INDEX

Explanations

terms and phrases related to deception or dishonesty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ê´

-0.16

ylko

-0.15

 Singleton

-0.14

 Pazar

-0.14

bek

-0.14

Ø¡

-0.14

Î³ÎºÎµ

-0.14

 ê´

-0.14

ç¦®

-0.13

atar

-0.13

POSITIVE LOGITS

0.15

rah

0.15

ukt

0.14

olis

0.14

arto

0.14

chers

0.14

Tre

0.14

ARCH

0.13

Fi

0.13

 duplic

0.13

Activations Density 0.033%

terms and phrases related to deception or dishonesty

No Comments

No Known Activations