INDEX

Explanations

annotations and attributes in code

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

code attributes and annotations

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

Aid

-0.28

aid

-0.28

ATIO

-0.27

atio

-0.26

 owner

-0.26

 nemlig

-0.26

 later

-0.25

FetchType

-0.25

 CONS

-0.25

 action

-0.25

POSITIVE LOGITS

 BoxDecoration

0.80

 autorytatywna

0.71

RectangleBorder

0.63

يكب

0.63

 ComVisible

0.62

DataAnnotations

0.61

XmlAccessorType

0.59

RunWith

0.58

AutoScaleMode

0.58

ExtendWith

0.57

Activations Density 0.007%

annotations and attributes in code

code attributes and annotations

No Comments

No Known Activations