INDEX

Explanations

references to key figures or organizations associated with terrorism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

iversal

-0.15

 Lutheran

-0.15

Ø±ÙģØªÙĩ

-0.15

-0.14

Ø§ØŃØª

-0.14

allas

-0.14

IAS

-0.14

 Bourbon

-0.13

ÑĤÐ¸Ð²

-0.13

POSITIVE LOGITS

bin

0.22

al

0.22

 wanted

0.20

 sanct

0.20

 safe

0.20

AQ

0.20

 leadership

0.19

 leader

0.19

leader

0.19

 senior

0.18

Activations Density 0.016%

references to key figures or organizations associated with terrorism

No Comments

No Known Activations