INDEX

Explanations

comment blocks and documentation parts in code or technical writing

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

Code comments or license headers

np_acts-logits-general · gemini-2.0-flash

programming identifiers and code libraries

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

larımız

-0.59

“

-0.59

−

-0.54

way

-0.51

 Curran

-0.50

Straße

-0.50

-0.49

 Neff

-0.49

Sure

-0.48

-0.47

POSITIVE LOGITS

****************

1.11

 nahilalakip

0.80

:+:

0.77

VersionUID

0.73

 Мексичка

0.73

NSNotification

0.67

***************/

0.66

////////////////

0.66

enumi

0.65

ActionCreators

0.64

Activations Density 0.169%

comment blocks and documentation parts in code or technical writing

Code comments or license headers

programming identifiers and code libraries

No Comments

No Known Activations