INDEX

Explanations

terms related to scales and measurements in the context of miniature models or figurines

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

scale and size measurements

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

RegressionTest

-0.63

WithIOException

-0.62

 BorderSide

-0.60

zzleHttp

-0.57

 estekak

-0.56

Spoljašnje

-0.55

icitis

-0.51

новништво

-0.51

DotNetBar

-0.50

ArrowToggle

-0.49

POSITIVE LOGITS

 масшта

0.43

scaled

0.41

 scale

0.41

 scaled

0.41

ModelForm

0.40

 render

0.37

 escala

0.37

ModelSerializer

0.37

Scale

0.37

 miniature

0.36

Activations Density 0.004%

terms related to scales and measurements in the context of miniature models or figurines

scale and size measurements

No Comments

No Known Activations