INDEX

Explanations

terms related to political contexts and issues

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

RenderAtEndOf

-0.82

 Jefus

-0.81

 fevere

-0.80

 Endlich

-0.76

ValueStyle

-0.76

 Verr

-0.74

 Intervals

-0.72

 faſt

-0.71

 Majefty

-0.70

 ordinaires

-0.70

POSITIVE LOGITS

 political

1.86

 Political

1.70

political

1.62

Political

1.61

 POLITICAL

1.57

 politics

1.49

 Politics

1.38

 politically

1.33

 politic

1.33

Politics

1.30

Activations Density 0.055%

terms related to political contexts and issues

No Comments

No Known Activations