INDEX

Explanations

programming-related terms and structures

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

forms, mk, lens

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

Tikang

-0.62

ExtendWith

-0.58

onedDateTime

-0.55

AddTagHelper

-0.54

UnitTesting

-0.50

abstractmethod

-0.48

WebMethod

-0.47

MLLoader

-0.47

RequiresApi

-0.47

ftance

-0.47

POSITIVE LOGITS

 FormGroup

0.88

IntoConstraints

0.52

oredCriteria

0.49

InputTagHelper

0.46

itinéraire

0.44

anmoins

0.41

Personendaten

0.39

ModelForm

0.37

RuleContext

0.36

 informée

0.36

Activations Density 0.000%

programming-related terms and structures

forms, mk, lens

No Comments

No Known Activations