INDEX

Explanations

words and phrases associated with official positions, roles, or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Ð²Ð°Ð½Ð¾Ð²

-0.16

enheim

-0.15

eco

-0.15

VICE

-0.14

Į

-0.14

 zÃ¡vod

-0.14

McG

-0.14

 analogy

-0.14

à¹Ģà¸Ńà¸ĩ

-0.14

esity

-0.13

POSITIVE LOGITS

 Person

0.19

Rel

0.17

 person

0.17

rel

0.17

kategori

0.16

ilyn

0.16

 kategor

0.15

avl

0.15

Person

0.15

avn

0.15

Activations Density 0.013%

words and phrases associated with official positions, roles, or titles

No Comments

No Known Activations