INDEX

Explanations

different data types and fields associated with decoding processes in programming

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

field(

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

LookAnd

-0.69

 betweenstory

-0.57

 ویکی‌پدیای

-0.51

UserScript

-0.48

 createState

-0.47

mouseX

-0.47

farwyddwr

-0.47

 onOptions

-0.47

 विश्वसनीयता

-0.47

 jScrollPane

-0.47

POSITIVE LOGITS

Autoritní

0.44

 PeEnEo

0.39

CharField

0.38

ImageField

0.37

 autorytatywna

0.36

NameInMap

0.33

 Numerade

0.33



0.33

Rhestr

0.33

 InputDecoration

0.32

Activations Density 0.005%

different data types and fields associated with decoding processes in programming

field(

No Comments

No Known Activations