INDEX

Explanations

programming-related terms and data structure operations

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

set operations

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

 arreglos

-0.41

 cuadros

-0.41

 düz

-0.40

ReusableCell

-0.40

 Instruction

-0.39

 sürü

-0.39

 ويكيميديا

-0.39

 aprilie

-0.39

 Instructions

-0.39

 çekilen

-0.39

POSITIVE LOGITS

HashSet

0.66

 HashSet

0.64

WaitGroup

0.58

 TreeSet

0.56

 isSet

0.53

Gweler

0.52

stdc

0.50

Географиясе

0.50

WebServlet

0.49

TreeSet

0.49

Activations Density 0.023%

programming-related terms and data structure operations

set operations

No Comments

No Known Activations