logistic fitting for shot thresholds

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• The logistic fit for threshold behavior is a phenomenological surrogate for interpretability, not a mechanistic derivationclaim0.822
  Authors' explicit epistemic limitation on the threshold model
• Logistic surrogate fittingmethod0.812
  Fitting a logistic function to success probability as a function of S or shot count to estimate midpoints and widths.
• The logistic surrogate is a phenomenological fit, not a mechanistic derivationclaim0.766
  Caveat about the threshold analysis
• Logistic regression correctness probemethod0.755
  Logistic regression trained on GSM8k training set to predict answer correctness from projection features along reflection direction
• few-shot threshold (k50/θ50)concept0.751
  Shot count needed to reach 50% accuracy; reflects when anchoring strength crosses critical value.
• logistic surrogate modelmethod0.750
  Sigmoid fit linking S to success probability.
• S = ρd - dr - log k predicts shot midpoints across different bases, tasks, and modelsclaim0.736
  Predictive practical utility claim.
• Logistic Regression Probemethod0.732
  Standard linear probing technique; compared to mass-mean probing for classification accuracy and causal implication