "step": 32000 |
| }, |
| { |
| "epoch": 4.541701073492981, |
| "grad_norm": 14.57941722869873, |
| "learning_rate": 0.0003, |
| "loss": 41.3221171875, |
| "step": 33000 |
| }, |
| { |
| "epoch": 4.679328378750344, |
| "grad_norm": 15.291731834411621, |
| "learning_rate": 0.0003, |
| "loss": 41.276203125, |
| "step": 34000 |
| }, |
| { |
| "epoch": 4.679328378750344, |
| "eval_accuracy": 0.7223401458779132, |
| "eval_loss": 10.0703125, |
| "eval_runtime": 239.5218, |
| "eval_samples_per_second": 6902.793, |
| "eval_steps_per_second": 13.485, |
| "step": 34000 |
| }, |
| { |
| "epoch": 4.8169556840077075, |
| "grad_norm": 15.057552337646484, |
| "learning_rate": 0.0003, |
| "loss": 41.19701171875, |
| "step": 35000 |
| }, |
| { |
| "epoch": 4.95458298926507, |
| "grad_norm": 15.457907676696777, |
| "learning_rate": 0.0003, |
| "loss": 41.10438671875, |
| "step": 36000 |
| }, |
| { |
| "epoch": 4.95458298926507, |
| "eval_accuracy": 0.7230995451445803, |
| "eval_loss": 10.0546875, |
| "eval_runtime": 240.0931, |
| "eval_samples_per_second": 6886.367, |
| "eval_steps_per_second": 13.453, |
| "step": 36000 |
| }, |
| { |
| "epoch": 5.092210294522434, |
| "grad_norm": 15.539594650268555, |
| "learning_rate": 0.0003, |
| "loss": 40.93646875, |
| "step": 37000 |
| }, |
| { |
| "epoch": 5.229837599779796, |
| "grad_norm": 14.915628433227539, |
| "learning_rate": 0.0003, |
| "loss": 40.8286875, |
| "step": 38000 |
| }, |
| { |
| "epoch": 5.229837599779796, |
| "eval_accuracy": 0.7238966529952525, |
| "eval_loss": 10.0, |
| "eval_runtime": 239.9098, |
| "eval_samples_per_second": 6891.629, |
| "eval_steps_per_second": 13.463, |
| "step": 38000 |
| }, |
| { |
| "epoch": 5.367464905037159, |
| "grad_norm": 14.271048545837402, |
| "learning_rate": 0.0003, |
| "loss": 40.80625, |
| "step": 39000 |
| }, |
| { |
| "epoch": 5.505092210294523, |
| "grad_norm": 14.605119705200195, |
| "learning_rate": 0.0003, |
| "loss": 40.713796875, |
| "step": 40000 |
| }, |
| { |
| "epoch": 5.505092210294523, |
| "eval_accuracy": 0.7245612679427045, |
| "eval_loss": 9.9609375, |
| "eval_runtime": 240.0103, |
| "eval_samples_per_second": 6888.743, |
| "eval_steps_per_second": 13.458, |
| "step": 40000 |
| }, |
| { |
| "epoch": 5.642719515551885, |
| "grad_norm": 14.748287200927734, |
| "learning_rate": 0.0003, |
| "loss": 40.62338671875, |
| "step": 41000 |
| }, |
| { |
| "epoch": 5.780346820809249, |
| "grad_norm": 15.422652244567871, |
| "learning_rate": 0.0003, |
| "loss": 40.56144140625, |
| "step": 42000 |
| }, |
| { |
| "epoch": 5.780346820809249, |
| "eval_accuracy": 0.7251576961266964, |
| "eval_loss": 9.9375, |
| "eval_runtime": 240.638, |
| "eval_samples_per_second": 6870.772, |
| "eval_steps_per_second": 13.423, |
| "step": 42000 |
| }, |
| { |
| "epoch": 5.917974126066611, |
| "grad_norm": 15.326558113098145, |
| "learning_rate": 0.0003, |
| "loss": 40.5059375, |
| "step": 43000 |
| }, |
| { |
| "epoch": 6.055601431323975, |
| "grad_norm": 15.331598281860352, |
| "learning_rate": 0.0003, |
| "loss": 40.40818359375, |
| "step": 44000 |
| }, |
| { |
| "epoch": 6.055601431323975, |
| "eval_accuracy": 0.7254487554600376, |
| "eval_loss": 9.8984375, |
| "eval_runtime": 240.1599, |
| "eval_samples_per_second": 6884.449, |
| "eval_steps_per_second": 13.449, |
| "step": 44000 |
| }, |
| { |
| "epoch": 6.193228736581338, |
| "grad_norm": 14.527973175048828, |
| "learning_rate": 0.0003, |
| "loss": 40.3428828125, |
| "step": 45000 |
| }, |
| { |
| "epoch": 6.330856041838701, |
| "grad_norm": 15.686996459960938, |
| "learning_rate": 0.0003, |
| "loss": 40.3244765625, |
| "step": 46000 |
| }, |
| { |
| "epoch": 6.330856041838701, |
| "eval_accuracy": 0.7256079479674087, |
| "eval_loss": 9.8984375, |
| "eval_runtime": 239.085, |
| "eval_samples_per_second": 6915.403, |
| "eval_steps_per_second": 13.51, |
| "step": 46000 |
| }, |
| { |
| "epoch": 6.468483347096064, |
| "grad_norm": 14.848986625671387, |
| "learning_rate": 0.0003, |
| "loss": 40.312796875, |
| "step": 47000 |
| }, |
| { |
| "epoch": 6.6061106523534265, |
| "grad_norm": 14.275111198425293, |
| "learning_rate": 0.0003, |
| "loss": 40.28499609375, |
| "step": 48000 |
| }, |
| { |
| "epoch": 6.6061106523534265, |
