INDEX

Explanations

phrases indicating references or citations

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

citing sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

RenderAtEndOf

-0.52

 Händen

-0.44

 solcher

-0.43

 estekak

-0.42

 inappropriés

-0.41

 éstos

-0.41

 utilizamos

-0.40

 ésta

-0.40

 könny

-0.40

ambilan

-0.40

POSITIVE LOGITS

 NASA

0.63

 Bible

0.58

 science

0.58

 Wikipedia

0.58

 scientific

0.57

 sources

0.56

 IPCC

0.56

 scriptures

0.55

 official

0.55

NASA

0.55

Activations Density 0.056%

phrases indicating references or citations

citing sources

No Comments

No Known Activations

phrases indicating references or citations

citing sources

No Comments

No Known Activations