INDEX

Explanations

structured data elements like labels and colors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.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

 GenerationType

-0.56

 kaarangay

-0.54

WindowConstants

-0.52

 mergeFrom

-0.51

 Wicidata

-0.50

StructEnd

-0.47

Regi

-0.47

titleMargin

-0.47

enschappelijke

-0.46

EventListener

-0.44

POSITIVE LOGITS

 contextLoads

0.60

addContainerGap

0.60

 <<<<<<<<<<<<<<

0.59

edakan

0.57

 labouring

0.56

 Clutch

0.53

ByVersion

0.52

 réessayer

0.52

MatButtonModule

0.51

 tdci

0.51

Activations Density 0.182%

structured data elements like labels and colors

No Comments

No Known Activations