INDEX

Explanations

numerical expressions related to quantities and counts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 unfavorable

-0.22

 favors

-0.21

 favorable

-0.21

 colorful

-0.20

 caliber

-0.20

favor

-0.20

 leveling

-0.19

 favor

-0.19

 honors

-0.19

 neighborhood

-0.19

POSITIVE LOGITS

 metre

0.23

 further

0.22

 programmes

0.20

 Further

0.19

 Maths

0.19

Further

0.18

 metres

0.18

mins

0.18

etre

0.18

 adverts

0.18

Activations Density 0.203%

numerical expressions related to quantities and counts

No Comments

No Known Activations