INDEX

Explanations

references to people and their roles or relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ydk

-0.15

ebek

-0.14

"<?

-0.14

.scalablytyped

-0.14

ìĽ

-0.14

ampoo

-0.14

wur

-0.14

ì§Ģ

-0.14

wz

-0.13

phy

-0.13

POSITIVE LOGITS

Suc

0.15

antha

0.15

 Readonly

0.15

 upper

0.14

ï¿½t

0.14

ãĥ³ãĥĩ

0.14

Suc

0.14

Upper

0.14

ches

0.14

Activations Density 1.028%

references to people and their roles or relationships

No Comments

No Known Activations