INDEX

Explanations

references to brand names and their attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

bral

-0.17

ikh

-0.16

edes

-0.15

Eig

-0.15

bane

-0.15

riba

-0.15

ëŀ

-0.14

bras

-0.14

 Lennon

-0.14

elig

-0.14

POSITIVE LOGITS

-name

0.27

enburg

0.24

-new

0.24

ishing

0.24

ished

0.22

 name

0.21

name

0.20

 spanking

0.20

/type

0.17

Activations Density 0.034%

references to brand names and their attributes

No Comments

No Known Activations