INDEX

Explanations

references to in-person attendance and experiences

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

ippet

-0.15

çĲ´

-0.15

 hlav

-0.15

ÐµÐºÑĤÐ¾ÑĢÐ°

-0.15

izio

-0.14

ÑıÑĤÐ¸

-0.14

cente

-0.13

uy

-0.13

cbd

-0.13

esser

-0.13

POSITIVE LOGITS

 physical

0.61

 physically

0.55

 face

0.54

physical

0.53

 Physical

0.50

Physical

0.47

face

0.42

 fÃŃs

0.41

 Face

0.40

 person

0.40

Activations Density 0.255%

references to in-person attendance and experiences

No Comments

No Known Activations