INDEX

Explanations

references to "Star Wars" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

elho

-0.18

enne

-0.17

erson

-0.17

kaz

-0.17

ymous

-0.17

yses

-0.17

go

-0.16

estro

-0.15

abler

-0.15

Ú¯Ø§Ùĩ

-0.15

POSITIVE LOGITS

 Wars

0.35

 Trek

0.29

 wars

0.28

Wars

0.28

ship

0.25

ry

0.22

vation

0.21

 trek

0.21

ships

0.20

ategy

0.20

Activations Density 0.010%

references to "Star Wars" in various contexts

No Comments

No Known Activations