INDEX

Explanations

mathematical notation and expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 Majefty

-0.71

sidemargin

-0.69

 itſelf

-0.66

 Theſe

-0.65

 betweenstory

-0.65

 Jefus

-0.65

](#

-0.64

 MainAxisSize

-0.63

ContentAsync

-0.63

ſelf

-0.63

POSITIVE LOGITS

↵↵

0.64

<eos>

0.59

forChild

0.53

 rate

0.50

0.48

 taux

0.48

XmlRootElement

0.47

end

0.45

paire

0.45

Activations Density 0.388%

mathematical notation and expressions

No Comments

No Known Activations