INDEX

Explanations

terms related to programming concepts and code structure in various contexts

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

user interface elements

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

 propOrder

-0.78

IntoConstraints

-0.77

ロウィン

-0.75

principalColumn

-0.75

Personendaten

-0.74

 disambiguazione

-0.73

BufferException

-0.72

 contextLoads

-0.69

харш

-0.69

 فريبيس

-0.68

POSITIVE LOGITS

хьтан

0.31

 époux

0.27

 tokio

0.26

 Singh

0.26

btnBack

0.24

ագրություններ

0.24

onPressed

0.24

colorPrimary

0.23

дре

0.23

ENODEV

0.23

Activations Density 0.208%

terms related to programming concepts and code structure in various contexts

user interface elements

No Comments

No Known Activations