INDEX

Explanations

programming-related code constructs and commands

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

algebraic expressions

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

citep

-0.40

 Verſ

-0.39

 provas

-0.38

WebServlet

-0.38

قایناقلار

-0.37

 études

-0.36

Дереккөздер

-0.36

სქოლიო

-0.35

xase

-0.35

 Kapol

-0.35

POSITIVE LOGITS

ंदीखरीदारी

0.50

 TextInputType

0.49

 tartalomajánló

0.49

IFA

0.44

enumi

0.42

 Paglinawan

0.42

VIAF

0.42

 Exactos

0.41

 gainera

0.40

ashqai

0.40

Activations Density 0.080%

programming-related code constructs and commands

algebraic expressions

No Comments

No Known Activations