INDEX

Explanations

instances of calls to action, specifically urging the reader to click for more information or to take specific steps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

erring

-0.15

à¥Įà¤Ł

-0.15

hend

-0.14

autor

-0.14

URY

-0.14

gets

-0.14

uur

-0.14

uling

-0.14

_SEL

-0.14

ëĨĵ

-0.14

POSITIVE LOGITS

 here

0.29

 HERE

0.26

 Here

0.21

 ÙĩÙĨØ§

0.21

_here

0.20

 aqui

0.20

 below

0.19

 aquÃŃ

0.19

è¿ĻéĩĮ

0.18

Here

0.18

Activations Density 0.014%

instances of calls to action, specifically urging the reader to click for more information or to take specific steps

No Comments

No Known Activations