INDEX

Explanations

references to political affiliations and actions

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

political parties and candidates

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

 kasarigan

-0.61

TypedDataSet

-0.57

tagPool

-0.44

 gynhyrchwyd

-0.43

DoubleQuotes

-0.43

postIndex

-0.41

richTextPanel

-0.41

RegistryLite

-0.40

 primit

-0.38

Famille

-0.36

POSITIVE LOGITS

 defaultstate

0.44

明

0.43

SUPPORT

0.41

support

0.40

 supporter

0.39

 Unterstützung

0.39

 támogat

0.39

 magát

0.38

supporting

0.38

legate

0.38

Activations Density 0.022%

references to political affiliations and actions

political parties and candidates

No Comments

No Known Activations