INDEX

Explanations

terms related to programming and data processing

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

randomized controlled trial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

HasAnnotation

-0.49

 defaultstate

-0.46

angliski

-0.43

 елның

-0.42

ConfigureAwait

-0.42

 tiguan

-0.41

)="

-0.40

 uska

-0.40

Erreferentziak

-0.40

ditto

-0.40

POSITIVE LOGITS

router

0.97

 Router

0.80

Router

0.75

 router

0.73

 useRouter

0.72

󠁢

0.63

 useHistory

0.57

 useNavigate

0.54

 ROUTE

0.54

 ROUT

0.53

Activations Density 0.000%

terms related to programming and data processing

randomized controlled trial

No Comments

No Known Activations

terms related to programming and data processing

randomized controlled trial

No Comments

No Known Activations