INDEX

Explanations

connections between different locations and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ãĥ¼ãĥ

-0.17

ird

-0.16

esis

-0.15

 Geographic

-0.15

 geographic

-0.15

epar

-0.14

gap

-0.14

æ¿ĥ

-0.13

elan

-0.13

inar

-0.13

POSITIVE LOGITS

 elsewhere

0.35

 abroad

0.31

 beyond

0.27

Else

0.26

 Else

0.24

Beyond

0.23

 Beyond

0.22

eyond

0.21

 throughout

0.18

 Europe

0.18

Activations Density 0.116%

connections between different locations and entities

No Comments

No Known Activations