INDEX

Explanations

terms related to legal proceedings and evidence

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

legal and technical terms

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

 Ripley

-0.28

逆

-0.28

Miz

-0.27

なくても

-0.27

 sumpay

-0.27

oz

-0.27

mig

-0.26

 Miss

-0.26

 remis

-0.26

rem

-0.26

POSITIVE LOGITS

 nakalista

0.69

 oprot

0.66

⟬

0.66

expandindo

0.64

parsedMessage

0.58

 المعيارى

0.56

WriteTagHelper

0.55

 ویکی‌پدیا

0.54

:✨

0.53

::_('

0.53

Activations Density 0.218%

terms related to legal proceedings and evidence

legal and technical terms

No Comments

No Known Activations