INDEX

Explanations

phrases expressing needs, desires, and actions related to community and individual contributions

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

 Ð±ÑĭÑĤÑĮ

-0.16

ajs

-0.15

Being

-0.15

being

-0.15

oke

-0.15

can

-0.14

been

-0.14

ault

-0.14

izik

-0.14

å¯

-0.14

POSITIVE LOGITS

is

0.41

çļĦæĺ¯

0.39

are

0.28

was

0.25

 ÙĩÙĪ

0.24

,is

0.24

"is

0.23

ãģ®ãģ¯

0.23

$is

0.23

 besides

0.23

Activations Density 0.118%

phrases expressing needs, desires, and actions related to community and individual contributions

No Comments

No Known Activations