INDEX

Explanations

proper nouns, specifically names of people or entities

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

initials and acronyms

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

 quadril

-0.67

binar

-0.67

⭗

-0.63

஡

-0.63

Попис

-0.63

 ſehr

-0.63

ashian

-0.63

<unused23>

-0.62

<unused20>

-0.62

<unused11>

-0.62

POSITIVE LOGITS

 AssemblyCompany

0.41

 MainAxisSize

0.36

 invokingState

0.32

 amitié

0.31

akterysty

0.31

clearTimeout

0.28

 espoir

0.28

 investissement

0.28

 mépris

0.28

 financement

0.28

Activations Density 0.007%

proper nouns, specifically names of people or entities

initials and acronyms

No Comments

No Known Activations