INDEX

Explanations

phrases indicating comparisons and contrasts in data analysis

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

people and their preferences

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

 Italijanski

-0.43

 Biôgrafia

-0.38

enumii

-0.37

urangi

-0.34

 Garnish

-0.32

 Fixture

-0.30

afficheront

-0.30

𝒯

-0.30

drip

-0.30

itinéraire

-0.29

POSITIVE LOGITS

 препратки

0.53

出版年

0.52

openConnection

0.52

 communiquez

0.51

новниш

0.50

djangoproject

0.49

جستارهای

0.49

 AssemblyTitle

0.48

DockStyle

0.48

ThroughAttribute

0.47

Activations Density 0.069%

phrases indicating comparisons and contrasts in data analysis

people and their preferences

No Comments

No Known Activations