INDEX

Explanations

terms related to configuration or setup in programming or software environments

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

user input indicator

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

 <=",

-0.54

ReusableCell

-0.49

+#+#

-0.45

 surla

-0.44

 campeones

-0.44

 Verständ

-0.43

kuuta

-0.43

 "..\..\

-0.43

 Karlsson

-0.41

 navideño

-0.41

POSITIVE LOGITS

CppMethod

0.56

BindingSource

0.45

 TestBed

0.45

provide

0.43

 Provide

0.43

INCREF

0.43

DockStyle

0.41

Provide

0.41

 NgModule

0.40

QPalette

0.40

Activations Density 0.001%

terms related to configuration or setup in programming or software environments

user input indicator

No Comments

No Known Activations