INDEX

Explanations

references to database collections and operations within coding contexts

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

database queries

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

 يتيمه

-0.94

 MainAxisSize

-0.67

 للمعارف

-0.66

fjspx

-0.65

 defaultstate

-0.63

IContainer

-0.63

xase

-0.63

onAttach

-0.63

 Meksiku

-0.61

 ब्रेकडाउन

-0.60

POSITIVE LOGITS

WritableDatabase

0.37

 DataLoader

0.34

 spiaggia

0.34

ophagus

0.32

 レンズ

0.32

 UserDao

0.32

userDao

0.31

 praia

0.31

UserDao

0.30

cel

0.29

Activations Density 0.004%

references to database collections and operations within coding contexts

database queries

No Comments

No Known Activations