INDEX

Explanations

block comments and documentation in programming code

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åt

-0.35

 NUKAT

-0.31

niająca

-0.30

 muligt

-0.30

 eind

-0.30

FieldDescriptor

-0.29

 effort

-0.29

Zacks

-0.29

AutoField

-0.28

 Bíblia

-0.28

POSITIVE LOGITS

DockStyle

0.76

Хьажоргаш

0.73

__*/

0.71

GOTREF

0.67

queryInterface

0.66

principalTable

0.64

Datuak

0.63

 lenker

0.63

expandindo

0.62

MigrationBuilder

0.61

Activations Density 0.003%

block comments and documentation in programming code

No Comments

No Known Activations