INDEX

Explanations

technical jargon or expressions related to integrated circuits and semiconductor technology

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

Tokens after "ref" or numbers

np_acts-logits-general · gemini-2.0-flash

references in code comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

setVerticalGroup

-0.53

 tomorrow

-0.51

 amanhã

-0.48

Samedi

-0.47

Refunds

-0.46

 Сегодня

-0.46

Сегодня

-0.45

})).

-0.44

Filmo

-0.44

 musiciens

-0.44

POSITIVE LOGITS

ThroughAttribute

0.80

Personendaten

0.77

 utafitiHapana

0.75

SourceChecksum

0.70

 disambiguazione

0.69

addCriterion

0.60

 ujednoznacz

0.60

 HasFactory

0.60

LLocation

0.59

 Dish

0.58

Activations Density 0.458%

technical jargon or expressions related to integrated circuits and semiconductor technology

Tokens after "ref" or numbers

references in code comments

No Comments

No Known Activations