INDEX

Explanations

terms related to dimensions and measurements

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

dimensions and measurements

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

Tikang

-0.58

 tartalomajánló

-0.46

abestanden

-0.44

Xna

-0.38

horabuena

-0.38

im

-0.36

makeConstraints

-0.36

 кӀ

-0.34

joba

-0.33

 OFDb

-0.33

POSITIVE LOGITS

Gambas

0.49

 ujednoznacz

0.47

borderBottom

0.47

 ArrayAdapter

0.47

arraycopy

0.47

 Obrador

0.47

Зноскі

0.47

 Hände

0.47

outWeight

0.46

 مرئيه

0.46

Activations Density 0.194%

terms related to dimensions and measurements

dimensions and measurements

No Comments

No Known Activations