INDEX

Explanations

words related to permissions and legal concepts

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

permit and allow

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

rrggbb

-0.41

 jurado

-0.38

onAttach

-0.38

 mandiri

-0.35

 Präsidenten

-0.35

 pengantin

-0.35

 presidencia

-0.34

 venido

-0.33

 eivät

-0.33

 reputación

-0.32

POSITIVE LOGITS

 Allow

1.23

allow

1.20

 allowed

1.18

 Permit

1.16

 permit

1.15

 allow

1.14

 Allowed

1.11

 permis

1.10

 permitted

1.09

permit

1.09

Activations Density 0.024%

words related to permissions and legal concepts

permit and allow

No Comments

No Known Activations