| { | |
| "best_metric": 1.6226980686187744, | |
| "best_model_checkpoint": "/data/user_data/gonilude/cpp_and_text_pythia_410m/checkpoint-200", | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 237, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0, | |
| "eval_accuracy": 0.3, | |
| "eval_loss": 2.5906739234924316, | |
| "eval_runtime": 0.536, | |
| "eval_samples_per_second": 130.596, | |
| "eval_steps_per_second": 16.791, | |
| "num_input_tokens_seen": 0, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.012658227848101266, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.2691, | |
| "num_input_tokens_seen": 8192, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06329113924050633, | |
| "grad_norm": Infinity, | |
| "learning_rate": 0.0, | |
| "loss": 2.9877, | |
| "num_input_tokens_seen": 40960, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12658227848101267, | |
| "grad_norm": Infinity, | |
| "learning_rate": 1e-05, | |
| "loss": 2.2275, | |
| "num_input_tokens_seen": 81920, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.189873417721519, | |
| "grad_norm": 274.08111572265625, | |
| "learning_rate": 1.9999058994907564e-05, | |
| "loss": 3.2099, | |
| "num_input_tokens_seen": 122880, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.25316455696202533, | |
| "grad_norm": 166.55935668945312, | |
| "learning_rate": 1.99661424082419e-05, | |
| "loss": 2.5884, | |
| "num_input_tokens_seen": 163840, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.31645569620253167, | |
| "grad_norm": 110.16832733154297, | |
| "learning_rate": 1.9886352515311134e-05, | |
| "loss": 2.4323, | |
| "num_input_tokens_seen": 204800, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.379746835443038, | |
| "grad_norm": 82.22343444824219, | |
| "learning_rate": 1.9760064588305347e-05, | |
| "loss": 2.3275, | |
| "num_input_tokens_seen": 245760, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4430379746835443, | |
| "grad_norm": 56.37868118286133, | |
| "learning_rate": 1.9587872591512583e-05, | |
| "loss": 1.8227, | |
| "num_input_tokens_seen": 286720, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5063291139240507, | |
| "grad_norm": 54.27399444580078, | |
| "learning_rate": 1.9370586387753532e-05, | |
| "loss": 1.998, | |
| "num_input_tokens_seen": 327680, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.569620253164557, | |
| "grad_norm": 44.79275894165039, | |
| "learning_rate": 1.9109227929390378e-05, | |
| "loss": 1.972, | |
| "num_input_tokens_seen": 368640, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6329113924050633, | |
| "grad_norm": 46.56148910522461, | |
| "learning_rate": 1.8805026451824547e-05, | |
| "loss": 1.9728, | |
| "num_input_tokens_seen": 409600, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6329113924050633, | |
| "eval_accuracy": 0.18571428571428572, | |
| "eval_loss": 2.392421245574951, | |
| "eval_runtime": 0.367, | |
| "eval_samples_per_second": 190.718, | |
| "eval_steps_per_second": 24.521, | |
| "num_input_tokens_seen": 409600, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6962025316455697, | |
| "grad_norm": 35.83555221557617, | |
| "learning_rate": 1.8459412692089497e-05, | |
| "loss": 2.219, | |
| "num_input_tokens_seen": 450560, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.759493670886076, | |
| "grad_norm": 33.77033996582031, | |
| "learning_rate": 1.8074012159730034e-05, | |
| "loss": 1.8699, | |
| "num_input_tokens_seen": 491520, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8227848101265823, | |
| "grad_norm": 51.212223052978516, | |
| "learning_rate": 1.765063749161688e-05, | |
| "loss": 1.9769, | |
| "num_input_tokens_seen": 532480, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8860759493670886, | |
| "grad_norm": 25.946151733398438, | |
| "learning_rate": 1.719127992665376e-05, | |
| "loss": 1.8649, | |
| "num_input_tokens_seen": 573440, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9493670886075949, | |
| "grad_norm": 34.178245544433594, | |
| "learning_rate": 1.6698099940473644e-05, | |
| "loss": 1.8005, | |
| "num_input_tokens_seen": 614400, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0126582278481013, | |
