INDEX

Explanations

disciplinary titles and affiliations in the context of research or expert commentary

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

roles followed by of

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

 мәкал

-0.76

 ویکی‌پدی

-0.61

httphttps

-0.59

GEBURTSDATUM

-0.59

yntaxException

-0.57

 autorytatywna

-0.57

UrlResolution

-0.55

 kasarigan

-0.52

 queſta

-0.52

帖最后由

-0.52

POSITIVE LOGITS

 ebenfalls

0.41

 similarly

0.38

 επίσης

0.38

 autre

0.37

 Runde

0.37

autre

0.36

occhio

0.35

Ayrıca

0.34

CHAFT

0.34

 another

0.33

Activations Density 0.004%

disciplinary titles and affiliations in the context of research or expert commentary

roles followed by of

No Comments

No Known Activations