INDEX

Explanations

references to game mechanics and updates

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

aml

-0.17

ebra

-0.15

ÙħÙĦ

-0.15

sk

-0.15

bri

-0.14

 Kingdom

-0.14

//{{

-0.14

McN

-0.14

strap

-0.14

POSITIVE LOGITS

Ten

0.27

War

0.25

Ten

0.22

 Corpus

0.22

 Vall

0.20

 Solar

0.20

 Rail

0.20

 Fort

0.20

 Digital

0.20

 Sent

0.20

Activations Density 0.004%

references to game mechanics and updates

No Comments

No Known Activations