INDEX

Explanations

keywords and phrases associated with data categorization and methodologies

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

statistical significance markers

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

 Chwiliwch

-0.46

pertory

-0.40

 Gesetze

-0.37

Coun

-0.37

Życiorys

-0.36

 solteiro

-0.36

bellum

-0.35

 receives

-0.34

 słucha

-0.34

ligente

-0.33

POSITIVE LOGITS

Fizz

0.97

fizz

0.79

 Fizz

0.75

 fizz

0.56

Попис

0.54

toThrow

0.52

interopRequire

0.52

WriteTagHelper

0.51

ніципа

0.50

 AssemblyVersion

0.49

Activations Density 0.003%

keywords and phrases associated with data categorization and methodologies

statistical significance markers

No Comments

No Known Activations