INDEX

Explanations

instances of numerical values and formatted data

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

scientific citations and formatting

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

oredCriteria

-0.59

 noDo

-0.56

@[+][

-0.53

 fört

-0.53

 Italijani

-0.48

хьтан

-0.47

kuuta

-0.44

цездатний

-0.42

 Numerade

-0.42

AnchorTagHelper

-0.42

POSITIVE LOGITS

IntoConstraints

0.57

сылкі

0.41

 EventArgs

0.39

SystemColors

0.39

 Exteriores

0.39

 חיצוניים

0.39

autoconfigure

0.39

DockStyle

0.39

GEBURTSDATUM

0.38

 balans

0.37

Activations Density 0.005%

instances of numerical values and formatted data

scientific citations and formatting

No Comments

No Known Activations