INDEX

Explanations

programming annotations and documentation comments

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

`@` followed by code keywords

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

awaiter

-0.48

UpInside

-0.39

UserScript

-0.38

 AssemblyTitle

-0.38

ResponseWriter

-0.36

 Biôgrafia

-0.36

CreateModel

-0.36

reportWebVitals

-0.35

 withRouter

-0.34

GEBURTSDATUM

-0.34

POSITIVE LOGITS

 propOrder

0.62

 ComVisible

0.56

 defaultstate

0.52

 esternos

0.49

 كومونز

0.49

 disponibilités

0.48

 nahilalakip

0.48

AccessorTable

0.48

SerializedName

0.47

 ſta

0.47

Activations Density 0.007%

programming annotations and documentation comments

`@` followed by code keywords

No Comments

No Known Activations