INDEX

Explanations

references to value and worth in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

oug

-0.15

ahat

-0.15

evin

-0.15

cop

-0.15

liers

-0.13

QQ

-0.13

rq

-0.13

_campaign

-0.13

onn

-0.13

WI

-0.13

POSITIVE LOGITS

 worth

0.94

 Worth

0.79

worth

0.75

sworth

0.50

ä»·åĢ¼

0.48

 value

0.48

åĢ¼

0.46

value

0.43

 valued

0.43

-value

0.42

Activations Density 0.122%

references to value and worth in various contexts

No Known Activations