INDEX

Explanations

instances of authorship in texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

.wik

-0.14

sumer

-0.14

gregator

-0.14

CEE

-0.14

 demokrat

-0.14

 Wich

-0.14

CED

-0.13

ayi

-0.13

Ø¯ÛĮÚ¯Ø±

-0.13

ØµØ±

-0.13

POSITIVE LOGITS

 Feld

0.15

by

0.15

ftime

0.14

 istem

0.13

 Miles

0.13

 Unknown

0.13

uid

0.13

ÐĽÑİ

0.13

 Various

0.13

Activations Density 0.035%

instances of authorship in texts

No Comments

No Known Activations