INDEX

Explanations

numerical parameters related to scale and transformations in images

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

scale and scaling

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

 protoimpl

-0.54

 للاسماء

-0.53

Vidite

-0.51

RegistryLite

-0.47

UserScript

-0.47

hit

-0.45

 correctes

-0.45

itinéraire

-0.45

tvguidetime

-0.44



-0.43

POSITIVE LOGITS

localScale

0.81

 size

0.51

Scale

0.49

 größe

0.48

FontOfSize

0.47

 scale

0.46

Scaling

0.45

 tamaño

0.45

 taille

0.44

holdet

0.43

Activations Density 0.026%

numerical parameters related to scale and transformations in images

scale and scaling

No Comments

No Known Activations