INDEX

Explanations

references to royal figures, particularly those with the title "Prince."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

anmar

-0.19

ctype

-0.17

agogue

-0.17

 sapi

-0.15

acus

-0.15

isoft

-0.15

 povol

-0.14

RITE

-0.14

 Ð¿ÑĢÐ¸ÑĤ

-0.14

acity

-0.14

POSITIVE LOGITS

(ss

0.24

/ss

0.24

ps

0.23

 Charles

0.21

 Harry

0.21

 Edward

0.20

 Consort

0.20

 Rupert

0.20

 charming

0.20

 William

0.20

Activations Density 0.008%

references to royal figures, particularly those with the title "Prince."

No Comments

No Known Activations