INDEX

Explanations

mentions of historical events or periods

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

history followed by of

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

 gynhyrchwyd

-0.78

 Taktlose

-0.72

 Normdatei

-0.67

$_(

-0.66

 brome

-0.63

Rè

-0.63

RTEE

-0.63

 ddelwed

-0.62

PSE

-0.62

 queſta

-0.61

POSITIVE LOGITS

 history

0.81

 History

0.71

history

0.69

Története

0.68

History

0.65

 HISTORY

0.64

 historia

0.64

HISTORY

0.63

 História

0.61

 história

0.59

Activations Density 0.015%

mentions of historical events or periods

history followed by of

No Comments

No Known Activations