INDEX

Explanations

numerical values or ratings

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

CppMethod

-0.59

MigrationBuilder

-0.50

 betweenstory

-0.49

transQ

-0.48

 ProtoMessage

-0.48

)_/¯

-0.48

 оригіналу

-0.48

oredCriteria

-0.48

原始内容存档于

-0.47

 AssemblyCompany

-0.46

POSITIVE LOGITS

:✨

1.05

 nahilalakip

0.55

 gyhoeddwyd

0.44

adeloupe

0.41

 JADX

0.40

 gawas

0.40

 kasarigan

0.40

MacBook

0.39

VES

0.39

 Kade

0.38

Activations Density 0.000%

numerical values or ratings

No Comments

No Known Activations