INDEX

Explanations

references to groups of people, including demographics and roles within society

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

inci

-0.16

coni

-0.16

obao

-0.15

orge

-0.15

rung

-0.14

angan

-0.13

roti

-0.13

urd

-0.13

 prepend

-0.13

illis

-0.13

POSITIVE LOGITS

 alike

1.53

 equally

0.71

 respectively

0.57

ä¸Ģæł·

0.39

 respective

0.38

 equal

0.37

 gleich

0.36

 similarly

0.35

 igual

0.34

 Equal

0.30

Activations Density 0.160%

references to groups of people, including demographics and roles within society

No Comments

No Known Activations