INDEX

Explanations

dates and numerical information typically associated with legal documents or court cases

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

dates

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

rungsseite

-0.72

المناصب

-0.67

ImageContext

-0.64

<unused28>

-0.62

<unused8>

-0.62

[@BOS@]

-0.62

<unused23>

-0.61

<unused41>

-0.61

<unused16>

-0.61

<unused14>

-0.61

POSITIVE LOGITS

期刊论文

0.33

ostante

0.29

NewGuid

0.26

RefNanny

0.26

 Ники

0.26

 approval

0.25

indd

0.25

gründet

0.25

ineri

0.24

شهاد

0.24

Activations Density 0.004%

dates and numerical information typically associated with legal documents or court cases

dates

No Comments

No Known Activations