INDEX

Explanations

references to in-game mechanics and player interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

engin

-0.15

ence

-0.15

scribe

-0.14

 McKay

-0.14

tie

-0.14

phin

-0.14

ãĤ¸ãĤ¢

-0.14

ROTO

-0.14

 Warcraft

-0.13

zew

-0.13

POSITIVE LOGITS

 Completion

0.23

Completion

0.22

completion

0.21

 play

0.21

 completion

0.21

 save

0.20

 playing

0.19

save

0.18

 saves

0.18

playing

0.18

Activations Density 0.207%

references to in-game mechanics and player interactions

No Comments

No Known Activations