INDEX

Explanations

phrases that encourage or prompt the reader to take action, particularly to "check" or "look" at something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

itori

-0.15

/from

-0.15

-scalable

-0.15

alth

-0.15

uncios

-0.15

ÑģÑĤÐ¾ÑĢÐ¸Ñı

-0.14

apia

-0.14

atcher

-0.14

olis

-0.14

stime

-0.14

POSITIVE LOGITS

ered

0.32

 back

0.30

mate

0.26

mark

0.26

-in

0.25

lists

0.25

 below

0.22

 into

0.22

out

0.22

 mate

0.20

Activations Density 0.013%

phrases that encourage or prompt the reader to take action, particularly to "check" or "look" at something

No Comments

No Known Activations