INDEX

Explanations

references to fear and danger

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

 twink

-0.16

Ã¥de

-0.15

 natural

-0.14

ifi

-0.14

 Lyons

-0.13

 naturally

-0.13

 Burst

-0.13

 recommended

-0.13

POSITIVE LOGITS

à¥įà¤°à¤£

0.15

ä¼¼çļĦ

0.14

LBL

0.14

 è¡ĮæĶ¿

0.14

uncate

0.14

HIR

0.14

afen

0.14

intent

0.14

ÐºÐ°Ð¿

0.14

 INLINE

0.13

Activations Density 0.606%

references to fear and danger

No Comments

No Known Activations