INDEX

Explanations

terms related to body parts and their specific conditions or applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

etur

-0.16

steller

-0.16

fab

-0.15

 finger

-0.15

elli

-0.14

á»±c

-0.14

 hand

-0.14

ACA

-0.14

ween

-0.14

kla

-0.14

POSITIVE LOGITS

bone

0.24

/back

0.23

band

0.22

/head

0.22

éĥ¨

0.22

bones

0.21

-high

0.21

/el

0.20

-length

0.19

laces

0.19

Activations Density 0.055%

terms related to body parts and their specific conditions or applications

No Comments

No Known Activations