INDEX

Explanations

programming-related syntax and keywords

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

 CreateTagHelper

-0.87

tagext

-0.72

rungsseite

-0.71

Personendaten

-0.69

errHandler

-0.67

RegressionTest

-0.65

 noDo

-0.63

 surla

-0.62

niſſe

-0.62

 iNdEx

-0.61

POSITIVE LOGITS

piecze

0.34

 psic

0.31

 nous

0.27

···

0.26

 redé

0.25

 feuille

0.24

 masas

0.23

Lähde

0.22

pé

0.22

Wymiary

0.22

Activations Density 0.003%

programming-related syntax and keywords

No Comments

No Known Activations