| "eval_accuracy": 0.7262142861047188, |
| "eval_loss": 9.875, |
| "eval_runtime": 240.5807, |
| "eval_samples_per_second": 6872.409, |
| "eval_steps_per_second": 13.426, |
| "step": 48000 |
| }, |
| { |
| "epoch": 6.74373795761079, |
| "grad_norm": 14.665587425231934, |
| "learning_rate": 0.0003, |
| "loss": 40.18369921875, |
| "step": 49000 |
| }, |
| { |
| "epoch": 6.881365262868153, |
| "grad_norm": 14.547246932983398, |
| "learning_rate": 0.0003, |
| "loss": 40.1498828125, |
| "step": 50000 |
| }, |
| { |
| "epoch": 6.881365262868153, |
| "eval_accuracy": 0.7269716959581425, |
| "eval_loss": 9.8515625, |
| "eval_runtime": 241.4549, |
| "eval_samples_per_second": 6847.527, |
| "eval_steps_per_second": 13.377, |
| "step": 50000 |
| }, |
| { |
| "epoch": 7.018992568125516, |
| "grad_norm": 14.525768280029297, |
| "learning_rate": 0.0003, |
| "loss": 40.1036328125, |
| "step": 51000 |
| }, |
| { |
| "epoch": 7.156619873382879, |
| "grad_norm": 14.632113456726074, |
| "learning_rate": 0.0003, |
| "loss": 39.9834296875, |
| "step": 52000 |
| }, |
| { |
| "epoch": 7.156619873382879, |
| "eval_accuracy": 0.7272316426626143, |
| "eval_loss": 9.828125, |
| "eval_runtime": 239.3181, |
| "eval_samples_per_second": 6908.667, |
| "eval_steps_per_second": 13.497, |
| "step": 52000 |
| }, |
| { |
| "epoch": 7.2942471786402425, |
| "grad_norm": 14.982499122619629, |
| "learning_rate": 0.0003, |
| "loss": 39.9509375, |
| "step": 53000 |
| }, |
| { |
| "epoch": 7.431874483897605, |
| "grad_norm": 16.801025390625, |
| "learning_rate": 0.0003, |
| "loss": 39.891859375, |
| "step": 54000 |
| }, |
| { |
| "epoch": 7.431874483897605, |
| "eval_accuracy": 0.7271305788939304, |
| "eval_loss": 9.828125, |
| "eval_runtime": 240.159, |
| "eval_samples_per_second": 6884.477, |
| "eval_steps_per_second": 13.449, |
| "step": 54000 |
| }, |
| { |
| "epoch": 7.569501789154968, |
| "grad_norm": 14.868009567260742, |
| "learning_rate": 0.0003, |
| "loss": 39.88668359375, |
| "step": 55000 |
| }, |
| { |
| "epoch": 7.707129094412331, |
| "grad_norm": 14.595479011535645, |
| "learning_rate": 0.0003, |
| "loss": 39.821890625, |
| "step": 56000 |
| }, |
| { |
| "epoch": 7.707129094412331, |
| "eval_accuracy": 0.7280901536840519, |
| "eval_loss": 9.7734375, |
| "eval_runtime": 238.9096, |
| "eval_samples_per_second": 6920.478, |
| "eval_steps_per_second": 13.52, |
| "step": 56000 |
| }, |
| { |
| "epoch": 7.844756399669695, |
| "grad_norm": 13.92586612701416, |
| "learning_rate": 0.0003, |
| "loss": 39.78269921875, |
| "step": 57000 |
| }, |
| { |
| "epoch": 7.982383704927058, |
| "grad_norm": 15.85058307647705, |
| "learning_rate": 0.0003, |
| "loss": 39.72277734375, |
| "step": 58000 |
| }, |
| { |
| "epoch": 7.982383704927058, |
| "eval_accuracy": 0.7287356832938983, |
| "eval_loss": 9.7578125, |
| "eval_runtime": 239.7822, |
| "eval_samples_per_second": 6895.295, |
| "eval_steps_per_second": 13.471, |
| "step": 58000 |
| }, |
| { |
| "epoch": 8.12001101018442, |
| "grad_norm": 15.202603340148926, |
| "learning_rate": 0.0003, |
| "loss": 39.6687421875, |
| "step": 59000 |
| }, |
| { |
| "epoch": 8.257638315441783, |
| "grad_norm": 14.994338989257812, |
| "learning_rate": 0.0003, |
| "loss": 39.60739453125, |
| "step": 60000 |
| }, |
| { |
| "epoch": 8.257638315441783, |
| "eval_accuracy": 0.7289930926403759, |
| "eval_loss": 9.7265625, |
| "eval_runtime": 241.1318, |
| "eval_samples_per_second": 6856.702, |
| "eval_steps_per_second": 13.395, |
| "step": 60000 |
| }, |
| { |
| "epoch": 8.395265620699147, |
| "grad_norm": 15.15245532989502, |
| "learning_rate": 0.0003, |
| "loss": 39.57180859375, |
| "step": 61000 |
| }, |
| { |
| "epoch": 8.53289292595651, |
| "grad_norm": 15.941924095153809, |
| "learning_rate": 0.0003, |
| "loss": 39.5704296875, |
| "step": 62000 |
| }, |
| { |
| "epoch": 8.53289292595651, |
| "eval_accuracy": 0.7289831970926051, |
| "eval_loss": 9.734375, |
| "eval_runtime": 241.0009, |
| "eval_samples_per_second": 6860.426, |
| "eval_steps_per_second": 13.402, |
| "step": 62000 |
| }, |
| { |
| "epoch": 8.670520231213873, |
| "grad_norm": 14.842296600341797, |
| "learning_rate": 0.0003, |
| "loss": 39.53778125, |
| "step": 63000 |
| }, |
| { |
| "epoch": 8.808147536471235, |
