INDEX

Explanations

occurrences of the word "many" and similar quantifying terms

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

quantifier and time duration

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

Tembelea

-0.65

Datuak

-0.57

 nakalista

-0.47

IGraphics

-0.46

GEBURTSDATUM

-0.45

 ujednoznacz

-0.45

 Administrativna

-0.44

TestBed

-0.44

 propOrder

-0.43

SharedDtor

-0.43

POSITIVE LOGITS

 years

0.66

 decades

0.63

 años

0.53

 YEARS

0.52

 jahre

0.52

years

0.51

 jaren

0.50

 dłu

0.48

decade

0.48

 Años

0.48

Activations Density 0.021%

occurrences of the word "many" and similar quantifying terms

quantifier and time duration

No Comments

No Known Activations