INDEX

Explanations

text from scientific or governmental documents that contain technical or legal terms.

oai_token-act-pair · gemini-2.0-flash

"direct" related words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+#

-0.94

 nahilalakip

-0.88

PerformLayout

-0.83

WebElementEntity

-0.78

SourceChecksum

-0.78

featureID

-0.77

XmlEnum

-0.75

AutoScaleMode

-0.74

$_['

-0.73

 NSCoder

-0.72

POSITIVE LOGITS

set

0.50

 Direkt

0.49

 directo

0.49

 Direct

0.48

dis

0.45

Direct

0.44

 Farm

0.43

 direct

0.43

 payments

0.43

 farm

0.42

Activations Density 1.209%

text from scientific or governmental documents that contain technical or legal terms.

"direct" related words

No Comments

No Known Activations