INDEX

Explanations

programming-related terms and structures

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 tartalomajánló

-0.81

 Мексичка

-0.80

 snippetHide

-0.71

MigrationBuilder

-0.71

 ModelExpression

-0.70

Slf

-0.70

 defaultstate

-0.66

OrBuilder

-0.66

 فريبيس

-0.66

\{\\

-0.65

POSITIVE LOGITS

0.52

0.51

0.50

 Obrador

0.50

Leary

0.48

 Moriarty

0.47

Activations Density 0.394%

programming-related terms and structures

No Comments

No Known Activations