INDEX

Explanations

statements regarding political actions and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 prévention

-0.68

 migli

-0.59

 civilisation

-0.57

 insegn

-0.56

 éduc

-0.54

 innamor

-0.53

runOnUiThread

-0.52

 secoli

-0.52

 wonderful

-0.51

 poliester

-0.51

POSITIVE LOGITS

WithIOException

0.82

 political

0.75

NameInMap

0.73

political

0.71

 politically

0.69

 POLITICAL

0.68

 undercut

0.67

 kritik

0.67

 tacit

0.67

 Political

0.66

Activations Density 0.594%

statements regarding political actions and their implications

No Comments

No Known Activations