INDEX

Explanations

terms related to computational complexity and tractability of problems, specifically within the context of Boolean circuits and tree-width

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

InstrumentedTest

-0.66

BeginInit

-0.59

enderror

-0.59

AndEndTag

-0.59

EndInit

-0.55

raszamy

-0.53

 defaultstate

-0.51

 mapStateToProps

-0.51

ParallelGroup

-0.50

ñor

-0.50

POSITIVE LOGITS

 متعلقه

0.63

Probl

0.52

 Bäume

0.51

oday

0.51

 mineures

0.51

 colori

0.51

 setPassword

0.50

 Burrows

0.50

 الدولى

0.49

 alberi

0.48

Activations Density 1.775%

terms related to computational complexity and tractability of problems, specifically within the context of Boolean circuits and tree-width

No Comments

No Known Activations