| "grad_norm": 26.01968765258789, | |
| "learning_rate": 1.6173417084171537e-05, | |
| "loss": 1.6863, | |
| "num_input_tokens_seen": 655360, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0759493670886076, | |
| "grad_norm": 17.587852478027344, | |
| "learning_rate": 1.5619699074864864e-05, | |
| "loss": 1.6306, | |
| "num_input_tokens_seen": 696320, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.139240506329114, | |
| "grad_norm": 25.147842407226562, | |
| "learning_rate": 1.50395501893913e-05, | |
| "loss": 1.5408, | |
| "num_input_tokens_seen": 737280, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2025316455696202, | |
| "grad_norm": 32.01517868041992, | |
| "learning_rate": 1.4435699015731449e-05, | |
| "loss": 1.9078, | |
| "num_input_tokens_seen": 778240, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.2658227848101267, | |
| "grad_norm": 22.242891311645508, | |
| "learning_rate": 1.3810985619764573e-05, | |
| "loss": 1.6016, | |
| "num_input_tokens_seen": 819200, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2658227848101267, | |
| "eval_accuracy": 0.14285714285714285, | |
| "eval_loss": 1.9393833875656128, | |
| "eval_runtime": 0.4244, | |
| "eval_samples_per_second": 164.94, | |
| "eval_steps_per_second": 21.207, | |
| "num_input_tokens_seen": 819200, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3291139240506329, | |
| "grad_norm": 36.45238494873047, | |
| "learning_rate": 1.3168348187715353e-05, | |
| "loss": 1.7416, | |
| "num_input_tokens_seen": 860160, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.3924050632911391, | |
| "grad_norm": 14.048996925354004, | |
| "learning_rate": 1.2510809207115666e-05, | |
| "loss": 1.6521, | |
| "num_input_tokens_seen": 901120, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4556962025316456, | |
| "grad_norm": 52.22154235839844, | |
| "learning_rate": 1.1841461251275868e-05, | |
| "loss": 1.7971, | |
| "num_input_tokens_seen": 942080, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.518987341772152, | |
| "grad_norm": 48.804534912109375, | |
| "learning_rate": 1.1163452434124773e-05, | |
| "loss": 1.7401, | |
| "num_input_tokens_seen": 983040, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5822784810126582, | |
| "grad_norm": 31.043750762939453, | |
| "learning_rate": 1.0479971603828001e-05, | |
| "loss": 1.6436, | |
| "num_input_tokens_seen": 1024000, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.6455696202531644, | |
| "grad_norm": 28.90459632873535, | |
| "learning_rate": 9.79423334482279e-06, | |
| "loss": 1.6048, | |
| "num_input_tokens_seen": 1064960, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.7088607594936709, | |
| "grad_norm": 33.064090728759766, | |
| "learning_rate": 9.109462858808586e-06, | |
| "loss": 1.6833, | |
| "num_input_tokens_seen": 1105920, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.7721518987341773, | |
| "grad_norm": 37.580997467041016, | |
| "learning_rate": 8.428880795801965e-06, | |
| "loss": 1.5788, | |
| "num_input_tokens_seen": 1146880, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.8354430379746836, | |
| "grad_norm": 36.66419219970703, | |
| "learning_rate": 7.75568810659924e-06, | |
| "loss": 1.7446, | |
| "num_input_tokens_seen": 1187840, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.8987341772151898, | |
| "grad_norm": 35.965087890625, | |
| "learning_rate": 7.093050987889547e-06, | |
| "loss": 1.7223, | |
| "num_input_tokens_seen": 1228800, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8987341772151898, | |
| "eval_accuracy": 0.14285714285714285, | |
| "eval_loss": 1.6300362348556519, | |
| "eval_runtime": 0.3719, | |
| "eval_samples_per_second": 188.224, | |
| "eval_steps_per_second": 24.2, | |
| "num_input_tokens_seen": 1228800, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.9620253164556962, | |
| "grad_norm": 36.96841812133789, | |
| "learning_rate": 6.444085990825338e-06, | |
| "loss": 1.628, | |
| "num_input_tokens_seen": 1269760, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.0253164556962027, | |
| "grad_norm": 20.582347869873047, | |
