INDEX

Explanations

pronouns, especially personal or demonstrative ones

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

non-english words

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

IsContent

-0.65

pantalones

-0.61

Hochspringen

-0.57

adaptiveStyles

-0.57

+#+

-0.54

 Fré

-0.53

 témoig

-0.52

DataSnapshot

-0.52

 Schott

-0.51

⪜

-0.50

POSITIVE LOGITS

 ويكيپيديا

0.36

featureID

0.35

 Familienname

0.31

 Gerade

0.30

interopRequire

0.30

 obcí

0.29

 puentes

0.29

Zgod

0.29

 články

0.29

Dabei

0.29

Activations Density 0.001%

pronouns, especially personal or demonstrative ones

non-english words

No Comments

No Known Activations