INDEX

Explanations

terms related to law enforcement and legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ackbar

-0.15

cif

-0.14

antan

-0.14

odon

-0.14

 Contract

-0.14

kre

-0.14

amburger

-0.14

bens

-0.14

Įĵ

-0.13

atrice

-0.13

POSITIVE LOGITS

PC

0.26

 const

0.23

PCs

0.22

 Const

0.22

PC

0.21

Pc

0.21

(PC

0.20

mag

0.19

‘

0.19

 magistrate

0.17

Activations Density 0.002%

terms related to law enforcement and legal contexts

No Comments

No Known Activations