INDEX

Explanations

references to user profiles and account management

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

au

-0.16

 elsewhere

-0.15

amps

-0.15

 others

-0.15

etri

-0.14

 theirs

-0.14

uch

-0.13

 periodically

-0.13

à¸µà¹Ī

-0.12

OTHER

-0.12

POSITIVE LOGITS

 every

0.56

 each

0.52

each

0.45

æ¯ı

0.44

 ÐºÐ°Ð¶Ð´

0.44

every

0.42

 chaque

0.41

 ogni

0.40

 kaÅ¼

0.40

 má»Ĺi

0.39

Activations Density 0.186%

references to user profiles and account management

No Comments

No Known Activations