INDEX

Explanations

references to boundaries or distinctions, particularly relating to geographical or political lines

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

 kasarigan

-0.72

ReusableCell

-0.64

PerformLayout

-0.52

inaison

-0.51

:✨

-0.50

AsUp

-0.49

CppCodeGen

-0.47

 škoda

-0.46

Manbalar

-0.45

 cherchés

-0.44

POSITIVE LOGITS

 boundary

2.13

 boundaries

2.02

 line

1.87

 border

1.80

 borders

1.76

 Boundaries

1.73

boundary

1.70

 Boundary

1.67

boundaries

1.65

 lines

1.64

Activations Density 0.616%

references to boundaries or distinctions, particularly relating to geographical or political lines

No Comments

No Known Activations