finding
active
finding:b10-phase-width-k-1-21-0-18B10 phase width Δk = 1.21 ± 0.18
Transition width (k90 – k10) for B10.
Source paper
extracted_from(2025) · Edward Yi Chang · Kaya, Zeyneb N. · Ethan Chang
Neighborhood — ranked by edge-count
Claims (1)
claim
- Transition widths ∆k increase with mismatch D(P0 ∥ PT), evidenced by wider widths from B10 to B9supportsInterpretive claim linking phase width in E2 to mismatch term in UCCT
Communities (3)
community
- Few-shot anchoring & latent structuremembers_ofHow minimal examples disambiguate and recruit latent arithmetic/reasoning interpretations in LLMs
- Empirical characterization of k50 midpoints and transition widths across transformer models, tracking how pretraining density ρd/dr predicts in-context learning thresholds.
- Few-shot arithmetic learning thresholdsmembers_ofk50 and phase width metrics for multi-base addition across bases 8, 9, and 10
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Widest transition in E2; consistent with lower prior density requiring more shots for reliable threshold crossing
- 10–90% width of the logistic transition; wider when anchoring is weaker.
- Lowest threshold condition in E2; near-zero/one-shot threshold consistent with high pretraining density
- Accuracy at k=16 shots for B10.
- Shot midpoint ordering k50(B10) < k50(B8) ≈ k50(B9) and transition widths correlate with mismatch D(P0∥PT)hypothesis0.721Testable prediction for Experiment 2
- Hypothesis: Shot midpoint ordering k50(B10) < k50(B8) ≈ k50(B9) follows pretraining exposure densityhypothesis0.721E2 prediction that bases with higher pretraining exposure require fewer shots to cross threshold
- Performance with a different backbone network.
- Monotone ordering consistent with k50 ∝ dr/ρd.