INDEX

Explanations

references to deception and disguise in interpersonal contexts

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

action, authority, message, known, headers

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

gius

-0.36

 soeur

-0.36

 málaga

-0.36

 للاسماء

-0.33

PrivateRoute

-0.33

ắp

-0.32

Portale

-0.32

期刊论文

-0.31

戦

-0.31

capulco

-0.31

POSITIVE LOGITS

 linkovi

0.57

OGND

0.55

BagConstraints

0.52

ViewImports

0.49

 defStyle

0.49

 دیکھیے

0.44

msgTypes

0.44

قایناقلار

0.43

onedDateTime

0.43

 Diplomacy

0.42

Activations Density 0.103%

references to deception and disguise in interpersonal contexts

action, authority, message, known, headers

No Comments

No Known Activations