INDEX

Explanations

references to innocence and victims of violence

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

adan

-0.16

PropertyValue

-0.15

inary

-0.15

lfw

-0.15

alog

-0.15

ly

-0.14

istrovstvÃŃ

-0.14

 ÙħÙĪÙĤØ¹

-0.14

è·¡

-0.14

POSITIVE LOGITS

 innocent

0.25

 Innoc

0.25

 bystand

0.25

 innoc

0.23

 innocence

0.21

 harmless

0.18

 victims

0.18

 civilians

0.17

 inoc

0.17

uous

0.16

Activations Density 0.013%

references to innocence and victims of violence

No Comments

No Known Activations