INDEX

Explanations

words related to political policies and their implications

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

policy priorities and positions

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

Reprodução

-0.50

 undangan

-0.44

 bailarina

-0.41

PreExecute

-0.41

 Betrach

-0.39

 lembran

-0.39

Assista

-0.39

 Paglinawan

-0.39

 Frankrijk

-0.39

 dawned

-0.39

POSITIVE LOGITS

 policies

0.68

 kaarangay

0.65

 Policies

0.62

policies

0.60

 policy

0.60

政策

0.55

policy

0.55

Policies

0.55

Personendaten

0.53

UserScript

0.49

Activations Density 0.014%

words related to political policies and their implications

policy priorities and positions

No Comments

No Known Activations