INDEX

Explanations

phrases expressing conditionality or limitations

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

Followed by "most"

np_acts-logits-general · gemini-2.0-flash

for the most

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Personendaten

-0.81

TestingModule

-0.76

 viewType

-0.72

 onOptions

-0.71

AlterField

-0.67

 StatelessWidget

-0.67

 %@",

-0.63

 nakalista

-0.61

NameInMap

-0.60

thâu

-0.60

POSITIVE LOGITS

 most

1.78

most

1.64

Most

1.36

 Most

1.29

MOST

1.18

 MOST

1.16

 moft

1.09

 moſt

0.98

最

0.83

 mest

0.81

Activations Density 0.105%

phrases expressing conditionality or limitations

Followed by "most"

for the most

No Comments

No Known Activations