INDEX

Explanations

periods or full stops in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

    č↵č↵

-0.15

à¸£à¸²à¸¢

-0.14

gress

-0.14

weigh

-0.14

ktor

-0.14

htdocs

-0.14

Ã¨ge

-0.14

-0.13

éĬ

-0.13

POSITIVE LOGITS

boy

0.15

zek

0.14

ADA

0.14

abus

0.14

ÂĿ

0.13

hots

0.13

 BaÄŁ

0.13

ulu

0.13

 Lakes

0.13

 Cook

0.13

Activations Density 0.028%

periods or full stops in text

No Comments

No Known Activations