INDEX

Explanations

mathematical expressions involving negative numbers

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

previous term is 1

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

aarrggbb

-0.45

исленность

-0.44

 Мексичка

-0.41

makeConstraints

-0.40

httphttps

-0.39

LayoutConstraint

-0.39

 GlobalKey

-0.38

تفصیلات

-0.38

 toek

-0.37

 trwa

-0.36

POSITIVE LOGITS

previous

0.64

Previous

0.62

 précédents

0.61

 previous

0.57

 précédente

0.55

 précédent

0.54

 للمعارف

0.52

 Previous

0.52

RTSN

0.52

 PREVIOUS

0.51

Activations Density 0.020%

mathematical expressions involving negative numbers

previous term is 1

No Comments

No Known Activations