INDEX

Explanations

instances of the word "just" and variations involving frequency or intensity

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

inden

-0.15

OUCH

-0.15

woff

-0.14

ë°į

-0.14

undy

-0.14

esson

-0.14

ITHER

-0.14

 atleast

-0.14

antha

-0.14

otine

-0.14

POSITIVE LOGITS

 days

0.23

 hours

0.22

 DAYS

0.20

 feet

0.20

 weeks

0.20

 steps

0.19

 minutes

0.19

 mere

0.19

one

0.19

 months

0.19

Activations Density 0.073%

instances of the word "just" and variations involving frequency or intensity

No Comments

No Known Activations