| "grad_norm": 17.454763412475586, |
| "learning_rate": 0.0003, |
| "loss": 39.540921875, |
| "step": 64000 |
| }, |
| { |
| "epoch": 8.808147536471235, |
| "eval_accuracy": 0.7290886771189041, |
| "eval_loss": 9.7109375, |
| "eval_runtime": 240.0306, |
| "eval_samples_per_second": 6888.158, |
| "eval_steps_per_second": 13.457, |
| "step": 64000 |
| }, |
| { |
| "epoch": 8.9457748417286, |
| "grad_norm": 13.98570442199707, |
| "learning_rate": 0.0003, |
| "loss": 39.512796875, |
| "step": 65000 |
| }, |
| { |
| "epoch": 9.083402146985962, |
| "grad_norm": 18.010318756103516, |
| "learning_rate": 0.0003, |
| "loss": 39.4786171875, |
| "step": 66000 |
| }, |
| { |
| "epoch": 9.083402146985962, |
| "eval_accuracy": 0.7294749784251455, |
| "eval_loss": 9.7109375, |
| "eval_runtime": 240.1115, |
| "eval_samples_per_second": 6885.838, |
| "eval_steps_per_second": 13.452, |
| "step": 66000 |
| }, |
| { |
| "epoch": 9.221029452243325, |
| "grad_norm": 15.137900352478027, |
| "learning_rate": 0.0003, |
| "loss": 39.4073359375, |
| "step": 67000 |
| }, |
| { |
| "epoch": 9.358656757500688, |
| "grad_norm": 18.228130340576172, |
| "learning_rate": 0.0003, |
| "loss": 39.3549765625, |
| "step": 68000 |
| }, |
| { |
| "epoch": 9.358656757500688, |
| "eval_accuracy": 0.7301181665976199, |
| "eval_loss": 9.671875, |
| "eval_runtime": 239.7862, |
| "eval_samples_per_second": 6895.179, |
| "eval_steps_per_second": 13.47, |
| "step": 68000 |
| }, |
| { |
| "epoch": 9.49628406275805, |
| "grad_norm": 16.575559616088867, |
| "learning_rate": 0.0003, |
| "loss": 39.3098828125, |
| "step": 69000 |
| }, |
| { |
| "epoch": 9.633911368015415, |
| "grad_norm": 14.635740280151367, |
| "learning_rate": 0.0003, |
| "loss": 39.35287890625, |
| "step": 70000 |
| }, |
| { |
| "epoch": 9.633911368015415, |
| "eval_accuracy": 0.7295560826467988, |
| "eval_loss": 9.6875, |
| "eval_runtime": 241.3494, |
| "eval_samples_per_second": 6850.521, |
| "eval_steps_per_second": 13.383, |
| "step": 70000 |
| }, |
| { |
| "epoch": 9.771538673272778, |
| "grad_norm": 14.436244010925293, |
| "learning_rate": 0.0003, |
| "loss": 39.29956640625, |
| "step": 71000 |
| }, |
| { |
| "epoch": 9.90916597853014, |
| "grad_norm": 14.493698120117188, |
| "learning_rate": 0.0003, |
| "loss": 39.31009375, |
| "step": 72000 |
| }, |
| { |
| "epoch": 9.90916597853014, |
| "eval_accuracy": 0.7304579509385183, |
| "eval_loss": 9.6484375, |
| "eval_runtime": 245.6684, |
| "eval_samples_per_second": 6730.085, |
| "eval_steps_per_second": 13.148, |
| "step": 72000 |
| }, |
| { |
| "epoch": 10.046793283787503, |
| "grad_norm": 15.077356338500977, |
| "learning_rate": 0.0003, |
| "loss": 39.2335546875, |
| "step": 73000 |
| }, |
| { |
| "epoch": 10.184420589044867, |
| "grad_norm": 13.661473274230957, |
| "learning_rate": 0.0003, |
| "loss": 39.09965625, |
| "step": 74000 |
| }, |
| { |
| "epoch": 10.184420589044867, |
| "eval_accuracy": 0.7312455280822778, |
| "eval_loss": 9.625, |
| "eval_runtime": 239.6349, |
| "eval_samples_per_second": 6899.534, |
| "eval_steps_per_second": 13.479, |
| "step": 74000 |
| }, |
| { |
| "epoch": 10.32204789430223, |
| "grad_norm": 15.429136276245117, |
| "learning_rate": 0.0003, |
| "loss": 39.147140625, |
| "step": 75000 |
| }, |
| { |
| "epoch": 10.459675199559593, |
| "grad_norm": 15.229757308959961, |
| "learning_rate": 0.0003, |
| "loss": 39.1339453125, |
| "step": 76000 |
| }, |
| { |
| "epoch": 10.459675199559593, |
| "eval_accuracy": 0.731277762807936, |
| "eval_loss": 9.609375, |
| "eval_runtime": 238.7439, |
| "eval_samples_per_second": 6925.282, |
| "eval_steps_per_second": 13.529, |
| "step": 76000 |
| }, |
| { |
| "epoch": 10.597302504816955, |
| "grad_norm": 14.771382331848145, |
| "learning_rate": 0.0003, |
| "loss": 39.1441796875, |
| "step": 77000 |
| }, |
| { |
| "epoch": 10.734929810074318, |
| "grad_norm": 13.703607559204102, |
| "learning_rate": 0.0003, |
| "loss": 39.141265625, |
| "step": 78000 |
| }, |
| { |
| "epoch": 10.734929810074318, |
| "eval_accuracy": 0.7310708531463442, |
| "eval_loss": 9.609375, |
| "eval_runtime": 239.4162, |
| "eval_samples_per_second": 6905.836, |
| "eval_steps_per_second": 13.491, |
| "step": 78000 |
| }, |
| { |
