INDEX

Explanations

phrases focused on capabilities and potential actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 GenerationType

-0.76

Sucesor

-0.72

økt

-0.66

 Italijani

-0.63

 disambiguazione

-0.62

 Meksiku

-0.60

 chi̍t

-0.58

UnknownFields

-0.57

 препратки

-0.57

 unknownFields

-0.56

POSITIVE LOGITS

 möglichst

0.83

 hopefully

0.81

 puissiez

0.69

hopefully

0.67

can

0.66

 possa

0.65

 pueda

0.64

 avoid

0.63

 puedan

0.62

 puissent

0.60

Activations Density 0.153%

phrases focused on capabilities and potential actions

No Comments

No Known Activations