INDEX

Explanations

terms and components related to electrical circuits, specifically diodes and transistors

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 StringTokenizer

-0.44

 teneur

-0.41

 Ambiental

-0.41

Katar

-0.40

 balkon

-0.39

 Gestão

-0.39

 maîtrise

-0.39

 promotion

-0.39

 cofre

-0.39

 papier

-0.39

POSITIVE LOGITS

 diode

0.65

 diodes

0.64

 Diode

0.56

diode

0.56

 InputDecoration

0.52



0.44

 <>",

0.42

addCriterion

0.40

Dio

0.40

 sumpay

0.39

Activations Density 0.018%

terms and components related to electrical circuits, specifically diodes and transistors

No Comments

No Known Activations