INDEX

Explanations

references to legal courts and judicial processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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¤¤à¤¹

-0.16

etter

-0.15

 vice

-0.15

stroy

-0.15

Äįan

-0.14

è¼Ķ

-0.14

::$_

-0.14

-license

-0.14

-validator

-0.14

lund

-0.14

POSITIVE LOGITS

 Circuit

0.30

 circuit

0.29

 circuits

0.27

 Ninth

0.22

 Circ

0.22

Ele

0.22

 Second

0.22

DC

0.21

 Third

0.20

 Fifth

0.20

Activations Density 0.029%

references to legal courts and judicial processes

No Comments

No Known Activations