INDEX

Explanations

references to the term "World" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

 xmlhttp

-0.16

anges

-0.15

dfs

-0.14

æĹ

-0.14

 Walters

-0.14

ibt

-0.14

arn

-0.14

à¥įà¤¤à¤ķ

-0.14

 White

-0.13

èħķ

-0.13

POSITIVE LOGITS

 Wide

0.26

Wide

0.24

wide

0.23

 wide

0.23

-wide

0.23

-Class

0.18

-class

0.17

idor

0.17

-ren

0.16

(World

0.16

Activations Density 0.026%

references to the term "World" in various contexts

No Comments

No Known Activations