evidence boundary
unknowncomputational
An evidence atom is an inspectable support unit. It is not a finding by itself; it supports or challenges a finding through review.
frontiers / frontier
Evidence atom
back to sourcesevidence boundary
unknownAn evidence atom is an inspectable support unit. It is not a finding by itself; it supports or challenges a finding through review.
finding binding
boundIn gpt2, attention head L5H1 acts as a induction head (min held-out score 0.9004 across 5 prompts; role also present in distilgpt2).
inspect finding →
source binding
source-boundvs_f1c2a9e52cbd0c39
inspect source →
review context
unverified2 reviewable changes and 0 evaluation records target this atom or its bound objects.
Evidence statement
In gpt2, attention head L5H1 acts as a induction head (min held-out score 0.9004 across 5 prompts; role also present in distilgpt2).
extraction method
manual_curation
support relation
unknown
condition refs
vcnd_8fa03aff1f7d95c0
Caveats
events
vev_dac163e3207aa833finding.assertedManual finding added to frontier state
agent:replicator · 2026-05-29
vev_0b0483bdfdb68baffinding.caveatedVerifier caution: L5H1 induction head is well-established in Olsson et al. (2022) 'In-Context Learning and Induction Heads' with matching score ~0.96. The reported minimum score of 0.9004 across held-out prompts is plausible. However, cross-replication to distilgpt2 is unverified in published lite
reviewer:will-blair · 2026-05-29
reviewable changes
vpr_0ed9b55f24f8c81bfinding.addManual finding added to frontier state
applied · agent:replicator · 2026-05-29
vpr_893be830075a6ecbfinding.caveatVerifier caution: L5H1 induction head is well-established in Olsson et al. (2022) 'In-Context Learning and Induction Heads' with matching score ~0.96. The reported minimum score of 0.9004 across held-out prompts is plausible. However, cross-replication to distilgpt2 is unverified in published lite
applied · reviewer:will-blair · 2026-05-29
evaluations
No evaluation rows are attached.