INDEX

Explanations

references to terrorist organizations and their activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

Licensed

-0.18

pendicular

-0.15

 Licensed

-0.14

sdale

-0.14

arta

-0.14

Bou

-0.14

 McCabe

-0.14

çīĩ

-0.13

 Glock

-0.13

-independent

-0.13

POSITIVE LOGITS

 sympath

0.24

-controlled

0.22

 leader

0.21

 forces

0.21

 controlled

0.21

-held

0.20

-Control

0.19

controlled

0.19

 Leader

0.18

-aff

0.18

Activations Density 0.092%

references to terrorist organizations and their activities

No Comments

No Known Activations