finding
active
finding:llama-sample-efficiency-multiplier-at-1-token-spacing-22-1-at-s-5-7-3-at-s-30-1-6-at-s-100Llama sample-efficiency multiplier at 1-token spacing: 22.1× at S=5, 7.3× at S=30, 1.6× at S=100.
Efficiency gain.
Source paper
extracted_from(2026) · Eric Bigelow · Amir Zur · Satchel Grant · Tal Haklay +6
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Discussion.
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.
- Llama sample-efficiency multiplier at 4-token spacing: 5.0× at S=5, 3.3× at S=30, 1.0× at S=100.finding0.971Efficiency gain.
- Noise scaling.
- Validates Assumption D.6 that base models assign higher typicality ratings to representative (diverse) sequences
- Headline.
- Supporting finding showing ESR is driven by both higher multi-attempt rates and comparable improvement rates
- Quantitative argument for the richness of quasi-psychological connections enabled by attention streams
- Empirical observation establishing that Llama's behavior for days-of-week tasks has circular structure.
- Multi-attempt improvement rate peaks at 83% around -1.0σ below threshold in Llama-3.3-70Bfinding0.769Shows slightly weaker steering allows more successful corrections, characterizing optimal ESR conditions