| "learning_rate": 5.811845363088477e-06, | |
| "loss": 1.6389, | |
| "num_input_tokens_seen": 1310720, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.088607594936709, | |
| "grad_norm": 40.09123229980469, | |
| "learning_rate": 5.199302693391958e-06, | |
| "loss": 1.5496, | |
| "num_input_tokens_seen": 1351680, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.151898734177215, | |
| "grad_norm": 21.198436737060547, | |
| "learning_rate": 4.609338925934743e-06, | |
| "loss": 1.5789, | |
| "num_input_tokens_seen": 1392640, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.2151898734177213, | |
| "grad_norm": 42.79294967651367, | |
| "learning_rate": 4.044728810587406e-06, | |
| "loss": 1.7237, | |
| "num_input_tokens_seen": 1433600, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.278481012658228, | |
| "grad_norm": 25.977487564086914, | |
| "learning_rate": 3.508127852536698e-06, | |
| "loss": 1.6412, | |
| "num_input_tokens_seen": 1474560, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.3417721518987342, | |
| "grad_norm": 12.27081298828125, | |
| "learning_rate": 3.0020598227682794e-06, | |
| "loss": 1.5877, | |
| "num_input_tokens_seen": 1515520, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.4050632911392404, | |
| "grad_norm": 29.93262481689453, | |
| "learning_rate": 2.5289048881289256e-06, | |
| "loss": 1.5655, | |
| "num_input_tokens_seen": 1556480, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.4683544303797467, | |
| "grad_norm": 11.503835678100586, | |
| "learning_rate": 2.090888416795582e-06, | |
| "loss": 1.6816, | |
| "num_input_tokens_seen": 1597440, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.5316455696202533, | |
| "grad_norm": 32.70045852661133, | |
| "learning_rate": 1.69007051180199e-06, | |
| "loss": 1.6031, | |
| "num_input_tokens_seen": 1638400, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5316455696202533, | |
| "eval_accuracy": 0.15714285714285714, | |
| "eval_loss": 1.6226980686187744, | |
| "eval_runtime": 0.3713, | |
| "eval_samples_per_second": 188.508, | |
| "eval_steps_per_second": 24.237, | |
| "num_input_tokens_seen": 1638400, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5949367088607596, | |
| "grad_norm": 25.854286193847656, | |
| "learning_rate": 1.3283363218493962e-06, | |
| "loss": 1.5904, | |
| "num_input_tokens_seen": 1679360, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.6582278481012658, | |
| "grad_norm": 19.382221221923828, | |
| "learning_rate": 1.0073871749720221e-06, | |
| "loss": 1.5788, | |
| "num_input_tokens_seen": 1720320, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.721518987341772, | |
| "grad_norm": 22.248796463012695, | |
| "learning_rate": 7.287325767579756e-07, | |
| "loss": 1.6045, | |
| "num_input_tokens_seen": 1761280, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.7848101265822782, | |
| "grad_norm": 23.392297744750977, | |
| "learning_rate": 4.936831107599749e-07, | |
| "loss": 1.5288, | |
| "num_input_tokens_seen": 1802240, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.848101265822785, | |
| "grad_norm": 22.441410064697266, | |
| "learning_rate": 3.033442744870685e-07, | |
| "loss": 1.5869, | |
| "num_input_tokens_seen": 1843200, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.911392405063291, | |
| "grad_norm": 31.935455322265625, | |
| "learning_rate": 1.5861127996827597e-07, | |
| "loss": 1.5976, | |
| "num_input_tokens_seen": 1884160, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.9746835443037973, | |
| "grad_norm": 13.112102508544922, | |
| "learning_rate": 6.016484334238515e-08, | |
| "loss": 1.613, | |
| "num_input_tokens_seen": 1925120, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "num_input_tokens_seen": 1941504, | |
| "step": 237, | |
| "total_flos": 3521692676653056.0, | |
| "train_loss": 1.8215426835329724, | |
| "train_runtime": 107.3388, | |
| "train_samples_per_second": 17.552, | |
| "train_steps_per_second": 2.208 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 237, | |
| "num_input_tokens_seen": 1941504, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3521692676653056.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |