INDEX

Explanations

references to drinking beverages, specifically alcohol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 AssemblyCulture

-1.02

 Weldon

-0.76

modelBuilder

-0.73

CDATA

-0.70

Personendaten

-0.70

ConstraintMaker

-0.69

]")]

-0.68

 Aces

-0.67

 blé

-0.67

 Seeder

-0.67

POSITIVE LOGITS

 drinking

2.90

 drink

2.87

 Drink

2.69

 DRINK

2.60

Drink

2.59

drink

2.56

 Drinking

2.56

drinking

2.48

Drinking

2.41

 drank

2.34

Activations Density 0.061%

references to drinking beverages, specifically alcohol

No Comments

No Known Activations