INDEX

Explanations

references to specific academic identifiers or metrics related to research publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

onen

-0.16

 another

-0.16

 these

-0.15

if

-0.15

rella

-0.15

dad

-0.14

 latter

-0.14

.Formatting

-0.14

 Refer

-0.14

jadi

-0.14

POSITIVE LOGITS

 Purpose

0.26

 PURPOSE

0.25

 purpose

0.24

BACKGROUND

0.23

 Objective

0.23

 OBJECT

0.23

_Object

0.22

æĳĺè¦ģ

0.22

 Background

0.22

<j

0.21

Activations Density 0.049%

references to specific academic identifiers or metrics related to research publications

No Comments

No Known Activations