INDEX

Explanations

expressions of opinion and belief related to various topics

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

ÅĪ

-0.17

 Vill

-0.17

Duy

-0.16

instein

-0.15

elda

-0.15

orp

-0.15

 cycles

-0.15

xo

-0.15

wan

-0.14

apa

-0.14

POSITIVE LOGITS

vore

0.18

chluss

0.15

erdem

0.15

Sez

0.14

adaki

0.14

 Kurd

0.13

 Haven

0.13

yses

0.13

aton

0.13

Activations Density 0.110%

expressions of opinion and belief related to various topics

No Comments

No Known Activations