INDEX

Explanations

keywords and numerical data related to research topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

inton

-0.15

 GÃ¼

-0.15

ingu

-0.15

cq

-0.14

-gnu

-0.14

_weather

-0.14

 Deus

-0.14

agan

-0.13

ahu

-0.13

CRE

-0.13

POSITIVE LOGITS

æľĭ

0.17

olu

0.15

 ãģ¿

0.15

/books

0.15

 ãģį

0.15

figcaption

0.14

apons

0.14

Î¿Ïģ

0.14

Kee

0.14

Ø§Ø¹Ø¯

0.14

Activations Density 0.033%

keywords and numerical data related to research topics

No Comments

No Known Activations