INDEX

Explanations

terms related to properties or characteristics of objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 consultato

-0.50

 survi

-0.49

يكب

-0.42

 Survi

-0.40

 SIAM

-0.36

 SQLITE

-0.36

execSQL

-0.35

 proves

-0.35

تقاوى

-0.35

Calcul

-0.35

POSITIVE LOGITS

 prop

2.06

 Property

2.03

Prop

2.00

 Prop

1.98

 Properties

1.97

 property

1.94

prop

1.92

Property

1.90

Properties

1.88

 properties

1.87

Activations Density 0.212%

terms related to properties or characteristics of objects

No Comments

No Known Activations