INDEX

Explanations

terms related to medical data and patient characteristics

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

initial conditions or outcomes

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

 zile

-0.31

@[+][

-0.31

 icing

-0.29

 formaggio

-0.29

 installieren

-0.29

 half

-0.29

ref

-0.28

 vapa

-0.28

 chi̍t

-0.28

 mencu

-0.27

POSITIVE LOGITS

Бахар

0.59

IBOutlet

0.59

 betweenstory

0.57

yntaxException

0.54

OOTDTY

0.54

 gynhyrchwyd

0.51

قایناق‌لار

0.50

󠀠

0.50

 setUsername

0.50

ferrer

0.49

Activations Density 0.390%

terms related to medical data and patient characteristics

initial conditions or outcomes

No Comments

No Known Activations