INDEX

Explanations

terms related to JSON data handling and manipulation

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-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

 reader

-1.27

 readers

-0.97

 leitor

-0.88

 snippetHide

-0.76

?')

-0.72

 lectores

-0.70

 lecteur

-0.68

.)}

-0.68

 readership

-0.66

 READER

-0.64

POSITIVE LOGITS

SpringRunner

0.75

NewUrlParser

0.68

voorbeeld

0.66

 Roskov

0.64

rowave

0.63

 FileReader

0.62

rosia

0.60

PRP

0.59

chuan

0.59

ageable

0.58

Activations Density 0.022%

terms related to JSON data handling and manipulation

No Comments

No Known Activations