| "epoch": 10.872557115331682, |
| "grad_norm": 19.041141510009766, |
| "learning_rate": 0.0003, |
| "loss": 39.0934140625, |
| "step": 79000 |
| }, |
| { |
| "epoch": 11.010184420589045, |
| "grad_norm": 17.401290893554688, |
| "learning_rate": 0.0003, |
| "loss": 39.113875, |
| "step": 80000 |
| }, |
| { |
| "epoch": 11.010184420589045, |
| "eval_accuracy": 0.7312406631974454, |
| "eval_loss": 9.6015625, |
| "eval_runtime": 238.9522, |
| "eval_samples_per_second": 6919.246, |
| "eval_steps_per_second": 13.517, |
| "step": 80000 |
| }, |
| { |
| "epoch": 11.147811725846408, |
| "grad_norm": 14.292427062988281, |
| "learning_rate": 0.0003, |
| "loss": 39.012484375, |
| "step": 81000 |
| }, |
| { |
| "epoch": 11.28543903110377, |
| "grad_norm": 15.462931632995605, |
| "learning_rate": 0.0003, |
| "loss": 39.04391796875, |
| "step": 82000 |
| }, |
| { |
| "epoch": 11.28543903110377, |
| "eval_accuracy": 0.7316027472794033, |
| "eval_loss": 9.6015625, |
| "eval_runtime": 240.4477, |
| "eval_samples_per_second": 6876.21, |
| "eval_steps_per_second": 13.433, |
| "step": 82000 |
| }, |
| { |
| "epoch": 11.423066336361135, |
| "grad_norm": 17.796772003173828, |
| "learning_rate": 0.0003, |
| "loss": 38.957421875, |
| "step": 83000 |
| }, |
| { |
| "epoch": 11.560693641618498, |
| "grad_norm": 17.314067840576172, |
| "learning_rate": 0.0003, |
| "loss": 38.9495234375, |
| "step": 84000 |
| }, |
| { |
| "epoch": 11.560693641618498, |
| "eval_accuracy": 0.7321146855990825, |
| "eval_loss": 9.578125, |
| "eval_runtime": 239.2384, |
| "eval_samples_per_second": 6910.967, |
| "eval_steps_per_second": 13.501, |
| "step": 84000 |
| }, |
| { |
| "epoch": 11.69832094687586, |
| "grad_norm": 16.145645141601562, |
| "learning_rate": 0.0003, |
| "loss": 38.91906640625, |
| "step": 85000 |
| }, |
| { |
| "epoch": 11.835948252133223, |
| "grad_norm": 13.51314640045166, |
| "learning_rate": 0.0003, |
| "loss": 38.91014453125, |
| "step": 86000 |
| }, |
| { |
| "epoch": 11.835948252133223, |
| "eval_accuracy": 0.732051599943418, |
| "eval_loss": 9.5546875, |
| "eval_runtime": 240.3949, |
| "eval_samples_per_second": 6877.722, |
| "eval_steps_per_second": 13.436, |
| "step": 86000 |
| }, |
| { |
| "epoch": 11.973575557390586, |
| "grad_norm": 15.877927780151367, |
| "learning_rate": 0.0003, |
| "loss": 38.933609375, |
| "step": 87000 |
| }, |
| { |
| "epoch": 12.11120286264795, |
| "grad_norm": 15.215489387512207, |
| "learning_rate": 0.0003, |
| "loss": 38.8452265625, |
| "step": 88000 |
| }, |
| { |
| "epoch": 12.11120286264795, |
| "eval_accuracy": 0.7323534973774022, |
| "eval_loss": 9.546875, |
| "eval_runtime": 240.1188, |
| "eval_samples_per_second": 6885.629, |
| "eval_steps_per_second": 13.452, |
| "step": 88000 |
| }, |
| { |
| "epoch": 12.248830167905313, |
| "grad_norm": 15.539190292358398, |
| "learning_rate": 0.0003, |
| "loss": 38.8104296875, |
| "step": 89000 |
| }, |
| { |
| "epoch": 12.386457473162675, |
| "grad_norm": 15.577831268310547, |
| "learning_rate": 0.0003, |
| "loss": 38.80796875, |
| "step": 90000 |
| }, |
| { |
| "epoch": 12.386457473162675, |
| "eval_accuracy": 0.7324531616408847, |
| "eval_loss": 9.546875, |
| "eval_runtime": 240.2465, |
| "eval_samples_per_second": 6881.969, |
| "eval_steps_per_second": 13.445, |
| "step": 90000 |
| }, |
| { |
| "epoch": 12.524084778420038, |
| "grad_norm": 14.47063159942627, |
| "learning_rate": 0.0003, |
| "loss": 38.865859375, |
| "step": 91000 |
| }, |
| { |
| "epoch": 12.661712083677402, |
| "grad_norm": 13.968493461608887, |
| "learning_rate": 0.0003, |
| "loss": 38.81719921875, |
| "step": 92000 |
| }, |
| { |
| "epoch": 12.661712083677402, |
| "eval_accuracy": 0.7321305936040334, |
| "eval_loss": 9.546875, |
| "eval_runtime": 239.1976, |
| "eval_samples_per_second": 6912.148, |
| "eval_steps_per_second": 13.503, |
| "step": 92000 |
| }, |
| { |
| "epoch": 12.799339388934765, |
| "grad_norm": 28.390636444091797, |
| "learning_rate": 0.0003, |
| "loss": 38.815578125, |
| "step": 93000 |
| }, |
| { |
| "epoch": 12.936966694192128, |
| "grad_norm": 27.102386474609375, |
| "learning_rate": 0.0003, |
| "loss": 38.82604296875, |
| "step": 94000 |
| }, |
| { |
| "epoch": 12.936966694192128, |
