INDEX

Explanations

names and identifiers, including proper nouns and titles

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

-0.18

ï¼ĪæĺŃåĴĮ

-0.14

Arizona

-0.13

 svens

-0.13

 behaviours

-0.13

 Arizona

-0.13

peon

-0.13

"](

-0.13

 (Â«

-0.12

PN

-0.12

POSITIVE LOGITS

 Marx

0.35

Heg

0.31

 Marxist

0.25

 Marxism

0.24

 Lenin

0.23

 Kant

0.20

 Rousse

0.20

 Karl

0.20

 liberalism

0.20

 Enlightenment

0.20

Activations Density 0.003%

names and identifiers, including proper nouns and titles

No Comments

No Known Activations