INDEX

Explanations

code comments and structure in programming documents

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

code comments and descriptions

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

dafx

-0.60

 kaynağından

-0.57

tagHelperRunner

-0.57

 Waſſer

-0.54

 typelib

-0.54

postsleuth

-0.54

elemField

-0.54

 geſch

-0.53

featureID

-0.52

 Geiſt

-0.52

POSITIVE LOGITS

 nahilalakip

0.35

 aims

0.34

0.32

lik

0.31

#!/

0.31

 describes

0.30

nehm

0.30

ながら

0.30

 this

0.29

исленность

0.29

Activations Density 0.011%

code comments and structure in programming documents

code comments and descriptions

No Comments

No Known Activations