INDEX

Explanations

terms and concepts related to human rights and their violations

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

hm

-0.15

áº¬

-0.14

hek

-0.14

imentos

-0.14

otts

-0.14

 prÃ³

-0.14

lemen

-0.13

ha

-0.13

 acquaintance

-0.13

gebn

-0.13

POSITIVE LOGITS

 violations

0.31

 violation

0.30

 viol

0.27

 Viol

0.27

 abuses

0.26

viol

0.23

 defenders

0.23

 abuse

0.23

 defender

0.22

 violated

0.22

Activations Density 0.026%

terms and concepts related to human rights and their violations

No Comments

No Known Activations