INDEX

Explanations

programming syntax elements and expressions

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

code variable expansion

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

뀜

-0.43

 Roskov

-0.41

 BitmapFactory

-0.41

ContentAlignment

-0.39

<bos>

-0.39

 Willen

-0.36

tvguidetime

-0.35

 AspNetCore

-0.35

ViewFeatures

-0.34

endregion

-0.34

POSITIVE LOGITS

RTEE

0.57

DockStyle

0.56

RegistryLite

0.53

 &___

0.52

exitRule

0.51

findpost

0.48

RTLI

0.48

GIVEREF

0.48

 postIndex

0.47

INCREF

0.47

Activations Density 0.102%

programming syntax elements and expressions

code variable expansion

No Comments

No Known Activations