INDEX

Explanations

numbers and numeric expressions, particularly in academic or formal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 SEDS

-0.62

новниш

-0.60

 unknownFields

-0.60

Jereo

-0.58

ništ

-0.57

ništvo

-0.57

Geplaatst

-0.57

öfe

-0.56

ApiProperty

-0.56

copg

-0.55

POSITIVE LOGITS

年

1.01

년

0.88

 yılında

0.70

 yılı

0.70

年の

0.68

年

0.68

 елның

0.67

年的

0.65

 summer

0.64

년

0.64

Activations Density 0.321%

numbers and numeric expressions, particularly in academic or formal contexts

No Comments

No Known Activations