INDEX

Explanations

references to web-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

leo

-0.19

onis

-0.16

rive

-0.15

ylan

-0.15

iaux

-0.14

èĢħçļĦ

-0.14

DCALL

-0.14

fy

-0.14

oppins

-0.14

lando

-0.14

POSITIVE LOGITS

iste

0.26

inars

0.25

isodes

0.24

isode

0.22

 presence

0.22

 site

0.21

inar

0.20

-based

0.20

inaire

0.20

istes

0.20

Activations Density 0.020%

references to web-related topics

No Comments

No Known Activations