INDEX

Explanations

mathematical terms related to matrix projections and their properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

SourceChecksum

-0.76

Dra

-0.54

 Theſe

-0.53

FTFY

-0.52

 Plural

-0.52

*/),

-0.51

 revision

-0.51

duty

-0.51

cibility

-0.51

 dragging

-0.50

POSITIVE LOGITS

 matrix

0.90

 matrices

0.84

 Matrix

0.77

matrix

0.70

Matrix

0.70

 matrice

0.69

 matriz

0.69

 Matrices

0.67

 InputDecoration

0.66

矩阵

0.65

Activations Density 1.754%

mathematical terms related to matrix projections and their properties

No Comments

No Known Activations