INDEX

Explanations

references to coding or programming concepts and specific numerical identifiers

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

Numbers, including single digits

np_acts-logits-general · gemini-2.0-flash

numbers and counts

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

ieR

-0.42

 Factors

-0.36

omitempty

-0.36

 Citations

-0.36

ANTLR

-0.36

 <>",

-0.35

Referencies

-0.35

 Bioaccumulative

-0.35

__*/

-0.35

phosa

-0.35

POSITIVE LOGITS

addCriterion

0.53

 incluido

0.51

 navíc

0.49

 nonUne

0.49

 included

0.48

BagConstraints

0.47

GEBURTSDATUM

0.45

 juſ

0.44

 excluded

0.43

 inclus

0.42

Activations Density 0.730%

references to coding or programming concepts and specific numerical identifiers

Numbers, including single digits

numbers and counts

No Comments

No Known Activations