INDEX

Explanations

references to visitors or guests in various contexts

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

visitor or visitors

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

 Leib

-0.47

 Holtz

-0.46

 MacDonald

-0.45

the

-0.45

-0.43

 hard

-0.42

<h1>

-0.42

 Schne

-0.42

McN

-0.42

 tough

-0.41

POSITIVE LOGITS

 Visitor

1.15

 visitor

1.12

visitors

1.10

 Visitors

1.09

visitor

1.05

Visitor

1.04

 visitors

1.03

Visitors

1.00

 visiteurs

0.94

 bezoekers

0.84

Activations Density 0.006%

references to visitors or guests in various contexts

visitor or visitors

No Comments

No Known Activations