INDEX

Explanations

names of influential figures and prominent titles associated with them

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 يتيمه

-0.62

 Induction

-0.47

 Patrick

-0.45

Induction

-0.42

patrick

-0.41

hand

-0.40

induction

-0.40

lection

-0.40

Patrick

-0.39

 patrick

-0.39

POSITIVE LOGITS

 greateſt

0.67

 juſ

0.65

 Majefty

0.60

 ſever

0.58

 ſtate

0.57

 houſe

0.57

 purpoſe

0.57

ſelf

0.56

 ſtre

0.56

contentLoaded

0.55

Activations Density 0.576%

names of influential figures and prominent titles associated with them

No Comments

No Known Activations