INDEX

Explanations

terms related to software functionalities and features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

egov

-0.17

avir

-0.16

ixel

-0.15

stadt

-0.15

ebi

-0.15

Ð¾ÑĢÐ¾Ð½

-0.14

å»ł

-0.14

 rebut

-0.14

ersed

-0.14

à¸Ļà¸ķ

-0.13

POSITIVE LOGITS

0.31

Pro

0.30

 Professional

0.29

 Lite

0.28

 lite

0.26

 Plus

0.25

 Deluxe

0.24

pro

0.24

 Classic

0.24

PRO

0.23

Activations Density 0.196%

terms related to software functionalities and features

No Comments

No Known Activations