INDEX

Explanations

references to programming concepts and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

########.

-0.65

 JpaRepository

-0.52

namics

-0.51

PEG

-0.47

LSM

-0.47

ichen

-0.47

▲

-0.46

parsedMessage

-0.46

.*")]

-0.46

bett

-0.46

POSITIVE LOGITS

 default

0.40

 automaticamente

0.39

vů

0.39

 jinak

0.39

 кӀ

0.39

 unless

0.37

 automatically

0.37

 Schluß

0.36

 hurtig

0.35

unless

0.34

Activations Density 1.218%

references to programming concepts and documentation

No Comments

No Known Activations