INDEX

Explanations

references to note-taking and documentation processes

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

Taking or scribbling notes

np_acts-logits-general · gemini-2.0-flash

recording notes and details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Roskov

-0.74

 betweenstory

-0.72

Демографія

-0.65

argout

-0.65

antMatchers

-0.64

StructEnd

-0.62

 Baillargeon

-0.58

couvrir

-0.55

InputBorder

-0.55

buttonBar

-0.55

POSITIVE LOGITS

 notes

2.02

 Notes

1.70

 note

1.70

notes

1.62

Notes

1.54

jot

1.46

 NOTES

1.41

note

1.36

 notebook

1.36

 Note

1.31

Activations Density 0.324%

references to note-taking and documentation processes

Taking or scribbling notes

recording notes and details

No Comments

No Known Activations