INDEX

Explanations

references to "ladies" or female-oriented terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ters

-0.17

itional

-0.16

ummings

-0.16

à¸±à¸Ļà¸Ļ

-0.15

leo

-0.15

bie

-0.14

hea

-0.14

Girl

-0.14

nist

-0.14

POSITIVE LOGITS

bug

0.24

/man

0.23

bugs

0.21

 Gaga

0.19

folk

0.19

oir

0.18

/g

0.18

bower

0.17

ships

0.17

gent

0.16

Activations Density 0.019%

references to "ladies" or female-oriented terms

No Comments

No Known Activations