INDEX

Explanations

terms related to lung cancer classifications and types

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

lung cancer or carcinoma

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

protoimpl

-0.40

 للاسماء

-0.39

 llenar

-0.38

ComponentName

-0.37

 riemp

-0.36

 impresión

-0.35

 Yesterday

-0.34

Cubit

-0.33

NameInMap

-0.33

ScopeManager

-0.32

POSITIVE LOGITS

othelioma

0.76

 Italijani

0.62

 cancer

0.60

 cancers

0.58

 lymphoma

0.57

cancer

0.54

endfor

0.52

 Cancer

0.52

arcoma

0.52

 carcinoma

0.52

Activations Density 0.090%

terms related to lung cancer classifications and types

lung cancer or carcinoma

No Comments

No Known Activations