INDEX

Explanations

mentions of government-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Missile

-0.69

 scrat

-0.69

 Dinah

-0.68

 Urqu

-0.67

ergies

-0.67

 estoppel

-0.65

 jitter

-0.64

 trespass

-0.64

orexia

-0.64

 Purdy

-0.63

POSITIVE LOGITS

 Government

2.27

 government

2.13

Government

2.07

government

1.86

 GOVERNMENT

1.80

GOVERNMENT

1.71

 Govt

1.65

 governments

1.60

 govt

1.55

 Governments

1.55

Activations Density 0.056%

mentions of government-related terms

No Comments

No Known Activations