INDEX

Explanations

mathematical terminology and formalism related to theorems and propositions

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

Text within brackets, especially theorem-like environments

np_acts-logits-general · gemini-2.0-flash

theorem, proposition, lemma labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 BorderSide

-0.40

ὺς

-0.39

AddTagHelper

-0.39

"]);

-0.38

 kalın

-0.37

fall

-0.37

 drenaje

-0.37

 universelle

-0.37

 darkness

-0.36

 démocr

-0.36

POSITIVE LOGITS

newtheorem

1.10

 GenerationType

0.52

 ProtoMessage

0.51

 autorytatywna

0.51

IRQn

0.47

 stanovnika

0.44

 numberOfRows

0.44

 cherchés

0.43

المناصب

0.43

baijan

0.42

Activations Density 0.846%

mathematical terminology and formalism related to theorems and propositions

Text within brackets, especially theorem-like environments

theorem, proposition, lemma labels

No Comments

No Known Activations