| { | |
| "best_metric": 0.5201743841171265, | |
| "best_model_checkpoint": "deberta-classifier-feedback-1024-pseudo-final/checkpoint-170", | |
| "epoch": 1.9915254237288136, | |
| "global_step": 470, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9576271186440678e-05, | |
| "loss": 0.5814, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.5888153910636902, | |
| "eval_runtime": 18.2433, | |
| "eval_samples_per_second": 23.022, | |
| "eval_steps_per_second": 2.905, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.9152542372881357e-05, | |
| "loss": 0.5521, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.5736112594604492, | |
| "eval_runtime": 18.7271, | |
| "eval_samples_per_second": 22.427, | |
| "eval_steps_per_second": 2.83, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.8728813559322033e-05, | |
| "loss": 0.5685, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.5809019804000854, | |
| "eval_runtime": 17.2407, | |
| "eval_samples_per_second": 24.361, | |
| "eval_steps_per_second": 3.074, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.8305084745762713e-05, | |
| "loss": 0.6052, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.5701586008071899, | |
| "eval_runtime": 18.6909, | |
| "eval_samples_per_second": 22.471, | |
| "eval_steps_per_second": 2.836, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.788135593220339e-05, | |
| "loss": 0.5532, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.5571172833442688, | |
| "eval_runtime": 18.369, | |
| "eval_samples_per_second": 22.865, | |
| "eval_steps_per_second": 2.885, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.745762711864407e-05, | |
| "loss": 0.6177, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.5848062634468079, | |
| "eval_runtime": 18.5061, | |
| "eval_samples_per_second": 22.695, | |
| "eval_steps_per_second": 2.864, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.7033898305084745e-05, | |
| "loss": 0.6196, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.5464363098144531, | |
| "eval_runtime": 18.5102, | |
| "eval_samples_per_second": 22.69, | |
| "eval_steps_per_second": 2.863, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.6610169491525424e-05, | |
| "loss": 0.5772, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 0.5307226777076721, | |
| "eval_runtime": 18.3662, | |
| "eval_samples_per_second": 22.868, | |
| "eval_steps_per_second": 2.886, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.6186440677966104e-05, | |
| "loss": 0.5805, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.554991602897644, | |
| "eval_runtime": 17.9687, | |
| "eval_samples_per_second": 23.374, | |
| "eval_steps_per_second": 2.95, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.576271186440678e-05, | |
| "loss": 0.6453, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.5466664433479309, | |
| "eval_runtime": 18.0919, | |
| "eval_samples_per_second": 23.215, | |
| "eval_steps_per_second": 2.929, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.533898305084746e-05, | |
| "loss": 0.5756, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.5586597919464111, | |
| "eval_runtime": 18.6353, | |
| "eval_samples_per_second": 22.538, | |
| "eval_steps_per_second": 2.844, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.4915254237288137e-05, | |
| "loss": 0.5901, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.5481747388839722, | |
| "eval_runtime": 18.326, | |
| "eval_samples_per_second": 22.918, | |
| "eval_steps_per_second": 2.892, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.4491525423728813e-05, | |
| "loss": 0.568, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.5262647867202759, | |
| "eval_runtime": 18.2108, | |
| "eval_samples_per_second": 23.063, | |
| "eval_steps_per_second": 2.91, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.4067796610169493e-05, | |
| "loss": 0.5452, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": 0.5698090195655823, | |
| "eval_runtime": 18.2551, | |
| "eval_samples_per_second": 23.007, | |
| "eval_steps_per_second": 2.903, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.364406779661017e-05, | |
| "loss": 0.5949, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.5483840107917786, | |
| "eval_runtime": 18.0824, | |
| "eval_samples_per_second": 23.227, | |
| "eval_steps_per_second": 2.931, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.3220338983050848e-05, | |
| "loss": 0.5537, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.578332781791687, | |
| "eval_runtime": 18.2057, | |
| "eval_samples_per_second": 23.07, | |
| "eval_steps_per_second": 2.911, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.2796610169491526e-05, | |
| "loss": 0.5327, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 0.5201743841171265, | |
| "eval_runtime": 18.1319, | |
| "eval_samples_per_second": 23.164, | |
| "eval_steps_per_second": 2.923, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.2372881355932205e-05, | |
| "loss": 0.5449, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.5272189378738403, | |
| "eval_runtime": 18.125, | |
| "eval_samples_per_second": 23.172, | |
| "eval_steps_per_second": 2.924, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.1949152542372882e-05, | |
| "loss": 0.5345, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 0.5621271133422852, | |
| "eval_runtime": 18.0129, | |
| "eval_samples_per_second": 23.317, | |
| "eval_steps_per_second": 2.942, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1.1525423728813561e-05, | |
| "loss": 0.5837, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.55014967918396, | |
| "eval_runtime": 18.0302, | |
| "eval_samples_per_second": 23.294, | |
| "eval_steps_per_second": 2.94, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.1101694915254237e-05, | |
| "loss": 0.5969, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.5470077395439148, | |
| "eval_runtime": 17.9721, | |
| "eval_samples_per_second": 23.37, | |
| "eval_steps_per_second": 2.949, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.0677966101694917e-05, | |
| "loss": 0.5905, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.5924287438392639, | |
| "eval_runtime": 18.0444, | |
| "eval_samples_per_second": 23.276, | |
| "eval_steps_per_second": 2.937, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.0254237288135593e-05, | |
| "loss": 0.5481, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.5415045022964478, | |
| "eval_runtime": 17.9371, | |
| "eval_samples_per_second": 23.415, | |
| "eval_steps_per_second": 2.955, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.830508474576272e-06, | |
| "loss": 0.5035, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.5320823788642883, | |
