INDEX

Explanations

elements related to HTML or JSX structure in code

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

user dialogue or code structure

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

лтамалар

-0.52

󠁮

-0.44

 intptr

-0.44

Personendaten

-0.44

 IndexPath

-0.42

RTGC

-0.42

 disambiguazione

-0.42

 nakalista

-0.41

rungsseite

-0.39

 Samir

-0.39

POSITIVE LOGITS

CardContent

0.51

 betweenstory

0.46

 InputDecoration

0.46

FontOfSize

0.45

DIPSETTING

0.42

DispatchToProps

0.41

исленность

0.40

InstrumentedTest

0.38

 AsyncTask

0.38

addCriterion

0.37

Activations Density 0.002%

elements related to HTML or JSX structure in code

user dialogue or code structure

No Comments

No Known Activations