| "eval_accuracy": 0.732027704335829, |
| "eval_loss": 9.546875, |
| "eval_runtime": 240.3497, |
| "eval_samples_per_second": 6879.014, |
| "eval_steps_per_second": 13.439, |
| "step": 94000 |
| }, |
| { |
| "epoch": 13.07459399944949, |
| "grad_norm": 14.193507194519043, |
| "learning_rate": 0.0003, |
| "loss": 38.72788671875, |
| "step": 95000 |
| }, |
| { |
| "epoch": 13.212221304706853, |
| "grad_norm": 18.604595184326172, |
| "learning_rate": 0.0003, |
| "loss": 38.6876171875, |
| "step": 96000 |
| }, |
| { |
| "epoch": 13.212221304706853, |
| "eval_accuracy": 0.7321750300843878, |
| "eval_loss": 9.546875, |
| "eval_runtime": 240.1101, |
| "eval_samples_per_second": 6885.879, |
| "eval_steps_per_second": 13.452, |
| "step": 96000 |
| }, |
| { |
| "epoch": 13.349848609964218, |
| "grad_norm": 16.717756271362305, |
| "learning_rate": 0.0003, |
| "loss": 38.7415390625, |
| "step": 97000 |
| }, |
| { |
| "epoch": 13.48747591522158, |
| "grad_norm": 13.74322509765625, |
| "learning_rate": 0.0003, |
| "loss": 38.704234375, |
| "step": 98000 |
| }, |
| { |
| "epoch": 13.48747591522158, |
| "eval_accuracy": 0.7335116918906991, |
| "eval_loss": 9.4921875, |
| "eval_runtime": 240.4214, |
| "eval_samples_per_second": 6876.962, |
| "eval_steps_per_second": 13.435, |
| "step": 98000 |
| }, |
| { |
| "epoch": 13.625103220478943, |
| "grad_norm": 17.836227416992188, |
| "learning_rate": 0.0003, |
| "loss": 38.6647890625, |
| "step": 99000 |
| }, |
| { |
| "epoch": 13.762730525736306, |
| "grad_norm": 20.256298065185547, |
| "learning_rate": 0.0003, |
| "loss": 38.654390625, |
| "step": 100000 |
| }, |
| { |
| "epoch": 13.762730525736306, |
| "eval_accuracy": 0.7328628073699861, |
| "eval_loss": 9.5078125, |
| "eval_runtime": 240.901, |
| "eval_samples_per_second": 6863.272, |
| "eval_steps_per_second": 13.408, |
| "step": 100000 |
| }, |
| { |
| "epoch": 13.762730525736306, |
| "eval_accuracy": 0.7328718517886109, |
| "eval_loss": 9.5078125, |
| "eval_runtime": 257.3667, |
| "eval_samples_per_second": 6424.176, |
| "eval_steps_per_second": 12.55, |
| "step": 100000 |
| }, |
| { |
| "epoch": 13.90035783099367, |
| "grad_norm": 17.175275802612305, |
| "learning_rate": 0.0003, |
| "loss": 38.7094453125, |
| "step": 101000 |
| }, |
| { |
| "epoch": 14.037985136251033, |
| "grad_norm": 30.791107177734375, |
| "learning_rate": 0.0003, |
| "loss": 38.7431796875, |
| "step": 102000 |
| }, |
| { |
| "epoch": 14.037985136251033, |
| "eval_accuracy": 0.732853775052298, |
| "eval_loss": 9.53125, |
| "eval_runtime": 244.7843, |
| "eval_samples_per_second": 6754.392, |
| "eval_steps_per_second": 13.195, |
| "step": 102000 |
| }, |
| { |
| "epoch": 14.175612441508395, |
| "grad_norm": 15.07434368133545, |
| "learning_rate": 0.0003, |
| "loss": 38.5621875, |
| "step": 103000 |
| }, |
| { |
| "epoch": 14.313239746765758, |
| "grad_norm": 16.333436965942383, |
| "learning_rate": 0.0003, |
| "loss": 38.6172734375, |
| "step": 104000 |
| }, |
| { |
| "epoch": 14.313239746765758, |
| "eval_accuracy": 0.7328456338360237, |
| "eval_loss": 9.515625, |
| "eval_runtime": 243.7048, |
| "eval_samples_per_second": 6784.311, |
| "eval_steps_per_second": 13.254, |
| "step": 104000 |
| }, |
| { |
| "epoch": 14.45086705202312, |
| "grad_norm": 14.872163772583008, |
| "learning_rate": 0.0003, |
| "loss": 38.61624609375, |
| "step": 105000 |
| }, |
| { |
| "epoch": 14.588494357280485, |
| "grad_norm": 15.491616249084473, |
| "learning_rate": 0.0003, |
| "loss": 38.5978203125, |
| "step": 106000 |
| }, |
| { |
| "epoch": 14.588494357280485, |
| "eval_accuracy": 0.7325266385860558, |
| "eval_loss": 9.53125, |
| "eval_runtime": 241.371, |
| "eval_samples_per_second": 6849.906, |
| "eval_steps_per_second": 13.382, |
| "step": 106000 |
| }, |
| { |
| "epoch": 14.726121662537848, |
| "grad_norm": 14.945006370544434, |
| "learning_rate": 0.0003, |
| "loss": 38.621796875, |
| "step": 107000 |
| }, |
| { |
| "epoch": 14.86374896779521, |
| "grad_norm": 14.714298248291016, |
| "learning_rate": 0.0003, |
| "loss": 38.5805546875, |
| "step": 108000 |
| }, |
| { |
| "epoch": 14.86374896779521, |
| "eval_accuracy": 0.7336863429887471, |
| "eval_loss": 9.484375, |
| "eval_runtime": 243.3178, |
| "eval_samples_per_second": 6795.1, |
