INDEX

Explanations

the occurrences of the abbreviation "N" followed by a number, typically indicating a numerical reference or a designation in a list

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

Ä±sÄ±ndan

-0.18

ivent

-0.16

uish

-0.15

ÑĭÑĪ

-0.15

uitka

-0.15

achten

-0.14

Ä±sÄ±na

-0.14

ÑıÐ²

-0.14

Hat

-0.14

 crack

-0.14

POSITIVE LOGITS

iger

0.33

igeria

0.30

ollywood

0.28

aira

0.27

nam

0.27

dig

0.26

npc

0.23

ai

0.23

asar

0.22

wan

0.21

Activations Density 0.008%

the occurrences of the abbreviation "N" followed by a number, typically indicating a numerical reference or a designation in a list

No Comments

No Known Activations