INDEX

Explanations

references to businesses and their history

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

following years

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

IntoConstraints

-0.82

⟬

-0.66

:✨

-0.49

Diwedd

-0.49

oredCriteria

-0.49

makeConstraints

-0.46

arashtra

-0.46

 ब्रेकडाउन

-0.46

白石

-0.46

 <>",

-0.46

POSITIVE LOGITS

WebServlet

0.55

ExtendWith

0.51

 defaultstate

0.47

Today

0.47

+#+

0.44

later

0.43

Later

0.42

 later

0.40

 continuer

0.38

today

0.38

Activations Density 0.009%

references to businesses and their history

following years

No Comments

No Known Activations