INDEX

Explanations

conditions and rules related to variable naming in programming

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

identifier definition

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

 Mail

-0.39

 palette

-0.38

 panels

-0.37

 Palette

-0.37

 mailing

-0.36

トーン

-0.36

 provin

-0.36

 skrzy

-0.35

setFill

-0.35

 Palla

-0.35

POSITIVE LOGITS

rawDesc

0.61

########.

0.59

afficheront

0.56

Atoi

0.54

windowFixed

0.53

LookAnd

0.52

 dAtA

0.52

ViewFeatures

0.51

aarrggbb

0.48

 متعلقه

0.47

Activations Density 0.005%

conditions and rules related to variable naming in programming

identifier definition

No Comments

No Known Activations