INDEX

Explanations

structural or functional elements related to written documents and content formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

-0.17

imir

-0.16

sonian

-0.15

.getOwnProperty

-0.15

oji

-0.15

EdgeInsets

-0.15

etti

-0.14

LineColor

-0.14

ijken

-0.14

lena

-0.14

POSITIVE LOGITS

kas

0.17

kÃ¡

0.15

agr

0.15

zung

0.15

 burg

0.15

cht

0.14

 Reign

0.14

aÅŁ

0.14

/flutter

0.14

 SÄ±

0.14

Activations Density 0.035%

structural or functional elements related to written documents and content formatting

No Comments

No Known Activations