INDEX

Explanations

components and parameters related to programming or coding contexts

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

mathematical expressions and code

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

IMAL

-0.44

бора

-0.39

 driving

-0.39

 Thurston

-0.37

itary

-0.37

 pause

-0.37

 paddle

-0.36

 cima

-0.36

原

-0.36

prechung

-0.36

POSITIVE LOGITS

AISSEE

0.73

EDEFAULT

0.57

oredCriteria

0.53

 كومونز

0.52

 resourceCulture

0.51

 мәкал

0.51

 disambiguazione

0.50

 PagesJaunes

0.47

цездатний

0.47

 يتيمه

0.46

Activations Density 0.028%

components and parameters related to programming or coding contexts

mathematical expressions and code

No Comments

No Known Activations