INDEX

Explanations

references to publication details and citations

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

before journal publishers

np_acts-logits-general · gemini-2.0-flash

scientific publishers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 AttributeSet

-0.50

ственных

-0.47

phylla

-0.42

ाप्त

-0.41

 Roughly

-0.41

 grads

-0.40

And

-0.40

ШЕ

-0.40

nbt

-0.39

fin

-0.39

POSITIVE LOGITS

 تضيفلها

0.80

 BoxDecoration

0.78

MemoryWarning

0.75

 lenker

0.73

Datuak

0.70

المكان

0.70

Viitteet

0.70

CPtr

0.70

CloseOperation

0.69

kloped

0.69

Activations Density 0.585%

references to publication details and citations

before journal publishers

scientific publishers

No Comments

No Known Activations