INDEX

Explanations

words related to packaging or containing items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ÐµÑĢ

-0.23

ÑĢÐ°Ð²

-0.23

erk

-0.21

-0.20

ÑĢÐ°Ð²Ð¸

-0.20

it

-0.19

ÎµÎ¹

-0.18

POSITIVE LOGITS

rio

0.18

resents

0.18

isode

0.17

ixmap

0.17

ooled

0.17

atri

0.17

roud

0.16

acific

0.15

ocket

0.15

pearance

0.15

Activations Density 0.019%

words related to packaging or containing items

No Comments

No Known Activations