INDEX

Explanations

technical terms and phrases related to science and research

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

docker container

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ьаж

-0.56

ftagPool

-0.54

protoimpl

-0.54

Tembelea

-0.52

########.

-0.50

makeConstraints

-0.47

RegistryLite

-0.47

NgModule

-0.45

featureID

-0.44

AppMethodBeat

-0.44

POSITIVE LOGITS

HostException

0.38

 barista

0.38

dasarkan

0.38

 Telefon

0.37

DeleteBehavior

0.37

 snipers

0.36

 TextAppearance

0.36

PerformLayout

0.36

JvmStatic

0.36

 iſt

0.36

Activations Density 0.067%

technical terms and phrases related to science and research

docker container

No Comments

No Known Activations