INDEX

Explanations

references to object-oriented programming elements such as objects, methods, and expressions

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

programming related terms

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

OCCURRED

-0.61

OGND

-0.60

 esternos

-0.59

 PeEnEo

-0.57

 BoxFit

-0.56

WriteTagHelper

-0.53

aarrggbb

-0.52

FieldBuilder

-0.52

Билгалдахарш

-0.52

AddHtmlAttribute

-0.50

POSITIVE LOGITS

styleable

0.42

 betweenstory

0.40

 cuyo

0.35

 diseño

0.35

feitura

0.33

 acusado

0.32

 diseñ

0.32

 cujo

0.32

 فا

0.31

RetentionPolicy

0.31

Activations Density 0.045%

references to object-oriented programming elements such as objects, methods, and expressions

programming related terms

No Comments

No Known Activations