INDEX

Explanations

technical terms and elements related to data management and support systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Ingen

-0.15

esser

-0.15

Poll

-0.15

infeld

-0.14

 onNext

-0.14

 Blond

-0.14

ateau

-0.14

ÐµÐ¼Ð¾Ð½

-0.14

braco

-0.14

POSITIVE LOGITS

alist

0.17

 Perry

0.14

ted

0.14

 Advis

0.14

åĩ¡

0.13

BX

0.13

 Yates

0.13

<Map

0.13

angkan

0.13

 McCartney

0.13

Activations Density 0.007%

technical terms and elements related to data management and support systems

No Comments

No Known Activations