INDEX

Explanations

references to legal charges or criminal activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

wor

-0.14

æĺŃ

-0.14

à¥ģà¤²à¤¨

-0.14

fik

-0.14

imon

-0.14

sue

-0.13

ipop

-0.13

Ãªnh

-0.13

ÑĪÐ¸Ð±

-0.13

vor

-0.13

POSITIVE LOGITS

 charges

0.26

bond

0.24

 charge

0.24

 warrants

0.24

 bond

0.23

 charged

0.23

 booking

0.22

 charging

0.22

 warrant

0.21

 arrest

0.21

Activations Density 0.093%

references to legal charges or criminal activities

No Comments

No Known Activations