INDEX

Explanations

references to file types and programming-related content

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

user specific content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 MotionEvent

-0.38

 JMenuItem

-0.36

 pressure

-0.35

 Mexican

-0.35

 frame

-0.35

 float

-0.34

SEK

-0.34

QT

-0.33

HttpFoundation

-0.33

 Venezuelan

-0.33

POSITIVE LOGITS

EFIAPI

0.68



0.65

/*++

0.61

 kasarigan

0.59

 AssemblyCompany

0.59

帖最后由

0.58

 defaultstate

0.56

最快更新

0.52

Samoa

0.52

 Samoa

0.51

Activations Density 0.013%

references to file types and programming-related content

user specific content

No Comments

No Known Activations