| "eval_runtime": 17.9879, | |
| "eval_samples_per_second": 23.349, | |
| "eval_steps_per_second": 2.946, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.40677966101695e-06, | |
| "loss": 0.4508, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.5371343493461609, | |
| "eval_runtime": 17.4824, | |
| "eval_samples_per_second": 24.024, | |
| "eval_steps_per_second": 3.032, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.983050847457628e-06, | |
| "loss": 0.4227, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.5276100635528564, | |
| "eval_runtime": 18.1362, | |
| "eval_samples_per_second": 23.158, | |
| "eval_steps_per_second": 2.922, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.559322033898306e-06, | |
| "loss": 0.4423, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.532426118850708, | |
| "eval_runtime": 17.9907, | |
| "eval_samples_per_second": 23.345, | |
| "eval_steps_per_second": 2.946, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 8.135593220338983e-06, | |
| "loss": 0.432, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_loss": 0.5377896428108215, | |
| "eval_runtime": 17.4953, | |
| "eval_samples_per_second": 24.006, | |
| "eval_steps_per_second": 3.029, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.711864406779663e-06, | |
| "loss": 0.4317, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 0.5301514863967896, | |
| "eval_runtime": 18.2183, | |
| "eval_samples_per_second": 23.054, | |
| "eval_steps_per_second": 2.909, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 7.288135593220339e-06, | |
| "loss": 0.46, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.5301567316055298, | |
| "eval_runtime": 18.4315, | |
| "eval_samples_per_second": 22.787, | |
| "eval_steps_per_second": 2.876, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 6.864406779661017e-06, | |
| "loss": 0.435, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 0.5325623750686646, | |
| "eval_runtime": 17.6821, | |
| "eval_samples_per_second": 23.753, | |
| "eval_steps_per_second": 2.997, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 6.440677966101695e-06, | |
| "loss": 0.3813, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.5431253910064697, | |
| "eval_runtime": 18.5006, | |
| "eval_samples_per_second": 22.702, | |
| "eval_steps_per_second": 2.865, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 6.0169491525423725e-06, | |
| "loss": 0.4422, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 0.5322949290275574, | |
| "eval_runtime": 18.4759, | |
| "eval_samples_per_second": 22.732, | |
| "eval_steps_per_second": 2.869, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.593220338983051e-06, | |
| "loss": 0.4298, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": 0.5574814677238464, | |
| "eval_runtime": 17.6896, | |
| "eval_samples_per_second": 23.743, | |
| "eval_steps_per_second": 2.996, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 5.169491525423729e-06, | |
| "loss": 0.5068, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.5528993606567383, | |
| "eval_runtime": 18.3232, | |
| "eval_samples_per_second": 22.922, | |
| "eval_steps_per_second": 2.893, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 4.745762711864408e-06, | |
| "loss": 0.4619, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 0.5589260458946228, | |
| "eval_runtime": 18.3038, | |
| "eval_samples_per_second": 22.946, | |
| "eval_steps_per_second": 2.896, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 4.322033898305085e-06, | |
| "loss": 0.4852, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 0.5255549550056458, | |
| "eval_runtime": 18.4908, | |
| "eval_samples_per_second": 22.714, | |
| "eval_steps_per_second": 2.866, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.898305084745763e-06, | |
| "loss": 0.3888, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.5730893611907959, | |
| "eval_runtime": 18.3761, | |
| "eval_samples_per_second": 22.856, | |
| "eval_steps_per_second": 2.884, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.474576271186441e-06, | |
| "loss": 0.4319, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_loss": 0.5334990620613098, | |
| "eval_runtime": 18.4738, | |
| "eval_samples_per_second": 22.735, | |
| "eval_steps_per_second": 2.869, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.0508474576271192e-06, | |
| "loss": 0.4422, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_loss": 0.5419171452522278, | |
| "eval_runtime": 18.5281, | |
| "eval_samples_per_second": 22.668, | |
| "eval_steps_per_second": 2.861, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.627118644067797e-06, | |
| "loss": 0.4522, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_loss": 0.5547201037406921, | |
| "eval_runtime": 18.4837, | |
| "eval_samples_per_second": 22.723, | |
| "eval_steps_per_second": 2.867, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.203389830508475e-06, | |
| "loss": 0.4276, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 0.5263144373893738, | |
| "eval_runtime": 18.6045, | |
| "eval_samples_per_second": 22.575, | |
| "eval_steps_per_second": 2.849, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.7796610169491526e-06, | |
| "loss": 0.3988, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 0.5480612516403198, | |
| "eval_runtime": 18.5789, | |
| "eval_samples_per_second": 22.606, | |
| "eval_steps_per_second": 2.853, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.3559322033898307e-06, | |
| "loss": 0.4063, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.5404064655303955, | |
| "eval_runtime": 18.6623, | |
| "eval_samples_per_second": 22.505, | |
| "eval_steps_per_second": 2.84, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 9.322033898305086e-07, | |
| "loss": 0.4141, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_loss": 0.5292345881462097, | |
| "eval_runtime": 18.5061, | |
| "eval_samples_per_second": 22.695, | |
| "eval_steps_per_second": 2.864, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 5.084745762711865e-07, | |
| "loss": 0.4149, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_loss": 0.5240865349769592, | |
| "eval_runtime": 18.7251, | |
| "eval_samples_per_second": 22.43, | |
| "eval_steps_per_second": 2.83, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 8.474576271186442e-08, | |
| "loss": 0.4104, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_loss": 0.5263239145278931, | |
| "eval_runtime": 18.5702, | |
| "eval_samples_per_second": 22.617, | |
| "eval_steps_per_second": 2.854, | |
| "step": 470 | |
| } | |
| ], | |
| "max_steps": 472, | |
| "num_train_epochs": 2, | |
| "total_flos": 1.3918818284081568e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |