INDEX

Explanations

code constructs and JavaScript programming concepts

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

aarrggbb

-0.83

 виправивши

-0.78

homonymie

-0.77

expandindo

-0.70

 HasFactory

-0.70

 saites

-0.67

TestingModule

-0.65

AutoScaleMode

-0.63

 utafitiHapana

-0.63

 Wiktionnaire

-0.62

POSITIVE LOGITS

 Object

1.23

Object

1.22

delete

0.80

typeof

0.65

 object

0.64

OBJECT

0.62

 delete

0.61

 OBJECT

0.59

bject

0.57

 typeof

0.56

Activations Density 0.093%

code constructs and JavaScript programming concepts

No Comments

No Known Activations