INDEX

Explanations

terms related to prisons and imprisonment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Administrativna

-0.42

-0.41

 Croft

-0.37

 Krä

-0.36

 Valk

-0.34

 Chrys

-0.34

 Selb

-0.34

 Sahne

-0.34

 scanning

-0.33

EDEFAULT

-0.33

POSITIVE LOGITS

 prison

0.93

 incarceration

0.89

 imprisonment

0.85

 Prison

0.84

MemoryWarning

0.83

 prisoner

0.83

Prison

0.83

 incarcerated

0.82

囚

0.82

 prisons

0.82

Activations Density 0.338%

terms related to prisons and imprisonment

No Comments

No Known Activations