INDEX

Explanations

references to branding and marketing strategies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ÏĦÎµ

-0.16

rane

-0.15

IMIT

-0.14

vaz

-0.14

Gad

-0.14

æıĲ

-0.13

.quest

-0.13

imit

-0.13

bilder

-0.13

_backward

-0.13

POSITIVE LOGITS

 collateral

0.27

 identity

0.26

_identity

0.23

 Identity

0.23

identity

0.22

 assets

0.21

Identity

0.21

 brand

0.20

 hero

0.19

 direction

0.19

Activations Density 0.118%

references to branding and marketing strategies

No Comments

No Known Activations