INDEX

Explanations

references to handles or grips on various objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ikk

-0.16

 Hust

-0.15

Hod

-0.14

tep

-0.14

vyk

-0.14

nem

-0.14

uracy

-0.14

ovu

-0.14

 Jaune

-0.14

á»§i

-0.14

POSITIVE LOGITS

cre

0.18

less

0.16

á»ĳc

0.15

 Satellite

0.15

MMM

0.14

 Firm

0.14

izu

0.14

âĢĮØ³ÛĮ

0.14

 satellite

0.14

à¸²à¸Ħà¸²à¸£

0.14

Activations Density 0.028%

references to handles or grips on various objects

No Comments

No Known Activations