INDEX

Explanations

references to violence in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

fjspx

-0.67

ContentAlignment

-0.66

 مرئيه

-0.59

брь

-0.56

 pocz

-0.54

NOPQRST

-0.52

Heu

-0.52

wą

-0.52

 Revival

-0.52

tre

-0.51

POSITIVE LOGITS

 violence

2.45

 Violence

2.22

violence

2.18

Violence

2.13

 violent

2.07

violent

1.94

Violent

1.90

 Violent

1.83

 violencia

1.74

 violen

1.71

Activations Density 0.127%

references to violence in various contexts

No Comments

No Known Activations