INDEX

Explanations

references to debates or discussions involving opposing viewpoints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

esters

-0.16

igan

-0.15

ÐºÐ°Ð²

-0.15

ikut

-0.15

orian

-0.14

ties

-0.14

ucha

-0.14

 rack

-0.14

Ä°Åŀ

-0.14

indow

-0.14

POSITIVE LOGITS

ative

0.25

 against

0.22

against

0.21

inine

0.20

arg

0.20

uably

0.20

=args

0.20

UMENT

0.19

atively

0.19

(argument

0.19

Activations Density 0.023%

references to debates or discussions involving opposing viewpoints

No Comments

No Known Activations