INDEX

Explanations

references to museums and museum-related activities

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

Ð½Ð¾Ñĩ

-0.16

annonce

-0.15

 sÃłng

-0.14

oulos

-0.14

ãĥ¼ãĤ¿

-0.13

 Apost

-0.13

ë³¸

-0.13

ombo

-0.13

-reply

-0.12

enta

-0.12

POSITIVE LOGITS

 museum

0.43

 exhibits

0.38

 Museum

0.35

 visitor

0.34

 visitors

0.32

 museums

0.32

 exhibit

0.31

 displays

0.30

useum

0.30

 Visitors

0.29

Activations Density 0.260%

references to museums and museum-related activities

No Comments

No Known Activations