INDEX

Explanations

keywords associated with object-oriented programming concepts and structure

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

programming class names

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

 nahilalakip

-0.91

Personendaten

-0.90

 мәкал

-0.87

évaluateur

-0.85

MLLoader

-0.76

interopRequire

-0.73

ьаж

-0.71

+#+

-0.68

хьтан

-0.68

 kasarigan

-0.66

POSITIVE LOGITS

cektir

0.34

 casera

0.32

ึ่ง

0.32

 plezier

0.32

føring

0.32

 similares

0.31

 médicament

0.30

 étudiant

0.30

 orgullo

0.30

 ouvriers

0.29

Activations Density 0.007%

keywords associated with object-oriented programming concepts and structure

programming class names

No Comments

No Known Activations