INDEX

Explanations

relationships between different entities or concepts

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

legal and technical contexts

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

OpenHelper

-0.36

 ſta

-0.36

tall

-0.35

 Rohing

-0.35

GIT

-0.35

 perſon

-0.35

 houſe

-0.34

AnchorStyles

-0.34

 Majefty

-0.33

pena

-0.33

POSITIVE LOGITS

:✨

0.61

 AspNetCore

0.54

 للاسماء

0.53

 logra

0.50

CloseOperation

0.48

//

0.47

createCanvas

0.45

<?

0.44

atisfaction

0.44

InjectAttribute

0.43

Activations Density 0.096%

relationships between different entities or concepts

legal and technical contexts

No Comments

No Known Activations