INDEX

Explanations

terms and phrases related to historical events and figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ucher

-0.19

 erotique

-0.17

edor

-0.16

 geschichten

-0.15

izzo

-0.15

?option

-0.15

ulur

-0.14

iliz

-0.14

 Wheat

-0.14

uve

-0.13

POSITIVE LOGITS

 belt

0.18

bou

0.17

erva

0.17

 probe

0.17

ke

0.16

bel

0.16

importe

0.15

kel

0.15

 vert

0.15

ha

0.15

Activations Density 0.016%

terms and phrases related to historical events and figures

No Comments

No Known Activations