INDEX

Explanations

references to specific games or gaming experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 King

-0.18

 Archie

-0.17

 Kings

-0.15

King

-0.15

lio

-0.14

ABA

-0.14

 kings

-0.14

ÙĤØ§Øª

-0.14

 Sammy

-0.14

ennon

-0.14

POSITIVE LOGITS

EVE

0.43

CCP

0.38

Eve

0.34

ccp

0.29

eve

0.28

eve

0.26

PLEX

0.24

 Citizen

0.23

 Citadel

0.22

 pilots

0.22

Activations Density 0.010%

references to specific games or gaming experiences

No Comments

No Known Activations