INDEX

Explanations

phrases that include the word "what's" and questions or inquiries regarding new content or happenings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

irected

-0.16

annon

-0.16

uby

-0.15

quals

-0.14

emer

-0.14

nemonic

-0.14

iá»ģn

-0.14

ooter

-0.14

okol

-0.13

 mÄ±

-0.13

POSITIVE LOGITS

 wrong

0.25

Wrong

0.22

 Wrong

0.21

 happened

0.21

 happening

0.21

wrong

0.20

 stopping

0.20

_wrong

0.19

Missing

0.18

inan

0.18

Activations Density 0.028%

phrases that include the word "what's" and questions or inquiries regarding new content or happenings

No Comments

No Known Activations