INDEX

Explanations

instances of the word "few" and variations of counting or limited quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Numerous

-0.20

è®¸å¤ļ

-0.18

 Plenty

-0.17

åĲĦç§į

-0.17

 numerous

-0.16

/Math

-0.16

Time

-0.15

 Many

-0.15

 plenty

-0.15

Many

-0.15

POSITIVE LOGITS

 dozen

0.42

 hundred

0.38

 thousand

0.29

 mÃ¡lo

0.27

 Hundred

0.26

 select

0.24

 choice

0.23

teenth

0.23

of

0.22

est

0.22

Activations Density 0.057%

instances of the word "few" and variations of counting or limited quantity

No Comments

No Known Activations