INDEX

Explanations

elements related to URLs or web links

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

URLs with percent encoding

np_acts-logits-general · gemini-2.0-flash

publication year

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

 versus

-0.69

vs

-0.68

 Versus

-0.47

vs

-0.45

DataSetChanged

-0.44

 effetto

-0.43

Oneof

-0.43

Precautionary

-0.43

versus

-0.43

 возраст

-0.42

POSITIVE LOGITS

GEBURTSDATUM

0.83

MigrationBuilder

0.73

出版年

0.71

 viewDidLoad

0.67

tvguidetime

0.65

 Majefty

0.63

 onCreateView

0.63

ніципалі

0.63

Clik

0.62

StoryboardSegue

0.62

Activations Density 0.131%

elements related to URLs or web links

URLs with percent encoding

publication year

No Comments

No Known Activations