INDEX

Explanations

references to various types of objects and their relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

GraphicsUnit

-0.80

}^{(\

-0.67

migrationBuilder

-0.65

 GenerationType

-0.65

 Spraw

-0.64

ثيق

-0.64

 SafeMath

-0.63

Autowired

-0.63

Numerology

-0.61

 mücade

-0.60

POSITIVE LOGITS

 objects

2.71

 object

2.64

 Objects

2.45

objects

2.32

 Object

2.31

 OBJECT

2.27

Objects

2.18

object

2.15

Object

1.98

OBJECT

1.94

Activations Density 0.084%

references to various types of objects and their relationships

No Comments

No Known Activations