INDEX

Explanations

references to incidents involving law enforcement actions and violence

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

lamaz

-0.16

Ã´me

-0.15

idges

-0.14

occo

-0.14

ughs

-0.14

WARDED

-0.14

ivas

-0.14

 Typeface

-0.13

anded

-0.13

HORT

-0.13

POSITIVE LOGITS

arer

0.15

æ¶ī

0.15

 gerade

0.15

char

0.14

 innocent

0.14

Popular

0.14

lee

0.14

 ironically

0.14

 Popular

0.13

nap

0.13

Activations Density 0.142%

references to incidents involving law enforcement actions and violence

No Comments

No Known Activations