INDEX

Explanations

references to scientific publications and authors

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

citations and academic references

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

 cast

-0.32

 basic

-0.30

 nemmeno

-0.29

ManagedBean

-0.29

 ویکی‌پدی

-0.28

SizeF

-0.28

FactoryBean

-0.28

 turn

-0.28

 asemenea

-0.28

pre

-0.28

POSITIVE LOGITS

NameInMap

0.69

 ब्रेकडाउन

0.66

Datuak

0.60

xico

0.57

AndEndTag

0.56

########.

0.55

 fieldNum

0.55

 PeEnEo

0.54

səhifə

0.53

 Administrativna

0.53

Activations Density 0.153%

references to scientific publications and authors

citations and academic references

No Comments

No Known Activations