INDEX

Explanations

variations of the letter 'U' in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

agate

-0.16

alsy

-0.15

Pri

-0.15

.defaults

-0.15

utow

-0.14

leton

-0.14

ioc

-0.13

ekce

-0.13

anded

-0.13

activ

-0.13

POSITIVE LOGITS

gly

0.26

pto

0.24

 kunt

0.20

á»ĳng

0.20

prising

0.20

ppers

0.19

isce

0.19

gly

0.19

haul

0.19

pl

0.19

Activations Density 0.032%

variations of the letter 'U' in different contexts

No Comments

No Known Activations