INDEX

Explanations

references to video game features and elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

idon

-0.17

idor

-0.16

itz

-0.16

role

-0.16

zilla

-0.15

mania

-0.15

arella

-0.15

 Readers

-0.14

ussy

-0.14

Role

-0.14

POSITIVE LOGITS

 game

0.29

 developers

0.21

 controls

0.21

 demo

0.20

developers

0.20

 devs

0.20

 campaign

0.20

game

0.19

 setting

0.18

 trailer

0.18

Activations Density 0.095%

references to video game features and elements

No Comments

No Known Activations