INDEX

Explanations

instances of hearsay or reported speech

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

heard or said

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

 itſelf

-0.62

 surla

-0.55

majánló

-0.54

stylers

-0.53

sizeCache

-0.53

 AssemblyCompany

-0.52

InputBorder

-0.52

BagConstraints

-0.52

 Gemeinsame

-0.49

aarrggbb

-0.49

POSITIVE LOGITS

.*")]

0.46

%}

0.38

 незавершена

0.36

[]>(

0.35

تقاوى

0.35

 numerus

0.34

听说

0.33

 tương

0.33

 kyse

0.31

Activations Density 0.033%

instances of hearsay or reported speech

heard or said

No Comments

No Known Activations