INDEX

Explanations

references to academic journals or research publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

pany

-0.17

Bil

-0.16

edic

-0.15

Î»ÏĮ

-0.15

iro

-0.14

Ø¬ÙĨ

-0.14

lobal

-0.14

utzer

-0.14

entionPolicy

-0.13

leine

-0.13

POSITIVE LOGITS

erk

0.16

ohl

0.16

º

0.16

ToUpper

0.14

sez

0.14

 tslint

0.14

owell

0.14

arget

0.14

keh

0.14

strup

0.14

Activations Density 0.002%

references to academic journals or research publications

No Comments

No Known Activations