INDEX

Explanations

references to legal offenses and their classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ffen

-0.18

ulis

-0.15

ucz

-0.15

ilter

-0.15

udiant

-0.15

ensem

-0.14

ÑĥÑĢÐ¸

-0.14

agos

-0.14

pling

-0.14

rier

-0.13

POSITIVE LOGITS

æ®¿

0.16

amus

0.15

actus

0.14

ORED

0.14

 Sims

0.14

$MESS

0.14

 gravity

0.13

sab

0.13

 committed

0.13

displayText

0.13

Activations Density 0.033%

references to legal offenses and their classifications

No Comments

No Known Activations