INDEX

Explanations

specific references to formal documents or academic language

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

start of text or phrase

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

 للمعارف

-0.53

 Infórmanos

-0.51

帖最后由

-0.47

 Autorizaciones

-0.45

 gyhoeddwyd

-0.44

 ModelExpression

-0.42

AddWithValue

-0.41

󠁣

-0.41

 Shovel

-0.40

 bacio

-0.40

POSITIVE LOGITS

SpringRunner

0.46

ViewFeatures

0.45

WebServlet

0.42

 arşivlendi

0.42

expandindo

0.39

følge

0.39

verwijspagina

0.36

 Inſ

0.36

 ويكيپيديا

0.36

OrNil

0.36

Activations Density 0.040%

specific references to formal documents or academic language

start of text or phrase

No Comments

No Known Activations