INDEX

Explanations

negative connotations or terms related to declines, with a focus on scientific or medical contexts

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

references to scientific papers

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

Izvori

-0.59

⚭

-0.58

brigens

-0.56

Kontra

-0.53

 Umständen

-0.52

บ้าง

-0.52

gany

-0.52

inephrine

-0.51

таратура

-0.51

Véxase

-0.50

POSITIVE LOGITS

featureID

0.63

 typelib

0.61

GEBURTS

0.56

 >=",

0.55

Personendaten

0.54

IonicModule

0.54

 presentato

0.53

®.

0.53

 AssemblyProduct

0.52

"](

0.52

Activations Density 0.030%

negative connotations or terms related to declines, with a focus on scientific or medical contexts

references to scientific papers

No Comments

No Known Activations