INDEX

Explanations

references to authority, UFO-related topics, and significant historical figures in science

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

[sizeof

-0.15

versible

-0.14

 transported

-0.14

 removable

-0.14

oyer

-0.14

 MOUSE

-0.14

Sad

-0.13

avanaugh

-0.13

 Brett

-0.13

áº¹

-0.13

POSITIVE LOGITS

 sightings

0.30

UFO

0.29

 sighting

0.28

uf

0.27

 paranormal

0.25

Sas

0.24

 Sight

0.22

UF

0.22

/Area

0.21

 Paran

0.20

Activations Density 0.127%

references to authority, UFO-related topics, and significant historical figures in science

No Comments

No Known Activations