INDEX

Explanations

terms related to administrative functions and roles in a system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dintr

-0.37

asas

-0.34

 Pende

-0.33

 natin

-0.33

botella

-0.32

 iemand

-0.31

Saludos

-0.31

chaqueta

-0.31

])))

-0.31

 Gulf

-0.30

POSITIVE LOGITS

 Admin

1.23

Admin

1.21

 admin

1.16

admin

1.15

ADMIN

1.00

 ADMIN

0.99

 admins

0.90

ModelAdmin

0.85

 Administ

0.79

Admins

0.79

Activations Density 0.163%

terms related to administrative functions and roles in a system

No Comments

No Known Activations