| "eval_steps_per_second": 13.275, |
| "step": 108000 |
| }, |
| { |
| "epoch": 15.001376273052573, |
| "grad_norm": 17.513687133789062, |
| "learning_rate": 0.0003, |
| "loss": 38.5988359375, |
| "step": 109000 |
| }, |
| { |
| "epoch": 15.139003578309937, |
| "grad_norm": 14.208888053894043, |
| "learning_rate": 0.0003, |
| "loss": 38.5494453125, |
| "step": 110000 |
| }, |
| { |
| "epoch": 15.139003578309937, |
| "eval_accuracy": 0.7334265583450897, |
| "eval_loss": 9.4921875, |
| "eval_runtime": 245.3975, |
| "eval_samples_per_second": 6737.512, |
| "eval_steps_per_second": 13.162, |
| "step": 110000 |
| }, |
| { |
| "epoch": 15.2766308835673, |
| "grad_norm": 20.13620376586914, |
| "learning_rate": 0.0003, |
| "loss": 38.51769140625, |
| "step": 111000 |
| }, |
| { |
| "epoch": 15.414258188824663, |
| "grad_norm": 14.885974884033203, |
| "learning_rate": 0.0003, |
| "loss": 38.52906640625, |
| "step": 112000 |
| }, |
| { |
| "epoch": 15.414258188824663, |
| "eval_accuracy": 0.7332003955432331, |
| "eval_loss": 9.4921875, |
| "eval_runtime": 246.8519, |
| "eval_samples_per_second": 6697.818, |
| "eval_steps_per_second": 13.085, |
| "step": 112000 |
| }, |
| { |
| "epoch": 15.551885494082025, |
| "grad_norm": 14.931363105773926, |
| "learning_rate": 0.0003, |
| "loss": 38.534203125, |
| "step": 113000 |
| }, |
| { |
| "epoch": 15.689512799339388, |
| "grad_norm": 15.144700050354004, |
| "learning_rate": 0.0003, |
| "loss": 38.5433125, |
| "step": 114000 |
| }, |
| { |
| "epoch": 15.689512799339388, |
| "eval_accuracy": 0.7337025970829132, |
| "eval_loss": 9.4765625, |
| "eval_runtime": 244.85, |
| "eval_samples_per_second": 6752.58, |
| "eval_steps_per_second": 13.192, |
| "step": 114000 |
| }, |
| { |
| "epoch": 15.827140104596753, |
| "grad_norm": 17.183073043823242, |
| "learning_rate": 0.0003, |
| "loss": 38.4901015625, |
| "step": 115000 |
| }, |
| { |
| "epoch": 15.964767409854115, |
| "grad_norm": 14.985239028930664, |
| "learning_rate": 0.0003, |
| "loss": 38.51575390625, |
| "step": 116000 |
| }, |
| { |
| "epoch": 15.964767409854115, |
| "eval_accuracy": 0.7338189696183159, |
| "eval_loss": 9.484375, |
| "eval_runtime": 245.1155, |
| "eval_samples_per_second": 6745.266, |
| "eval_steps_per_second": 13.177, |
| "step": 116000 |
| }, |
| { |
| "epoch": 16.10239471511148, |
| "grad_norm": 19.971887588500977, |
| "learning_rate": 0.0003, |
| "loss": 38.4035234375, |
| "step": 117000 |
| }, |
| { |
| "epoch": 16.24002202036884, |
| "grad_norm": 17.1956844329834, |
| "learning_rate": 0.0003, |
| "loss": 38.42918359375, |
| "step": 118000 |
| }, |
| { |
| "epoch": 16.24002202036884, |
| "eval_accuracy": 0.733730730614503, |
| "eval_loss": 9.46875, |
| "eval_runtime": 243.3012, |
| "eval_samples_per_second": 6795.566, |
| "eval_steps_per_second": 13.276, |
| "step": 118000 |
| }, |
| { |
| "epoch": 16.377649325626205, |
| "grad_norm": 15.118714332580566, |
| "learning_rate": 0.0003, |
| "loss": 38.507515625, |
| "step": 119000 |
| }, |
| { |
| "epoch": 16.515276630883566, |
| "grad_norm": 14.03774642944336, |
| "learning_rate": 0.0003, |
| "loss": 38.526671875, |
| "step": 120000 |
| }, |
| { |
| "epoch": 16.515276630883566, |
| "eval_accuracy": 0.733831136300071, |
| "eval_loss": 9.484375, |
| "eval_runtime": 241.9438, |
| "eval_samples_per_second": 6833.691, |
| "eval_steps_per_second": 13.35, |
| "step": 120000 |
| }, |
| { |
| "epoch": 16.515276630883566, |
| "eval_accuracy": 0.7365382984533457, |
| "eval_loss": 9.328125, |
| "eval_runtime": 320.3822, |
| "eval_samples_per_second": 1416.711, |
| "eval_steps_per_second": 2.769, |
| "step": 120000 |
| }, |
| { |
| "epoch": 60.65171074069432, |
| "grad_norm": 8.715871810913086, |
| "learning_rate": 0.0003, |
| "loss": 34.4526328125, |
| "step": 121000 |
| }, |
| { |
| "epoch": 61.15290136608598, |
| "grad_norm": 16.51197052001953, |
| "learning_rate": 0.0003, |
| "loss": 33.80009375, |
| "step": 122000 |
| }, |
| { |
| "epoch": 61.15290136608598, |
| "eval_accuracy": 0.7592972259433672, |
| "eval_loss": 8.328125, |
| "eval_runtime": 322.1474, |
| "eval_samples_per_second": 1408.948, |
| "eval_steps_per_second": 2.753, |
| "step": 122000 |
| }, |
| { |
| "epoch": 61.65421732046622, |
