INDEX

Explanations

features that enhance usability and performance in products

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

Å¾il

-0.18

zdy

-0.15

odom

-0.14

arrants

-0.14

iras

-0.14

asers

-0.14

ajs

-0.14

blings

-0.14

apolis

-0.14

á»§ng

-0.14

POSITIVE LOGITS

 thanks

0.35

 without

0.33

 while

0.29

thanks

0.29

without

0.28

while

0.26

 Thanks

0.26

 ideal

0.25

 WITHOUT

0.24

 wherever

0.24

Activations Density 0.351%

features that enhance usability and performance in products

No Comments

No Known Activations