INDEX

Explanations

identifiers and keywords commonly used in programming or data 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-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

Попис

-0.62

 محفوظة

-0.54

 lenker

-0.52

 arşivlendi

-0.52

SocketAddress

-0.52

幸いです

-0.50

 виправивши

-0.49

 kaarangay

-0.49

 kasarigan

-0.49

 HFILL

-0.49

POSITIVE LOGITS

Tikang

0.46

XmlAccessType

0.38

.*")]

0.37

Recall

0.36

模

0.36

 الحره

0.35

 BoxFit

0.35

 Klar

0.35

 Table

0.35

newtheorem

0.35

Activations Density 0.125%

identifiers and keywords commonly used in programming or data structures

No Comments

No Known Activations