INDEX

Explanations

words and phrases related to historical political entities and their characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.community

-0.17

á»ĳi

-0.16

 genie

-0.14

 Indones

-0.14

ç¤¾åĮº

-0.14

 Quarry

-0.14

-community

-0.13

_community

-0.13

IIC

-0.13

edeki

-0.13

POSITIVE LOGITS

 capital

0.29

 flag

0.26

 population

0.25

 flags

0.24

 Capital

0.23

Capital

0.22

 Gross

0.22

capital

0.22

 Flag

0.21

 Flags

0.21

Activations Density 0.121%

words and phrases related to historical political entities and their characteristics

No Comments

No Known Activations