INDEX

Explanations

names and titles of individuals and entities associated with notable actions or properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

 atlik

-0.37

 performed

-0.31

 καλ

-0.31

 regulares

-0.31

 gră

-0.30

hiv

-0.30

 regolare

-0.29

 sessions

-0.29

 dapur

-0.29

 flow

-0.28

POSITIVE LOGITS

 entrepreneur

1.35

 businessman

1.30

 entrepreneurs

1.23

 businessmen

1.21

 billionaire

1.20

 entrepreneurial

1.18

businessman

1.13

 billionaires

1.12

entrepreneur

1.11

 tycoon

1.10

Activations Density 0.649%

names and titles of individuals and entities associated with notable actions or properties

No Comments

No Known Activations