INDEX

Explanations

structured data formats, specifically XML and JSON elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

biÃªn

-0.15

obao

-0.14

iyah

-0.14

atre

-0.14

 ChÃŃ

-0.14

ertools

-0.14

-0.13

 Sexe

-0.13

ä¸ŃæĸĩåŃĹå¹ķ

-0.13

POSITIVE LOGITS

 test

0.28

 testing

0.27

0.25

0.23

 Testing

0.23

 Test

0.23

 Lorem

0.22

 testers

0.22

 tester

0.22

test

0.21

Activations Density 0.110%

structured data formats, specifically XML and JSON elements

No Comments

No Known Activations