INDEX

Explanations

programming code structures and definitions

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

programming literals and keywords

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

Vidite

-0.55

MLLoader

-0.53

 EconPapers

-0.46

 nahilalakip

-0.41

HasIndex

-0.40

DialogContent

-0.40

 coincide

-0.40

 autorytatywna

-0.40

сылкі

-0.40

 Référence

-0.39

POSITIVE LOGITS

VersionUID

0.47

msgSender

0.46

 MainAxisSize

0.45

protoimpl

0.45

 FormsModule

0.42

saraba

0.41

fjspx

0.41

RotationOrder

0.41

thenReturn

0.40

textTheme

0.40

Activations Density 0.046%

programming code structures and definitions

programming literals and keywords

No Comments

No Known Activations