INDEX

Explanations

references to navigation or article pagination

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 lekker

-0.15

eyse

-0.15

asion

-0.15

 cáº§m

-0.15

 Jahres

-0.15

_UNSIGNED

-0.14

celain

-0.14

ÑĢÑĥÑĪ

-0.14

Ð¡Ð¡Ðł

-0.14

ãĥĲãĥ¼

-0.14

POSITIVE LOGITS

 article

0.36

 articles

0.29

 artic

0.26

 Articles

0.24

article

0.23

/article

0.22

 Article

0.21

(article

0.19

articles

0.19

ar

0.19

Activations Density 0.002%

references to navigation or article pagination

No Comments

No Known Activations