INDEX

Explanations

mathematical expressions and notations associated with statistical analysis

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

Numbers in references, citations, or lists

np_acts-logits-general · gemini-2.0-flash

numbers as words

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Two

-0.43

TWO

-0.43

Two

-0.43

 THOUS

-0.41

Three

-0.40

One

-0.40

 Zwei

-0.39

Thousands

-0.39

 dedi

-0.39

 thousands

-0.38

POSITIVE LOGITS

 fourteen

1.08

 thirteen

1.04

 sixteen

1.03

 fifteen

1.02

 Thirteen

1.00

 eighteen

0.99

 twenty

0.99

 twelve

0.97

 Eighteen

0.96

Thirteen

0.95

Activations Density 1.523%

mathematical expressions and notations associated with statistical analysis

Numbers in references, citations, or lists

numbers as words

No Comments

No Known Activations