INDEX

Explanations

occurrences of implementation classes in the context of software libraries or frameworks

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

implementation test classes

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

transQ

-0.46

 Kugel

-0.44

ambilan

-0.42

 rör

-0.40

 Luar

-0.40

 entdecken

-0.40

 Grüsse

-0.40

Linq

-0.40

var

-0.39

getNumber

-0.39

POSITIVE LOGITS

ServiceImpl

1.44

DaoImpl

1.09

DAOImpl

0.81

mybatisplus

0.79

Impl

0.72

 AssemblyVersion

0.71

featureID

0.69

 Implementation

0.66

oredCriteria

0.65

Implementation

0.64

Activations Density 0.003%

occurrences of implementation classes in the context of software libraries or frameworks

implementation test classes

No Comments

No Known Activations