INDEX

Explanations

references to individuals and their professional accomplishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

usic

-0.15

ym

-0.14

ÐµÐºÑĥ

-0.14

ÃľM

-0.14

ÑĥÑĢÐ¸

-0.14

rowsable

-0.14

ogle

-0.13

lish

-0.13

.pretty

-0.13

echan

-0.13

POSITIVE LOGITS

 speaking

0.41

 speaker

0.39

 Speaking

0.36

 keynote

0.33

 speakers

0.33

 Speaker

0.32

Speaking

0.32

 Speakers

0.31

 motivational

0.29

-spe

0.29

Activations Density 0.110%

references to individuals and their professional accomplishments

No Comments

No Known Activations