INDEX

Explanations

terms related to health conditions and treatment factors

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

medical conditions and treatments

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للاسماء

-0.57

shadowOpacity

-0.50

 الحره

-0.49

 étoit

-0.47

LookAnd

-0.47

NameInMap

-0.47

Composable

-0.46

 CreateTagHelper

-0.45

 contextLoads

-0.45

 wikipagina

-0.45

POSITIVE LOGITS

 bandar

0.40

Via

0.39

 Advanced

0.38

 avance

0.38

 advanced

0.38

 ocio

0.37

Зноскі

0.36

 Bergmann

0.35

osti

0.35

advance

0.35

Activations Density 0.254%

terms related to health conditions and treatment factors

medical conditions and treatments

No Comments

No Known Activations