INDEX

Explanations

instances of the word "Read" indicating calls to action or references to further information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

eldorf

-0.17

/perl

-0.16

anner

-0.16

elder

-0.15

glich

-0.15

ples

-0.15

dra

-0.15

berman

-0.14

alf

-0.14

Reader

-0.14

POSITIVE LOGITS

mission

0.29

ily

0.28

ers

0.26

ings

0.25

missions

0.24

ystate

0.24

 more

0.23

ying

0.23

 More

0.23

iness

0.22

Activations Density 0.022%

instances of the word "Read" indicating calls to action or references to further information

No Comments

No Known Activations