INDEX

Explanations

technical terms and symbols related to data processing or programming languages

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

Acronyms/abbreviations in parentheses

np_acts-logits-general · gemini-2.0-flash

Tuamie producers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 poignée

-0.65

 réputation

-0.65

 poitrine

-0.64

 bibliothèque

-0.64

 leçon

-0.63

 Révolution

-0.62

 amitié

-0.60

 casquette

-0.58

 cathédrale

-0.56

 toiture

-0.56

POSITIVE LOGITS

 Bale

1.03

 Dane

0.91

 Cline

0.91

 Dube

0.90

 Wolfe

0.88

 Hale

0.88

 Tate

0.86

ACE

0.86

MCE

0.85

 Gale

0.84

Activations Density 0.956%

technical terms and symbols related to data processing or programming languages

Acronyms/abbreviations in parentheses

Tuamie producers

No Comments

No Known Activations