INDEX

Explanations

references to legal disputes and court rulings

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

à¸¨

-0.15

 discredit

-0.15

ÑĢÑĥÐ¶

-0.14

Disappear

-0.14

ibern

-0.14

ÙĪØ¨ÛĮ

-0.13

 Ð½ÐµÐ¿ÑĢÐ¸ÑıÑĤ

-0.13

 seedu

-0.13

ìĤ¬ë¥¼

-0.13

bert

-0.13

POSITIVE LOGITS

 violated

0.28

 discrim

0.28

 Viol

0.28

 viol

0.25

 imper

0.25

viol

0.24

 violates

0.24

vi

0.23

-viol

0.23

 discriminator

0.23

Activations Density 0.156%

references to legal disputes and court rulings

No Comments

No Known Activations