INDEX

Explanations

elements related to personal identification

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

names and countries

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

SuppressLint

-0.51

MessageOf

-0.40

 Sermons

-0.39

 addSubview

-0.39

brtc

-0.39

 sermons

-0.39

📷

-0.39

 TextFormField

-0.38

TextAlign

-0.38

+#+#

-0.38

POSITIVE LOGITS

 Roskov

0.62

RefNanny

0.60

ScopeManager

0.56

BASEPATH

0.50

ItemLayout

0.49

 disambiguazione

0.47

 vuestros

0.46

참고

0.46

msgSender

0.44

 województwie

0.44

Activations Density 0.000%

elements related to personal identification

names and countries

No Comments

No Known Activations