INDEX

Explanations

references to geographic locations and neighborhoods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

 è£

-0.17

 Columbia

-0.15

?url

-0.15

UILayout

-0.15

SW

-0.15

SW

-0.14

 ++)↵

-0.14

åĵ¥

-0.14

è²¡

-0.14

udson

-0.14

POSITIVE LOGITS

 Queens

0.27

 queens

0.21

Que

0.19

 Jamaica

0.19

0.18

 Nass

0.16

 Long

0.16

Kew

0.16

 Suffolk

0.16

eless

0.15

Activations Density 0.172%

references to geographic locations and neighborhoods

No Comments

No Known Activations