INDEX

Explanations

references to timelines and updates related to regular publications or series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ä»Ĭå¹´

-0.18

nech

-0.16

bsites

-0.15

ç±į

-0.14

atern

-0.13

ÑĪÐµÐ²

-0.13

ickets

-0.13

ãĥ¼ãĥŃ

-0.13

ufen

-0.13

usercontent

-0.13

POSITIVE LOGITS

 newsletter

0.33

 column

0.28

 weekly

0.28

 Weekly

0.27

 Newsletter

0.27

 newsletters

0.27

column

0.24

 update

0.23

 edition

0.23

newsletter

0.23

Activations Density 0.162%

references to timelines and updates related to regular publications or series

No Comments

No Known Activations