INDEX

Explanations

formatting characters typically used in programming code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

rane

-0.16

jde

-0.15

prior

-0.15

jom

-0.14

yme

-0.14

dato

-0.14

ineTransform

-0.13

achs

-0.13

POSITIVE LOGITS

REM

0.17

chu

0.14

IES

0.14

alah

0.14

ç¥Ŀ

0.14

nam

0.14

 numberWith

0.14

hy

0.14

 allen

0.13

Vie

0.13

Activations Density 0.003%

formatting characters typically used in programming code

No Comments

No Known Activations