INDEX

Explanations

references to violent attacks and explosives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ernals

-0.15

holm

-0.15

-0.14

jie

-0.14

 digits

-0.14

anax

-0.14

åĽ´

-0.13

digits

-0.13

ç¼

-0.13

POSITIVE LOGITS

 anonymously

0.15

 plot

0.15

SOLE

0.14

.xhtml

0.14

Ã¡n

0.14

erdale

0.14

IFA

0.14

ednou

0.14

lingen

0.14

TZ

0.13

Activations Density 0.038%

references to violent attacks and explosives

No Comments

No Known Activations