INDEX

Explanations

terms and concepts related to programming or technical documentation

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UserScript

-0.38

INSEE

-0.35

AndroidJUnit

-0.35

MeasureSpec

-0.34

URIComponent

-0.34

thole

-0.34

chables

-0.34

 CURIAM

-0.33

 Organisationen

-0.32

kian

-0.32

POSITIVE LOGITS

 otomatig

0.58

 utafitiHapana

0.54

abstractmethod

0.52

 queſta

0.52

RegistryLite

0.50

 intptr

0.50

majánló

0.49

 lenker

0.49

 للمعارف

0.47

matchCondition

0.47

Activations Density 1.338%

terms and concepts related to programming or technical documentation

No Comments

No Known Activations