INDEX

Explanations

references to law enforcement and affiliated entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

zych

-0.17

icha

-0.16

 solic

-0.16

Ð¼Ð°Ð·

-0.16

olic

-0.15

ovich

-0.15

 Richie

-0.15

ayne

-0.15

 uncont

-0.14

.appspot

-0.14

POSITIVE LOGITS

Ci

0.17

roc

0.15

ByKey

0.15

ROC

0.15

dio

0.14

ScreenState

0.14

arkan

0.14

 Loren

0.14

_frames

0.14

 Bullet

0.13

Activations Density 0.233%

references to law enforcement and affiliated entities

No Comments

No Known Activations