INDEX

Explanations

references to specific characters or cast members in a narrative

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

erer

-0.19

cke

-0.17

annes

-0.17

geh

-0.16

zeigen

-0.16

er

-0.16

udson

-0.16

jez

-0.16

suppress

-0.15

POSITIVE LOGITS

igated

0.33

aways

0.33

away

0.32

igation

0.31

igate

0.28

ellan

0.28

 iron

0.27

les

0.27

ings

0.26

ell

0.24

Activations Density 0.020%

references to specific characters or cast members in a narrative

No Comments

No Known Activations