INDEX

Explanations

references to body parts and injuries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 mouths

-0.18

_heads

-0.18

 sides

-0.17

odate

-0.17

 Hands

-0.16

 shoulders

-0.16

ãģ£ãģ±

-0.16

ahat

-0.16

 palms

-0.16

 backs

-0.16

POSITIVE LOGITS

leg

0.44

arm

0.35

 foot

0.33

leg

0.28

 knee

0.26

 ankle

0.25

ear

0.24

arm

0.24

 shin

0.24

toe

0.24

Activations Density 0.095%

references to body parts and injuries

No Comments

No Known Activations