INDEX

Explanations

references to family relationships and dynamics

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

 granddaughter

-0.26

 grandson

-0.25

 fathers

-0.23

 father

-0.19

 Fathers

-0.19

 Parent

-0.18

 mothers

-0.17

ansson

-0.17

 grandfather

-0.17

 dads

-0.17

POSITIVE LOGITS

 siblings

0.34

 brothers

0.33

 brother

0.33

 sisters

0.33

 sibling

0.31

 sister

0.29

 Brother

0.27

 Brothers

0.25

 frat

0.25

åħĦ

0.25

Activations Density 0.087%

references to family relationships and dynamics

No Comments

No Known Activations