INDEX

Explanations

proper nouns, including names and institutions

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

ifu

-0.15

iffin

-0.14

 Paperback

-0.13

ÑĢÐ°ÑĤÐ½Ð¾

-0.13

or

-0.13

Æ°Æ¡ng

-0.13

ÑĢÐµÐ²

-0.13

 mutually

-0.13

ãĤīãģĹ

-0.13

áº¯ng

-0.13

POSITIVE LOGITS

via

0.36

via

0.31

 courtesy

0.30

Via

0.24

Via

0.23

 Courtesy

0.23

Courtesy

0.23

Unless

0.22

 taken

0.21

cour

0.21

Activations Density 0.065%

proper nouns, including names and institutions

No Comments

No Known Activations