INDEX

Explanations

references to political affiliations and ideologies, particularly related to the Democratic Party

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 torebka

-0.36

 technology

-0.35

technology

-0.34

Technology

-0.33

 Technology

-0.33

ilarang

-0.33

 régler

-0.31

 bluzka

-0.31

 rodea

-0.31

 esclu

-0.30

POSITIVE LOGITS

<?

0.63

UserScript

0.63

دانشنامهٔ

0.57

featureID

0.56

LookAnd

0.54

 Савезне

0.54

 ویکی‌پدی

0.51

CompilerServices

0.51

 nakalista

0.50

ckså

0.47

Activations Density 0.087%

references to political affiliations and ideologies, particularly related to the Democratic Party

No Comments

No Known Activations