INDEX

Explanations

references to juries and jury-related concepts

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

jury verdict conviction trial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 propOrder

-0.57

 nahilalakip

-0.53

بوابة

-0.53

 MonoBehaviour

-0.50

Enllaces

-0.42

Felipe

-0.42

 bezit

-0.40

 kasarigan

-0.40

adpleegd

-0.39

 desplaz

-0.39

POSITIVE LOGITS

 Jury

1.35

Jury

1.31

 jury

1.31

jury

1.26

 juries

1.11

 jurors

0.99

 juror

0.87

 jurado

0.76

Jur

0.73

Jur

0.71

Activations Density 0.027%

references to juries and jury-related concepts

jury verdict conviction trial

No Comments

No Known Activations