INDEX

Explanations

references to familial or personal relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

que

-0.15

ovit

-0.15

 Ð¾Ð±Ð¾

-0.15

McG

-0.14

esel

-0.14

Both

-0.13

udoku

-0.13

ez

-0.13

 Dank

-0.13

ãĢĳ,

-0.13

POSITIVE LOGITS

 others

0.31

 other

0.24

others

0.23

 Others

0.22

Others

0.20

 crew

0.19

his

0.18

åħ¶ä»ĸ

0.17

 colleagues

0.17

 autres

0.16

Activations Density 0.077%

references to familial or personal relationships

No Comments

No Known Activations