INDEX

Explanations

key nouns or organizational terms that indicate a subject or main idea

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

ParallelGroup

-0.38

 abstraction

-0.37

MLLoader

-0.37

بوابة

-0.36

olism

-0.34

 generalization

-0.34

Obrázky

-0.34

Ufficio

-0.34

 opposition

-0.34

aplic

-0.33

POSITIVE LOGITS

хьтан

0.63

:✨

0.54

 AssemblyCulture

0.52

scryfall

0.51

 continúas

0.50

DockStyle

0.49

EDEFAULT

0.49

帖最后由

0.48

 Meksiku

0.48

 utafitiHapana

0.48

Activations Density 0.000%

key nouns or organizational terms that indicate a subject or main idea

No Comments

No Known Activations