/Users/simon-pierreboucher/Desktop/Recherche-UQO/ia/prisme-slm/training/train_controller.py:131: UserWarning: Converting a tensor with requires_grad=True to a scalar may lead to unexpected behavior.
Consider using tensor.detach() first. (Triggered internally at /Users/runner/work/pytorch/pytorch/torch/csrc/autograd/generated/python_variable_methods.cpp:821.)
  run.log(step=step, loss=float(loss), gain=float(L["gain"]), harm=float(L["harm"]), nll=float(L["nll"]))
step 0 loss 0.45115 gain 0.02663 harm 0.8158
step 20 loss 0.13172 gain 0.00543 harm 0.5627
  calib: {'huber': 0.0175, 'sign_acc': 0.5467, 'harm_brier': 0.1334, 'harm_ece': 0.0955, 'n_labels': 1156, 'oracle_opportunity_per_state': 0.2508, 'regret_per_state': 0.1381, 'captured_fraction': 0.4494, 'top1_match': 0.0, 'n_audit_states': 124}
step 40 loss 0.07597 gain 0.00382 harm 0.4947
step 59 loss 0.05327 gain 0.00333 harm 0.4847
  calib: {'huber': 0.0162, 'sign_acc': 0.7344, 'harm_brier': 0.1264, 'harm_ece': 0.0872, 'n_labels': 1156, 'oracle_opportunity_per_state': 0.2508, 'regret_per_state': 0.1129, 'captured_fraction': 0.5501, 'top1_match': 0.0, 'n_audit_states': 124}
restored controller from step 60 (calibration score -0.11286)
sweep {'lambda': 0.0, 'kappa': 0.0, 'loss': 2.059080786738835, 'edits_per_token': 2.5673758865248226, 'extra_ops_ratio': 1.4467924402115193}
sweep {'lambda': 0.3, 'kappa': 0.0, 'loss': 2.307212234388852, 'edits_per_token': 0.04964539007092199, 'extra_ops_ratio': 0.02571701665296622}
selected {'lambda': 0.3, 'kappa': 0.0, 'loss': 2.307212234388852, 'edits_per_token': 0.04964539007092199, 'extra_ops_ratio': 0.02571701665296622, 'loss_reduction': 0.04320390850094169, 'reduction_per_ops': 1.6799735787377388} 
run controller_mini_seed01_20261030005154