INDEX

Explanations

terms related to programming libraries and frameworks

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

code parsing and identifiers

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

 EconPapers

-0.71

IUrlHelper

-0.59

 kaarangay

-0.53

GEBURTS

-0.49

 plufieurs

-0.48

SharedDtor

-0.48

ResponseWriter

-0.47

+#+

-0.46

writeFieldEnd

-0.46

ContentAsync

-0.45

POSITIVE LOGITS

 autorytatywna

0.47

ModelSerializer

0.42

 HttpNotFound

0.38

جغرافيا

0.36

 CreateTagHelper

0.35

 CallOverrides

0.35

urysty

0.35

 ویکی‌پدیا

0.35

Carrière

0.33

brück

0.32

Activations Density 0.005%

terms related to programming libraries and frameworks

code parsing and identifiers

No Comments

No Known Activations