INDEX

Explanations

references to historical figures and philosophical concepts

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

historical figures and events

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

ThroughAttribute

-0.67

 informée

-0.63

iglesia

-0.62

 كورة

-0.59

 ***!

-0.57

oa̍t

-0.57

iſten

-0.55

 виправивши

-0.54

enablog

-0.53

XtraBars

-0.53

POSITIVE LOGITS

 historical

0.50

Historical

0.40

 Historical

0.40

 historic

0.39

 history

0.37

 antique

0.36

Historic

0.35

 historian

0.35

 legado

0.33

Pos

0.33

Activations Density 0.058%

references to historical figures and philosophical concepts

historical figures and events

No Comments

No Known Activations