INDEX

Explanations

structural elements and symbols used in formal or technical writing

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

analogy and similarity

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

 فريبيس

-0.82

 المعيارى

-0.68

 queſta

-0.63

⟬

-0.63

WebVitals

-0.62

 незавершена

-0.61

niſſe

-0.60

parsedMessage

-0.60

 gynhyrchwyd

-0.60

*/;

-0.60

POSITIVE LOGITS

Similar

0.48

 Similar

0.44

 similar

0.42

 Ähn

0.41

 аналоги

0.39

 ähn

0.39

 similarly

0.39

 podob

0.38

 semelhante

0.38

同样

0.36

Activations Density 0.013%

structural elements and symbols used in formal or technical writing

analogy and similarity

No Comments

No Known Activations