INDEX

Explanations

mentions of evidence in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

evidence

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kowalski

-0.57

 pasaran

-0.53

 princely

-0.52

mobileqq

-0.50

 useRouter

-0.49

monthly

-0.49

rrggbb

-0.48

 urusan

-0.48

 kaybet

-0.48

 alugar

-0.48

POSITIVE LOGITS

 evidence

1.13

Evidence

1.06

 Evidence

1.05

evidence

1.02

 EVIDENCE

0.90

 evidencia

0.82

 Evid

0.78

 eviden

0.75

 evid

0.70

 evidences

0.68

Activations Density 0.039%

mentions of evidence in various contexts

evidence

No Comments

No Known Activations