INDEX

Explanations

phrases indicating usage or participation across multiple organizations and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

Ä±ÅŁÄ±

-0.15

lor

-0.15

ä¸ĸ

-0.14

oru

-0.14

 century

-0.13

ãĤ¢ãĥ³

-0.13

Î½Îµ

-0.13

 Refuge

-0.13

west

-0.13

Ten

-0.13

POSITIVE LOGITS

 dozens

0.25

 several

0.21

 many

0.21

 multiple

0.20

many

0.19

 dozen

0.19

 countries

0.18

 across

0.18

ultiple

0.18

 five

0.17

Activations Density 0.144%

phrases indicating usage or participation across multiple organizations and events

No Comments

No Known Activations