INDEX

Explanations

locations or references to being in specific places and areas

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

in the context of visiting

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

 casó

-0.36

 savons

-0.35

rū

-0.35

nictví

-0.34

 ddelweddau

-0.33

 terlaris

-0.33

garten

-0.31

SAD

-0.31

 PROBLEMS

-0.31

 adap

-0.30

POSITIVE LOGITS

ⓧ

0.51

IndentedString

0.47

NameInMap

0.46

 &___

0.45

ADELPHIA

0.45

InputBorder

0.44

 disambiguazione

0.44

曖昧さ回避

0.44

جمعیت

0.43

WebElementEntity

0.43

Activations Density 0.005%

locations or references to being in specific places and areas

in the context of visiting

No Comments

No Known Activations