INDEX

Explanations

expressions indicating the act of indicating or showing something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ç¦»

-0.16

Ø®Øª

-0.16

rior

-0.16

appable

-0.16

throp

-0.15

thus

-0.14

zure

-0.14

erty

-0.14

Ð»ÐµÐº

-0.14

ventus

-0.14

POSITIVE LOGITS

 finger

0.28

finger

0.27

 fingers

0.26

 Finger

0.25

blank

0.23

ificate

0.22

 Ð¿Ð°Ð»ÑĮ

0.21

out

0.20

ificates

0.19

 towards

0.17

Activations Density 0.014%

expressions indicating the act of indicating or showing something

No Comments

No Known Activations