INDEX

Explanations

references to media sources and news outlets

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

news sources and data

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

 Schie

-0.32

 Wied

-0.31

nande

-0.30

 réputation

-0.30

 Eind

-0.30

 vogli

-0.30

 Bele

-0.30

 SOMETHING

-0.29

kí

-0.29

Wid

-0.29

POSITIVE LOGITS

 ویکی‌پدی

0.76

adpleegd

0.70

istoitu

0.68

 препратки

0.66

IsMutable

0.63

isContained

0.63

 يتيمه

0.62

ScopeManager

0.61

Rhestr

0.59

windowFixed

0.58

Activations Density 0.013%

references to media sources and news outlets

news sources and data

No Comments

No Known Activations