| { | |
| "best_metric": 1.6984472274780273, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved_llamaprosocial_dialog/checkpoint-341", | |
| "epoch": 2.9531799729364003, | |
| "global_step": 341, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.9915313720703125, | |
| "eval_runtime": 22.7483, | |
| "eval_samples_per_second": 87.919, | |
| "eval_steps_per_second": 1.407, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.000292814371257485, | |
| "loss": 2.0439, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.8029985427856445, | |
| "eval_runtime": 22.7395, | |
| "eval_samples_per_second": 87.953, | |
| "eval_steps_per_second": 1.407, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.7677525281906128, | |
| "eval_runtime": 22.7246, | |
| "eval_samples_per_second": 88.01, | |
| "eval_steps_per_second": 1.408, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00027485029940119756, | |
| "loss": 1.7686, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 1.7513495683670044, | |
| "eval_runtime": 22.772, | |
| "eval_samples_per_second": 87.827, | |
| "eval_steps_per_second": 1.405, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 1.7416645288467407, | |
| "eval_runtime": 22.7234, | |
| "eval_samples_per_second": 88.015, | |
| "eval_steps_per_second": 1.408, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00025688622754491017, | |
| "loss": 1.737, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 1.734699010848999, | |
| "eval_runtime": 22.7372, | |
| "eval_samples_per_second": 87.962, | |
| "eval_steps_per_second": 1.407, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.72977614402771, | |
| "eval_runtime": 22.74, | |
| "eval_samples_per_second": 87.951, | |
| "eval_steps_per_second": 1.407, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00023892215568862272, | |
| "loss": 1.7289, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 1.725810170173645, | |
| "eval_runtime": 22.7407, | |
| "eval_samples_per_second": 87.948, | |
| "eval_steps_per_second": 1.407, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 1.7222310304641724, | |
| "eval_runtime": 22.733, | |
| "eval_samples_per_second": 87.978, | |
| "eval_steps_per_second": 1.408, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00022095808383233533, | |
| "loss": 1.7172, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.7189042568206787, | |
| "eval_runtime": 22.7271, | |
| "eval_samples_per_second": 88.001, | |
| "eval_steps_per_second": 1.408, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00020299401197604788, | |
| "loss": 1.7127, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 1.7177015542984009, | |
| "eval_runtime": 22.7303, | |
| "eval_samples_per_second": 87.988, | |
| "eval_steps_per_second": 1.408, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 1.7150254249572754, | |
| "eval_runtime": 22.7281, | |
| "eval_samples_per_second": 87.997, | |
| "eval_steps_per_second": 1.408, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00018502994011976046, | |
| "loss": 1.7017, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.7136248350143433, | |
| "eval_runtime": 22.732, | |
| "eval_samples_per_second": 87.982, | |
| "eval_steps_per_second": 1.408, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 1.7119090557098389, | |
| "eval_runtime": 22.7069, | |
| "eval_samples_per_second": 88.079, | |
| "eval_steps_per_second": 1.409, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00016706586826347302, | |
| "loss": 1.7003, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 1.7102028131484985, | |
| "eval_runtime": 22.9016, | |
| "eval_samples_per_second": 87.33, | |
| "eval_steps_per_second": 1.397, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 1.708575963973999, | |
| "eval_runtime": 22.7175, | |
| "eval_samples_per_second": 88.038, | |
| "eval_steps_per_second": 1.409, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0001491017964071856, | |
| "loss": 1.7025, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 1.7076679468154907, | |
| "eval_runtime": 22.792, | |
| "eval_samples_per_second": 87.75, | |
| "eval_steps_per_second": 1.404, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_loss": 1.7062067985534668, | |
| "eval_runtime": 22.7054, | |
| "eval_samples_per_second": 88.085, | |
| "eval_steps_per_second": 1.409, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0001311377245508982, | |
| "loss": 1.6976, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_loss": 1.7049460411071777, | |
| "eval_runtime": 22.765, | |
| "eval_samples_per_second": 87.854, | |
| "eval_steps_per_second": 1.406, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.00011317365269461076, | |
| "loss": 1.6955, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_loss": 1.704172134399414, | |
| "eval_runtime": 22.8464, | |
| "eval_samples_per_second": 87.541, | |
| "eval_steps_per_second": 1.401, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.7033213376998901, | |
| "eval_runtime": 22.7378, | |
| "eval_samples_per_second": 87.959, | |
| "eval_steps_per_second": 1.407, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.520958083832335e-05, | |
| "loss": 1.6893, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_loss": 1.7023844718933105, | |
| "eval_runtime": 22.7421, | |
| "eval_samples_per_second": 87.942, | |
| "eval_steps_per_second": 1.407, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.7013720273971558, | |
| "eval_runtime": 22.7212, | |
| "eval_samples_per_second": 88.023, | |
| "eval_steps_per_second": 1.408, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 7.724550898203592e-05, | |
| "loss": 1.6853, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_loss": 1.7013109922409058, | |
| "eval_runtime": 22.7306, | |
| "eval_samples_per_second": 87.987, | |
| "eval_steps_per_second": 1.408, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_loss": 1.7006632089614868, | |
| "eval_runtime": 22.748, | |
| "eval_samples_per_second": 87.92, | |
| "eval_steps_per_second": 1.407, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 5.92814371257485e-05, | |
| "loss": 1.6877, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.7002646923065186, | |
| "eval_runtime": 22.7267, | |
| "eval_samples_per_second": 88.002, | |
| "eval_steps_per_second": 1.408, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": 1.6997302770614624, | |
| "eval_runtime": 22.7256, | |
| "eval_samples_per_second": 88.006, | |
| "eval_steps_per_second": 1.408, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.131736526946108e-05, | |
| "loss": 1.6834, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 1.6993123292922974, | |
| "eval_runtime": 22.7463, | |
| "eval_samples_per_second": 87.926, | |
| "eval_steps_per_second": 1.407, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 1.6989519596099854, | |
| "eval_runtime": 22.7888, | |
| "eval_samples_per_second": 87.762, | |
| "eval_steps_per_second": 1.404, | |
| "step": 319 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 2.3353293413173652e-05, | |
| "loss": 1.6854, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_loss": 1.698702335357666, | |
| "eval_runtime": 22.724, | |
| "eval_samples_per_second": 88.013, | |
| "eval_steps_per_second": 1.408, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 5.389221556886227e-06, | |
| "loss": 1.6858, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_loss": 1.6984472274780273, | |
| "eval_runtime": 22.7255, | |
| "eval_samples_per_second": 88.007, | |
| "eval_steps_per_second": 1.408, | |
| "step": 341 | |
| } | |
| ], | |
| "max_steps": 345, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.604629423378399e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |