INDEX

Explanations

references to innovative ideas and creative expressions in diverse contexts

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

Innovation and novelty

np_acts-logits-general · gemini-2.0-flash

new and unconventional ideas

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ModelExpression

-0.66

AutoScaleMode

-0.65

setVerticalGroup

-0.64

Autoritní

-0.61

IndentedString

-0.59

nonatomic

-0.58

美子

-0.58

#+#

-0.57

ConstraintMaker

-0.55

Vidite

-0.55

POSITIVE LOGITS

 innovative

1.86

 innov

1.72

 unconventional

1.72

 innovation

1.70

 innovate

1.66

innovative

1.65

Innovative

1.58

 unusual

1.55

 novelty

1.53

 Innovative

1.49

Activations Density 0.442%

references to innovative ideas and creative expressions in diverse contexts

Innovation and novelty

new and unconventional ideas

No Comments

No Known Activations