INDEX

Explanations

mathematical symbols and notations related to tensors and functions

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

IonicModule

-0.65

 ujednoznacz

-0.59

 AssemblyTitle

-0.56

rungsseite

-0.56

DropTable

-0.55

RenderAtEndOf

-0.55

IntentFilter

-0.54

blad

-0.53

ISupport

-0.52

IVEREF

-0.52

POSITIVE LOGITS

 point

1.78

 points

1.67

point

1.52

points

1.49

 Point

1.47

Point

1.42

 Points

1.40

 POINT

1.36

Points

1.30

 POINTS

1.27

Activations Density 1.160%

mathematical symbols and notations related to tensors and functions

No Comments

No Known Activations