INDEX

Explanations

elements related to legal proceedings and courtroom discussions

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

Follows "court", "I", "we", or "been"

np_acts-logits-general · gemini-2.0-flash

court has heard

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tagHelperRunner

-0.72

JspWriter

-0.70

IonicModule

-0.67

WebVitals

-0.56

حياته

-0.55

 AssemblyTitle

-0.54

EnableWeb

-0.54

usitis

-0.54

✨:

-0.52

ConverterFactory

-0.52

POSITIVE LOGITS

 heard

1.20

 hear

1.19

 hears

1.19

 told

1.03

heard

0.99

 hearing

0.91

 Hear

0.91

听说

0.91

hear

0.90

 HEAR

0.90

Activations Density 0.171%

elements related to legal proceedings and courtroom discussions

Follows "court", "I", "we", or "been"

court has heard

No Comments

No Known Activations