INDEX

Explanations

programming syntax related to component rendering and structure

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndEndTag

-0.64

sidemargin

-0.55

ManagedBean

-0.54

ValueStyle

-0.54

 enamel

-0.52

 setEmail

-0.51

 للمعارف

-0.51

NSIndexPath

-0.50

теризу

-0.49

 acrylique

-0.49

POSITIVE LOGITS

ArgsConstructor

0.84

__":

0.54

oluzione

0.51

曖昧さ回避

0.51

 nakalista

0.51

ixos

0.48

__":

0.48

UnsafeEnabled

0.45

DSL

0.45

ulo

0.44

Activations Density 0.324%

programming syntax related to component rendering and structure

No Comments

No Known Activations