INDEX

Explanations

mentions of recent studies or updates

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

"Recent" followed by a period of time

np_acts-logits-general · gemini-2.0-flash

recent time indications

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 eventually

-0.81

eventually

-0.79

 uiteindelijk

-0.76

 later

-0.70

 currently

-0.69

Eventually

-0.66

currently

-0.66

 allerede

-0.65

 redan

-0.65

 actuellement

-0.64

POSITIVE LOGITS

MemoryWarning

0.81

 resourceCulture

0.73

 spate

0.69

 addition

0.68

 additions

0.68

 OFDb

0.64

 BoxFit

0.63

__*/

0.62

addition

0.60

 Vergangenheit

0.60

Activations Density 0.098%

mentions of recent studies or updates

"Recent" followed by a period of time

recent time indications

No Comments

No Known Activations