| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.846153846153847, | |
| "eval_steps": 500, | |
| "global_step": 480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20512820512820512, | |
| "grad_norm": 0.68310546875, | |
| "learning_rate": 0.00019978589232386035, | |
| "loss": 1.5545, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.41025641025641024, | |
| "grad_norm": 0.71240234375, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 1.4203, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.6708984375, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 1.4322, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8205128205128205, | |
| "grad_norm": 0.69482421875, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 1.377, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 0.0001946930129495106, | |
| "loss": 1.3246, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.85107421875, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.2329, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.435897435897436, | |
| "grad_norm": 0.7978515625, | |
| "learning_rate": 0.00018968727415326884, | |
| "loss": 1.1856, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.641025641025641, | |
| "grad_norm": 0.8544921875, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.2045, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 0.8330078125, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 1.1925, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 0.8515625, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 1.159, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.2564102564102564, | |
| "grad_norm": 0.87353515625, | |
| "learning_rate": 0.00017518398074789775, | |
| "loss": 0.9939, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 1.0583, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 0.00016593458151000688, | |
| "loss": 1.0258, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.871794871794872, | |
| "grad_norm": 0.9384765625, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 1.0612, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.08984375, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 1.0167, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.282051282051282, | |
| "grad_norm": 1.1005859375, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.9019, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.4871794871794872, | |
| "grad_norm": 1.154296875, | |
| "learning_rate": 0.00014422886902190014, | |
| "loss": 0.8952, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 1.06640625, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.9395, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.8974358974358974, | |
| "grad_norm": 1.1083984375, | |
| "learning_rate": 0.00013214394653031616, | |
| "loss": 0.9138, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "grad_norm": 1.240234375, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 0.8003, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.1552734375, | |
| "learning_rate": 0.00011950903220161285, | |
| "loss": 0.7733, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.512820512820513, | |
| "grad_norm": 1.0068359375, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 0.7342, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.717948717948718, | |
| "grad_norm": 1.271484375, | |
| "learning_rate": 0.00010654031292301432, | |
| "loss": 0.7787, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8549, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.128205128205128, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 9.345968707698569e-05, | |
| "loss": 0.748, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 0.6502, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.538461538461538, | |
| "grad_norm": 1.18359375, | |
| "learning_rate": 8.049096779838719e-05, | |
| "loss": 0.678, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.743589743589744, | |
| "grad_norm": 1.2099609375, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 0.6956, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.948717948717949, | |
| "grad_norm": 1.3115234375, | |
| "learning_rate": 6.785605346968386e-05, | |
| "loss": 0.7146, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 1.2412109375, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.6016, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.358974358974359, | |
| "grad_norm": 1.318359375, | |
| "learning_rate": 5.577113097809989e-05, | |
| "loss": 0.6325, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.564102564102564, | |
| "grad_norm": 1.3349609375, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.6448, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "grad_norm": 1.2177734375, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.607, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.9743589743589745, | |
| "grad_norm": 1.166015625, | |
| "learning_rate": 3.9123857099127936e-05, | |
| "loss": 0.6006, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.17948717948718, | |
| "grad_norm": 1.2724609375, | |
| "learning_rate": 3.406541848999312e-05, | |
| "loss": 0.5849, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.384615384615385, | |
| "grad_norm": 1.4111328125, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.6052, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.589743589743589, | |
| "grad_norm": 1.2001953125, | |
| "learning_rate": 2.4816019252102273e-05, | |
| "loss": 0.5591, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.794871794871795, | |
| "grad_norm": 1.3798828125, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 0.5398, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.2548828125, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.531, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.205128205128204, | |
| "grad_norm": 1.146484375, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 0.5311, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.41025641025641, | |
| "grad_norm": 1.3310546875, | |
| "learning_rate": 1.0312725846731175e-05, | |
| "loss": 0.5371, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 8.615384615384615, | |
| "grad_norm": 1.365234375, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.511, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.820512820512821, | |
| "grad_norm": 1.302734375, | |
| "learning_rate": 5.306987050489442e-06, | |
| "loss": 0.5505, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 9.025641025641026, | |
| "grad_norm": 1.3994140625, | |
| "learning_rate": 3.40741737109318e-06, | |
| "loss": 0.556, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 1.921471959676957e-06, | |
| "loss": 0.5174, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.435897435897436, | |
| "grad_norm": 1.337890625, | |
| "learning_rate": 8.555138626189618e-07, | |
| "loss": 0.5164, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 9.64102564102564, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 2.141076761396521e-07, | |
| "loss": 0.498, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 0.0, | |
| "loss": 0.5447, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "step": 480, | |
| "total_flos": 3.1522610479104e+16, | |
| "train_loss": 0.833039856950442, | |
| "train_runtime": 560.469, | |
| "train_samples_per_second": 3.479, | |
| "train_steps_per_second": 0.856 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3.1522610479104e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |