INDEX

Explanations

phrases and terms related to "cleaning up" or "removal."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 StatefulWidget

-0.54

новниш

-0.53

PreferredItem

-0.51

 surla

-0.50

AndroidJUnit

-0.44

 Biôgrafia

-0.44

 ivelany

-0.43



-0.41

 raiſ

-0.41

ValueGenerated

-0.41

POSITIVE LOGITS

 cleanup

1.13

 Cleanup

1.03

cleanup

1.01

Cleanup

0.95

清理

0.61

 nettoyage

0.52

 messes

0.51

 manipulation

0.49

Avery

0.47

修正

0.46

Activations Density 0.397%

phrases and terms related to "cleaning up" or "removal."

No Comments

No Known Activations