INDEX

Explanations

expressions indicating the act of writing or creating

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

-0.17

 Buck

-0.16

atra

-0.15

eler

-0.14

jour

-0.14

Rub

-0.14

Ð²Ð¸Ð¹

-0.14

Representation

-0.14

-0.13

ĢìĿ´

-0.13

POSITIVE LOGITS

argon

0.17

uteur

0.16

.synthetic

0.16

æľ¬

0.15

 Leer

0.15

ãģĵãģĵ

0.15

uess

0.15

 Äĳang

0.15

 BÃłi

0.14

-wsj

0.14

Activations Density 0.162%

expressions indicating the act of writing or creating

No Comments

No Known Activations