INDEX

Explanations

references to specific malware and cyber espionage activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

Unsafe

-0.20

statusCode

-0.15

rape

-0.14

 desert

-0.14

olla

-0.14

éªĮ

-0.14

 Unsafe

-0.14

.modelo

-0.14

mom

-0.14

rim

-0.13

POSITIVE LOGITS

.rt

0.18

ÐºÐ¾Ð³

0.16

 Foreign

0.15

 Attribution

0.15

.Toolkit

0.15

ionage

0.15

 directed

0.15

ÑĪÐ¾Ð²

0.15

 attribution

0.15

.dds

0.15

Activations Density 0.137%

references to specific malware and cyber espionage activities

No Comments

No Known Activations