INDEX

Explanations

phrases that indicate the use of language as a means of persuasion or manipulation

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

exaggeration and negativity

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

 CreateTagHelper

-0.62

帖最后由

-0.58

المشاركات

-0.56

aarrggbb

-0.49

verwijspagina

-0.47

 MyBatis

-0.47

 Majefty

-0.47



-0.46

 Monfieur

-0.45

 ostavi

-0.45

POSITIVE LOGITS

bed

0.40

̈́

0.40

adins

0.37

 Tole

0.36

โท

0.36

 קל

0.36

 exaggeration

0.35

 exaggerating

0.35

 wings

0.35

 Vikipedi

0.35

Activations Density 0.058%

phrases that indicate the use of language as a means of persuasion or manipulation

exaggeration and negativity

No Comments

No Known Activations