INDEX

Explanations

expressions of gratitude and acknowledgments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ë®¤

-0.15

untime

-0.13

Ã¸r

-0.13

StringLength

-0.13

ê³

-0.13

swer

-0.13

.SDK

-0.13

Î´ÏĮ

-0.13

upe

-0.13

antal

-0.13

POSITIVE LOGITS

 thank

0.54

 thanks

0.47

 Thank

0.46

 THANK

0.43

 Thanks

0.43

Thanks

0.41

Thank

0.41

 thanked

0.39

 thanking

0.39

thanks

0.38

Activations Density 0.154%

expressions of gratitude and acknowledgments

No Comments

No Known Activations