INDEX

Explanations

phrases related to comfort and safety

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

strup

-0.16

ovah

-0.15

 uncomment

-0.15

hangi

-0.15

owered

-0.15

vard

-0.15

icago

-0.15

æĪ¸

-0.15

hung

-0.15

baugh

-0.15

POSITIVE LOGITS

ability

0.40

able

0.37

 zone

0.35

ab

0.33

ably

0.32

 zones

0.32

abl

0.31

ingly

0.29

-zone

0.28

zone

0.27

Activations Density 0.025%

phrases related to comfort and safety

No Comments

No Known Activations