INDEX

Explanations

references to the term "Southern" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

pare

-0.16

auge

-0.15

iac

-0.15

ensual

-0.14

enton

-0.14

olk

-0.14

 northwest

-0.14

elenium

-0.14

uar

-0.14

 sack

-0.14

POSITIVE LOGITS

most

0.32

 Hemisphere

0.25

-most

0.23

 Tier

0.23

 hemisphere

0.21

-tier

0.20

 tier

0.20

 Fried

0.20

 Exposure

0.19

 Baptist

0.18

Activations Density 0.011%

references to the term "Southern" in various contexts

No Comments

No Known Activations