INDEX

Explanations

references to females or female-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eer

-0.22

raman

-0.15

male

-0.15

eti

-0.15

ally

-0.15

 gentlemen

-0.15

ality

-0.15

osu

-0.15

flix

-0.15

 males

-0.15

POSITIVE LOGITS

hood

0.33

ie

0.30

/man

0.26

 scout

0.25

friends

0.25

/w

0.25

 Scout

0.24

ies

0.24

 Scouts

0.24

friend

0.23

Activations Density 0.044%

references to females or female-related terms

No Comments

No Known Activations

references to females or female-related terms

No Comments

No Known Activations