INDEX

Explanations

terms related to programming, enumeration, and error handling in code

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

interface types enum

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

featureID

-0.83

GTCX

-0.71

Personendaten

-0.66

parsedMessage

-0.65

PropertyChanging

-0.64

 ProtoMessage

-0.62

MLLoader

-0.62

 mergeFrom

-0.61

 protoimpl

-0.59

NameInMap

-0.59

POSITIVE LOGITS

urlpatterns

0.45

代表

0.39

 reprez

0.33

 саны

0.33

 Proctor

0.31

녀

0.30

QUADS

0.30

isValid

0.30

 categor

0.29

 representa

0.29

Activations Density 0.098%

terms related to programming, enumeration, and error handling in code

interface types enum

No Comments

No Known Activations