INDEX

Explanations

preferences related to favorites in various categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

letal

-0.18

buz

-0.16

afc

-0.16

ollen

-0.15

ulum

-0.15

uluk

-0.15

arton

-0.15

embali

-0.15

creativecommons

-0.15

abin

-0.14

POSITIVE LOGITS

 favorite

0.28

 favourite

0.26

 Favorite

0.23

 favorites

0.22

favorite

0.21

 childhood

0.20

 Childhood

0.20

 favourites

0.19

Favorite

0.19

 underrated

0.16

Activations Density 0.095%

preferences related to favorites in various categories

No Comments

No Known Activations