INDEX

Explanations

references to the nose and related medical terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

 priesthood

-0.17

acro

-0.16

æİ

-0.15

éĴ

-0.15

argar

-0.15

iges

-0.15

åªĴ

-0.14

obl

-0.14

olis

-0.14

statt

-0.14

POSITIVE LOGITS

wick

0.16

 noses

0.15

andle

0.15

sey

0.14

ouri

0.14

.Accessible

0.14

 Timothy

0.14

 Evolution

0.14

ouden

0.14

ÑĦÑĸÐº

0.14

Activations Density 0.014%

references to the nose and related medical terminology

No Comments

No Known Activations