INDEX

Explanations

references to quantities and measurements, particularly in a comparative context

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

two children or experts

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

NameInMap

-0.41

 rhestr

-0.36

 dAtA

-0.34

 historical

-0.33

 <=",

-0.32

historical

-0.32

 pomys

-0.31

 ब्रेकडाउन

-0.31

migrationBuilder

-0.31

intahan

-0.29

POSITIVE LOGITS

HideFlags

0.65

 Biôgrafia

0.57

fromnode

0.54

bootstrapcdn

0.53

 IconData

0.52

 disambiguazione

0.52

 EClass

0.51

 continúas

0.50

 Anſ

0.49

cière

0.48

Activations Density 0.172%

references to quantities and measurements, particularly in a comparative context

two children or experts

No Comments

No Known Activations