INDEX

Explanations

code-related concepts and programming structures

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

domain-specific technical jargon and identifier-like terms, especially in code, database schemas, and other formal documentation contexts.

oai_token-act-pair · gpt-5 Triggered by @arielgoldsteinlab

code snippets

np_max-act-logits · gpt-5 Triggered by @arielgoldsteinlab

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 ویکی‌پدی

-1.03

EndGlobalSection

-0.78

 noDo

-0.78

ьаж

-0.74

IndentedString

-0.69

AutoScaleMode

-0.68

ंदीखरीदारी

-0.68

msgTypes

-0.67

Personensuche

-0.65

featureID

-0.62

POSITIVE LOGITS

extAlignment

0.59

 surla

0.52

 rhestr

0.48

󠁢

0.47



0.46

 fortæ

0.45

 Вікіпе

0.44

 Bioaccumulative

0.43

ClientRect

0.42

fastjson

0.41

Activations Density 42.178%

code-related concepts and programming structures

domain-specific technical jargon and identifier-like terms, especially in code, database schemas, and other formal documentation contexts.

code snippets

No Comments

No Known Activations

code-related concepts and programming structures

domain-specific technical jargon and identifier-like terms, especially in code, database schemas, and other formal documentation contexts.

code snippets

No Comments

No Known Activations