INDEX

Explanations

terms and concepts related to governmental and organizational policies

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

"policy"

np_acts-logits-general · gemini-2.0-flash

policy targets and reasons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

withCredentials

-0.42

 mystérie

-0.40

The

-0.39

Jetzt

-0.36

 Hors

-0.36

 namorado

-0.35

Sweat

-0.35

具备

-0.35

Weblinks

-0.34

from

-0.34

POSITIVE LOGITS

 POLICY

1.35

 policy

1.34

 Policy

1.31

policy

1.27

Policy

1.24

POLICY

1.18

 Policies

1.11

 policies

1.10

Policies

1.03

 POLICIES

1.02

Activations Density 0.061%

terms and concepts related to governmental and organizational policies

"policy"

policy targets and reasons

No Comments

No Known Activations