INDEX

Explanations

class definitions and structural keywords in programming code

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

class definition context

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

 autorytatywna

-1.01

 '\\;'

-0.98

 فريبيس

-0.95

GEBURTS

-0.88

aarrggbb

-0.88

 kasarigan

-0.84

ViewImports

-0.84

principalColumn

-0.84

تقاوى

-0.83

styleType

-0.83

POSITIVE LOGITS

 Erben

0.28

 epä

0.25

 quæ

0.25

abaikan

0.25

 Jehová

0.24

<h3>

0.23

 arvio

0.23

 épaules

0.23

super

0.23

 nada

0.23

Activations Density 0.003%

class definitions and structural keywords in programming code

class definition context

No Comments

No Known Activations