INDEX

Explanations

programming and data structure terminology

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

code and data formatting

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

Бахар

-0.57

extAlignment

-0.56

CloseOperation

-0.54

脚注の使い方

-0.52

 EconPapers

-0.49

bootstrapcdn

-0.46

WindowConstants

-0.44

ReusableCell

-0.43

 مشين

-0.42

RTHOOK

-0.41

POSITIVE LOGITS

 Dete

0.44

KommentareTeilen

0.41

Dete

0.40

///</

0.39

 recognized

0.39

 appearances

0.39

ública

0.38

 recognizable

0.38

 signature

0.38

SpringRunner

0.38

Activations Density 0.350%

programming and data structure terminology

code and data formatting

No Comments

No Known Activations