INDEX

Explanations

references to matrices or matrix operations in a mathematical or computational context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

tvguidetime

-0.76

çalves

-0.57

</blockquote>

-0.56

ử

-0.48

 Might

-0.47

<eos>

-0.47

itinéraire

-0.46

 Williamson

-0.46

↵↵

-0.45

POSITIVE LOGITS

InjectAttribute

0.88

تقاوى

0.86

TintMode

0.79

EndContext

0.78

AsUp

0.76

CloseOperation

0.75

󠁣

0.75

BufferException

0.73

BeginInit

0.70

 >=",

0.70

Activations Density 0.033%

references to matrices or matrix operations in a mathematical or computational context

No Comments

No Known Activations