INDEX

Explanations

numeric expressions related to dimensions and sizes

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

Not in Any Lists

No Comments

Negative Logits

quent

-0.16

ãģĻãģĻ

-0.15

icycle

-0.15

.tencent

-0.15

egan

-0.15

umatic

-0.15

achi

-0.15

iov

-0.14

ruz

-0.14

mando

-0.14

POSITIVE LOGITS

#__

0.15

kÃ¼

0.15

 Tart

0.14

/~

0.14

 Seite

0.14

 decom

0.14

LP

0.14

Ã¤ft

0.14

plor

0.14

FO

0.13

Activations Density 0.022%

numeric expressions related to dimensions and sizes

No Comments

No Known Activations