INDEX

Explanations

references to user roles and permissions in a structured format

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

then [conditional code blocks]

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

 itſelf

-0.54

 تكبرها

-0.48

SPATH

-0.46

AFN

-0.43

큼

-0.43

 myſelf

-0.42

 slutet

-0.42

rotum

-0.42

 Kigali

-0.42

batore

-0.41

POSITIVE LOGITS

UnusedPrivate

0.53

expandindo

0.48

GrantedAuthority

0.47

timewa

0.41

ExecuteReader

0.40

CastException

0.40

__;

0.38

__()

0.38

 étrangères

0.36

merking

0.36

Activations Density 0.006%

references to user roles and permissions in a structured format

then [conditional code blocks]

No Comments

No Known Activations