INDEX

Explanations

references to sections, definitions, propositions, and theorems in technical documents

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

Reference to sections or theorems

np_acts-logits-general · gemini-2.0-flash

section and proposition numbers

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

PullParser

-0.52

complexContent

-0.51

tikzpicture

-0.50

 useDispatch

-0.50

DAD

-0.47

 száz

-0.46

 cérami

-0.46

 poffe

-0.46

 ſtate

-0.44

dad

-0.44

POSITIVE LOGITS

 EconPapers

0.62

twimg

0.60

 kasarigan

0.58

ToProps

0.56

 esternos

0.56

liosis

0.55

 pinulongan

0.54

acabana

0.52

ronpa

0.52

SupportActionBar

0.51

Activations Density 1.089%

references to sections, definitions, propositions, and theorems in technical documents

Reference to sections or theorems

section and proposition numbers

No Comments

No Known Activations