INDEX

Explanations

references to patent applications and legal documentation details

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

application No

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

 casó

-0.44

}{*}{}

-0.42

 Parigi

-0.42

 oprot

-0.42

 HasFactory

-0.41

expandindo

-0.40

 propOrder

-0.39

 ويكيپيديا

-0.39

Abitanti

-0.36

 recorrido

-0.35

POSITIVE LOGITS

 Chwiliwch

0.54

ConstraintMaker

0.53

 ModelRenderer

0.50

ValueStyle

0.48

TestingModule

0.48

Rhestr

0.47

Kaynakça

0.47

الدراسه

0.46

 resourceCulture

0.46

 مواليد

0.45

Activations Density 0.020%

references to patent applications and legal documentation details

application No

No Comments

No Known Activations