INDEX

Explanations

technical terms related to application development and frameworks

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

build init start USH

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

وانید

-0.45

windowFixed

-0.43

 betweenstory

-0.43

取締役

-0.42

 سكانية

-0.40

 Rockstar

-0.38

 transférez

-0.38

neous

-0.38

colgroup

-0.38

それから

-0.38

POSITIVE LOGITS

 ExecuteAsync

0.83

 nonatomic

0.82

findpost

0.75

 חיצוניים

0.75

Datuak

0.72

 consultato

0.69

ResumeLayout

0.69

 StatefulWidget

0.68

 mapStateToProps

0.66

 مشين

0.65

Activations Density 0.002%

technical terms related to application development and frameworks

build init start USH

No Comments

No Known Activations