INDEX

Explanations

patterns related to the word "most"

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_9/width_16k/average_l0_100

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

programme

-0.58

 Blanchard

-0.54

 Martel

-0.54

 Fable

-0.53

 Ename

-0.52

cheque

-0.51

Programme

-0.51

 Symbol

-0.51

phide

-0.51

 Papier

-0.50

POSITIVE LOGITS

 most

1.48

most

1.36

Most

1.35

 Most

1.31

 MOST

1.26

MOST

1.16

 moſt

1.02

 moft

1.01

 plupart

0.92

 meisten

0.89

Activations Density 0.180%

patterns related to the word "most"

No Known Activations