| "grad_norm": 8.568217277526855, |
| "learning_rate": 0.0003, |
| "loss": 33.51157421875, |
| "step": 123000 |
| }, |
| { |
| "epoch": 62.155407945857874, |
| "grad_norm": 13.904038429260254, |
| "learning_rate": 0.0003, |
| "loss": 33.3759140625, |
| "step": 124000 |
| }, |
| { |
| "epoch": 62.155407945857874, |
| "eval_accuracy": 0.7606993464209245, |
| "eval_loss": 8.2578125, |
| "eval_runtime": 310.5126, |
| "eval_samples_per_second": 1461.741, |
| "eval_steps_per_second": 2.857, |
| "step": 124000 |
| }, |
| { |
| "epoch": 62.65672390023813, |
| "grad_norm": 9.302454948425293, |
| "learning_rate": 0.0003, |
| "loss": 33.2303125, |
| "step": 125000 |
| }, |
| { |
| "epoch": 63.15791452562978, |
| "grad_norm": 10.245097160339355, |
| "learning_rate": 0.0003, |
| "loss": 33.114984375, |
| "step": 126000 |
| }, |
| { |
| "epoch": 63.15791452562978, |
| "eval_accuracy": 0.7620252803249203, |
| "eval_loss": 8.1953125, |
| "eval_runtime": 311.281, |
| "eval_samples_per_second": 1458.133, |
| "eval_steps_per_second": 2.85, |
| "step": 126000 |
| }, |
| { |
| "epoch": 63.659230480010024, |
| "grad_norm": 9.459521293640137, |
| "learning_rate": 0.0003, |
| "loss": 33.0674765625, |
| "step": 127000 |
| }, |
| { |
| "epoch": 64.16042110540168, |
| "grad_norm": 12.050172805786133, |
| "learning_rate": 0.0003, |
| "loss": 33.0123046875, |
| "step": 128000 |
| }, |
| { |
| "epoch": 64.16042110540168, |
| "eval_accuracy": 0.7628614283635131, |
| "eval_loss": 8.15625, |
| "eval_runtime": 309.2479, |
| "eval_samples_per_second": 1467.719, |
| "eval_steps_per_second": 2.868, |
| "step": 128000 |
| }, |
| { |
| "epoch": 64.66173705978193, |
| "grad_norm": 8.326544761657715, |
| "learning_rate": 0.0003, |
| "loss": 32.89726171875, |
| "step": 129000 |
| }, |
| { |
| "epoch": 65.16292768517359, |
| "grad_norm": 9.267374038696289, |
| "learning_rate": 0.0003, |
| "loss": 32.78715625, |
| "step": 130000 |
| }, |
| { |
| "epoch": 65.16292768517359, |
| "eval_accuracy": 0.7632605632607093, |
| "eval_loss": 8.1484375, |
| "eval_runtime": 313.0209, |
| "eval_samples_per_second": 1450.028, |
| "eval_steps_per_second": 2.834, |
| "step": 130000 |
| }, |
| { |
| "epoch": 65.66424363955383, |
| "grad_norm": 9.583052635192871, |
| "learning_rate": 0.0003, |
| "loss": 32.747501953125, |
| "step": 131000 |
| }, |
| { |
| "epoch": 66.16543426494549, |
| "grad_norm": 8.761311531066895, |
| "learning_rate": 0.0003, |
| "loss": 32.67369140625, |
| "step": 132000 |
| }, |
| { |
| "epoch": 66.16543426494549, |
| "eval_accuracy": 0.7639422135833412, |
| "eval_loss": 8.1015625, |
| "eval_runtime": 311.6656, |
| "eval_samples_per_second": 1456.333, |
| "eval_steps_per_second": 2.846, |
| "step": 132000 |
| }, |
| { |
| "epoch": 66.66675021932573, |
| "grad_norm": 8.83479118347168, |
| "learning_rate": 0.0003, |
| "loss": 32.617767578125, |
| "step": 133000 |
| }, |
| { |
| "epoch": 67.16794084471738, |
| "grad_norm": 8.598926544189453, |
| "learning_rate": 0.0003, |
| "loss": 32.5695625, |
| "step": 134000 |
| }, |
| { |
| "epoch": 67.16794084471738, |
| "eval_accuracy": 0.7644549296283725, |
| "eval_loss": 8.078125, |
| "eval_runtime": 308.4248, |
| "eval_samples_per_second": 1471.636, |
| "eval_steps_per_second": 2.876, |
| "step": 134000 |
| }, |
| { |
| "epoch": 67.66925679909762, |
| "grad_norm": 10.846793174743652, |
| "learning_rate": 0.0003, |
| "loss": 32.53196484375, |
| "step": 135000 |
| }, |
| { |
| "epoch": 68.17044742448928, |
| "grad_norm": 23.080833435058594, |
| "learning_rate": 0.0003, |
| "loss": 32.47344140625, |
| "step": 136000 |
| }, |
| { |
| "epoch": 68.17044742448928, |
| "eval_accuracy": 0.7638109525627774, |
| "eval_loss": 8.109375, |
| "eval_runtime": 312.4841, |
| "eval_samples_per_second": 1452.519, |
| "eval_steps_per_second": 2.839, |
| "step": 136000 |
| }, |
| { |
| "epoch": 68.67176337886953, |
| "grad_norm": 11.440296173095703, |
| "learning_rate": 0.0003, |
| "loss": 32.4546796875, |
| "step": 137000 |
| }, |
| { |
| "epoch": 69.17295400426119, |
| "grad_norm": 9.561952590942383, |
| "learning_rate": 0.0003, |
| "loss": 32.3915703125, |
| "step": 138000 |
| }, |
| { |
| "epoch": 69.17295400426119, |
| "eval_accuracy": 0.7654183586207545, |
