INDEX

Explanations

phrases indicating bureaucratic processes and official communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

/apt

-0.16

stu

-0.15

alent

-0.15

rone

-0.15

æĲŃ

-0.15

ilia

-0.14

cia

-0.14

heels

-0.14

gary

-0.14

PageIndex

-0.14

POSITIVE LOGITS

 Additional

0.29

Dy

0.25

Additional

0.23

nod

0.22

 additional

0.20

 Joint

0.19

 Head

0.18

Vig

0.18

 officer

0.18

 nods

0.18

Activations Density 0.112%

phrases indicating bureaucratic processes and official communication

No Comments

No Known Activations