INDEX

Explanations

references to individuals and their relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

Jew

-0.23

 Jewish

-0.21

 Jews

-0.17

Äįet

-0.16

 McCabe

-0.16

 Jude

-0.15

jerne

-0.15

quan

-0.15

ÙĨØ¬

-0.15

 Judaism

-0.14

POSITIVE LOGITS

Ya

0.30

Yo

0.29

Av

0.28

Mos

0.27

Ya

0.26

Men

0.25

 Nissan

0.25

Yak

0.24

It

0.24

Mos

0.23

Activations Density 0.059%

references to individuals and their relationships

No Comments

No Known Activations