INDEX

Explanations

elements and structure within a data model or document format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.40.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kinds

-1.02

Kinds

-0.96

 kinds

-0.84

kinds

-0.81

 Varieties

-0.75

 Sorten

-0.75

Tipos

-0.75

 Tipos

-0.73

 Types

-0.71

 Arten

-0.69

POSITIVE LOGITS

 type

1.00

type

0.98

型

0.57

tagHelperRunner

0.56

 tyle

0.46

 Lightboxes

0.44

كويكب

0.44

 physician

0.43

astify

0.43

 Réponses

0.42

Activations Density 0.300%

elements and structure within a data model or document format

No Comments

No Known Activations