INDEX

Explanations

instances of the word "pair" and its variations related to collaboration or teamwork

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ä¸īçº§

-0.17

edir

-0.17

rud

-0.17

.erb

-0.16

eti

-0.16

Ð½Ð°ÑĢ

-0.15

/ion

-0.15

pany

-0.15

Leaks

-0.15

Ã¡ÅĻ

-0.15

POSITIVE LOGITS

ings

0.41

INGS

0.27

/group

0.26

wise

0.24

-wise

0.22

ä¼į

0.21

tures

0.21

/groups

0.21

wis

0.20

/single

0.18

Activations Density 0.026%

instances of the word "pair" and its variations related to collaboration or teamwork

No Comments

No Known Activations