INDEX

Explanations

phrases related to dimensions or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

foot

-0.16

dol

-0.15

etat

-0.15

er

-0.15

kus

-0.15

mars

-0.15

 Affero

-0.15

ehler

-0.15

heet

-0.14

vip

-0.14

POSITIVE LOGITS

wise

0.37

ened

0.37

ening

0.36

iness

0.26

 wise

0.25

iest

0.25

-wise

0.24

ier

0.23

wis

0.22

ily

0.21

Activations Density 0.071%

phrases related to dimensions or measurements

No Comments

No Known Activations