INDEX

Explanations

references to evidence in a judicial context

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

cited unpublished dispositions

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

 Normdatei

-0.63

BeginContext

-0.56

 AssemblyTitle

-0.51

发表于

-0.51

Diweddarwch

-0.50

RectangleBorder

-0.48

帖最后由

-0.48

 AssemblyCompany

-0.47

сылкі

-0.47

 SIMBAD

-0.47

POSITIVE LOGITS

 incorporation

0.45

 incorporate

0.40

 incorporar

0.40

 incorporates

0.37

Jereo

0.36

 INCORPOR

0.36

 incorporating

0.36

 assimilation

0.35

 incorporated

0.35

 shots

0.34

Activations Density 0.206%

references to evidence in a judicial context

cited unpublished dispositions

No Comments

No Known Activations