INDEX

Explanations

topics related to medical research and treatment processes

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

Appears in or after uncommon words

np_acts-logits-general · gemini-2.0-flash

endence, bility, ied, oxidants, eable, bases, sistence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

enumi

-0.60

 Sorgen

-0.50

ρ

-0.49

 leksikon

-0.48

 कैसी

-0.47

 المعيارى

-0.47

 Tarn

-0.46

λω

-0.46

 binoculars

-0.46

Φ

-0.45

POSITIVE LOGITS

 للمعارف

1.15

 تانيه

0.87

0.84

UnsafeEnabled

0.81

 vnto

0.78



0.77

حياته

0.77

Efq

0.75

esModule

0.74

ⓧ

0.74

Activations Density 1.213%

topics related to medical research and treatment processes

Appears in or after uncommon words

endence, bility, ied, oxidants, eable, bases, sistence

No Comments

No Known Activations