| "eval_loss": 8.0234375, |
| "eval_runtime": 311.376, |
| "eval_samples_per_second": 1457.688, |
| "eval_steps_per_second": 2.849, |
| "step": 138000 |
| }, |
| { |
| "epoch": 69.67426995864143, |
| "grad_norm": 10.652801513671875, |
| "learning_rate": 0.0003, |
| "loss": 32.3813203125, |
| "step": 139000 |
| }, |
| { |
| "epoch": 70.17546058403309, |
| "grad_norm": 9.549755096435547, |
| "learning_rate": 0.0003, |
| "loss": 32.329857421875, |
| "step": 140000 |
| }, |
| { |
| "epoch": 70.17546058403309, |
| "eval_accuracy": 0.765731013146163, |
| "eval_loss": 8.015625, |
| "eval_runtime": 311.5237, |
| "eval_samples_per_second": 1456.997, |
| "eval_steps_per_second": 2.847, |
| "step": 140000 |
| }, |
| { |
| "epoch": 70.17546058403309, |
| "eval_accuracy": 0.7655576478890911, |
| "eval_loss": 8.03125, |
| "eval_runtime": 312.782, |
| "eval_samples_per_second": 1451.135, |
| "eval_steps_per_second": 2.836, |
| "step": 140000 |
| }, |
| { |
| "epoch": 70.67677653841334, |
| "grad_norm": 8.273364067077637, |
| "learning_rate": 0.0003, |
| "loss": 32.32880859375, |
| "step": 141000 |
| }, |
| { |
| "epoch": 71.177967163805, |
| "grad_norm": 11.310037612915039, |
| "learning_rate": 0.0003, |
| "loss": 32.2803671875, |
| "step": 142000 |
| }, |
| { |
| "epoch": 71.177967163805, |
| "eval_accuracy": 0.7654140428452689, |
| "eval_loss": 8.0234375, |
| "eval_runtime": 302.7715, |
| "eval_samples_per_second": 1499.114, |
| "eval_steps_per_second": 2.93, |
| "step": 142000 |
| }, |
| { |
| "epoch": 71.67928311818524, |
| "grad_norm": 9.46422004699707, |
| "learning_rate": 0.0003, |
| "loss": 32.241615234375, |
| "step": 143000 |
| }, |
| { |
| "epoch": 72.18047374357688, |
| "grad_norm": 9.287914276123047, |
| "learning_rate": 0.0003, |
| "loss": 32.22880078125, |
| "step": 144000 |
| }, |
| { |
| "epoch": 72.18047374357688, |
| "eval_accuracy": 0.7658155554395308, |
| "eval_loss": 8.015625, |
| "eval_runtime": 300.7976, |
| "eval_samples_per_second": 1508.951, |
| "eval_steps_per_second": 2.949, |
| "step": 144000 |
| }, |
| { |
| "epoch": 72.68178969795714, |
| "grad_norm": 9.183584213256836, |
| "learning_rate": 0.0003, |
| "loss": 32.233244140625, |
| "step": 145000 |
| }, |
| { |
| "epoch": 73.18298032334879, |
| "grad_norm": 9.008417129516602, |
| "learning_rate": 0.0003, |
| "loss": 32.181228515625, |
| "step": 146000 |
| }, |
| { |
| "epoch": 73.18298032334879, |
| "eval_accuracy": 0.76619202647217, |
| "eval_loss": 7.98828125, |
| "eval_runtime": 302.0251, |
| "eval_samples_per_second": 1502.819, |
| "eval_steps_per_second": 2.937, |
| "step": 146000 |
| }, |
| { |
| "epoch": 73.68429627772903, |
| "grad_norm": 8.19743537902832, |
| "learning_rate": 0.0003, |
| "loss": 32.162357421875, |
| "step": 147000 |
| }, |
| { |
| "epoch": 74.18548690312069, |
| "grad_norm": 8.455910682678223, |
| "learning_rate": 0.0003, |
| "loss": 32.091048828125, |
| "step": 148000 |
| }, |
| { |
| "epoch": 74.18548690312069, |
| "eval_accuracy": 0.7663843416476586, |
| "eval_loss": 7.97265625, |
| "eval_runtime": 301.7215, |
| "eval_samples_per_second": 1504.331, |
| "eval_steps_per_second": 2.94, |
| "step": 148000 |
| }, |
| { |
| "epoch": 74.68680285750094, |
| "grad_norm": 8.09157943725586, |
| "learning_rate": 0.0003, |
| "loss": 32.071322265625, |
| "step": 149000 |
| }, |
| { |
| "epoch": 75.1879934828926, |
| "grad_norm": 12.704072952270508, |
| "learning_rate": 0.0003, |
| "loss": 32.044611328125, |
| "step": 150000 |
| }, |
| { |
| "epoch": 75.1879934828926, |
| "eval_accuracy": 0.7670952482486783, |
| "eval_loss": 7.96484375, |
| "eval_runtime": 301.7456, |
| "eval_samples_per_second": 1504.211, |
| "eval_steps_per_second": 2.94, |
| "step": 150000 |
| }, |
| { |
| "epoch": 75.1879934828926, |
| "step": 150000, |
| "total_flos": 3.23779983669461e+19, |
| "train_loss": 2.1457560286458333, |
| "train_runtime": 27498.2172, |
| "train_samples_per_second": 11171.633, |
| "train_steps_per_second": 5.455 |
| } |
| ], |
| "logging_steps": 1000, |
| "max_steps": 150000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 76, |
| "save_steps": 10000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.23779983669461e+19, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
