INDEX

Explanations

terms associated with deception or falsehoods

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

things that are not true

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

)++;

-0.40

 smart

-0.32

Conf

-0.32

admir

-0.31

áld

-0.29

APPRO

-0.29

Smart

-0.29

inci

-0.28

Tapped

-0.28

bold

-0.28

POSITIVE LOGITS

WebVitals

1.00

InjectAttribute

0.77

 HasFactory

0.74

Diwedd

0.73

 CreateTagHelper

0.72

出版年

0.68

migrationBuilder

0.68

oa̍t

0.66

verwijspagina

0.66

inaison

0.62

Activations Density 0.058%

terms associated with deception or falsehoods

things that are not true

No Comments

No Known Activations