INDEX

Explanations

phrases or terms indicating conditions, requirements, or aspects related to completion and restrictions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

chg

-0.16

eneg

-0.15

ovable

-0.15

 Geld

-0.14

 MONEY

-0.14

tras

-0.14

hur

-0.14

dÃ¼k

-0.14

ertz

-0.14

VRT

-0.13

POSITIVE LOGITS

 exports

0.17

exports

0.17

 physical

0.17

physical

0.15

_lvl

0.15

 Book

0.15

 physically

0.14

 export

0.14

rition

0.14

Book

0.14

Activations Density 0.008%

phrases or terms indicating conditions, requirements, or aspects related to completion and restrictions

No Known Activations