INDEX

Explanations

references to planets or worlds 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 @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

outs

-0.18

erson

-0.16

erton

-0.16

edom

-0.15

poses

-0.15

erman

-0.15

sy

-0.14

aday

-0.14

ikel

-0.14

afi

-0.14

POSITIVE LOGITS

-wide

0.33

wide

0.27

esimal

0.26

arium

0.21

éĻħ

0.21

weit

0.20

 wide

0.19

/star

0.18

/world

0.17

Wide

0.17

Activations Density 0.053%

references to planets or worlds in various contexts

No Comments

No Known Activations