INDEX

Explanations

variables and mathematical expressions in a programming context

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

pairs of coordinates and variables

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

انتهای

-0.49

'{@

-0.47

 विश्वसनीयता

-0.45

posedge

-0.45

ništ

-0.43

Predecesor

-0.42

 Sangu

-0.42

 ویکی‌پدی

-0.42

SharedCtor

-0.42

getStringExtra

-0.41

POSITIVE LOGITS

 Beide

0.47

DeleteBehavior

0.47

Filmografia

0.46

acamata

0.45

 Bestand

0.44

 both

0.44

 दोनों

0.44

 HasFactory

0.44

 båda

0.43

 entrambi

0.42

Activations Density 0.094%

variables and mathematical expressions in a programming context

pairs of coordinates and variables

No Comments

No Known Activations