INDEX

Explanations

instances of the number '3'

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

number three

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

SharedDtor

-0.70

writeFieldEnd

-0.67

 AssemblyCulture

-0.61

bcryptjs

-0.60



-0.59

 Wicidata

-0.59

abestanden

-0.59

parsedMessage

-0.58

AsUp

-0.58

 Fazit

-0.57

POSITIVE LOGITS

two

0.97

 three

0.86

 four

0.73

 multiple

0.70

 five

0.67

TWO

0.67

two

0.65

 different

0.65

Two

0.65

 several

0.62

Activations Density 0.134%

instances of the number '3'

number three

No Comments

No Known Activations