INDEX

Explanations

patterns or codes related to data structures or object representations

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

Hexadecimal and other codes

np_acts-logits-general · gemini-2.0-flash

letters d, e, D

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

 Paglinawan

-0.73

 ostavi

-0.72

IUrlHelper

-0.69

 disambiguazione

-0.69

intios

-0.69

ParallelGroup

-0.67

 ProtoMessage

-0.63

QMetaType

-0.62

expandindo

-0.60

Personensuche

-0.60

POSITIVE LOGITS

IMPORTED

0.51

gatron

0.47

0.46

 Redacción

0.44

Saludos

0.43

ябре

0.42

0.41

uteurs

0.40

 ویکی‌پدیای

0.40

Activations Density 0.406%

patterns or codes related to data structures or object representations

Hexadecimal and other codes

letters d, e, D

No Comments

No Known Activations