INDEX

Explanations

references to political affiliations and actions within a legislative context

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

preceding "who"

np_acts-logits-general · gemini-2.0-flash

who support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

脚注の使い方

-0.73

windowFixed

-0.65

 AssemblyCulture

-0.61

 gynnwys

-0.60

 Paglinawan

-0.60

ragalactic

-0.58

 ModelExpression

-0.56

thâu

-0.55

pexpr

-0.54

 autorytatywna

-0.54

POSITIVE LOGITS

 supports

1.62

 supported

1.59

 support

1.56

supported

1.39

 Supports

1.38

 Supported

1.37

 supporting

1.35

support

1.34

 SUPPORT

1.34

 Support

1.33

Activations Density 0.495%

references to political affiliations and actions within a legislative context

preceding "who"

who support

No Comments

No Known Activations

references to political affiliations and actions within a legislative context

preceding "who"

who support

No Comments

No Known Activations