INDEX

Explanations

temporal references and phrases indicating progress or advancement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

 clones

-0.15

eller

-0.15

Ã¹i

-0.14

onomous

-0.14

olini

-0.14

Ð±Ð°Ñĩ

-0.14

Ø¶Ùħ

-0.13

 PROCUREMENT

-0.13

iosper

-0.13

 residues

-0.13

POSITIVE LOGITS

 eventually

0.34

 eventual

0.31

cul

0.29

 Eventually

0.28

Eventually

0.27

 ultimately

0.26

 finally

0.23

cul

0.22

ult

0.21

 until

0.20

Activations Density 0.289%

temporal references and phrases indicating progress or advancement

No Comments

No Known Activations