datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
PersonalizationV3enron_personalization_test
Dataset Card for "enron_personalization_test"
More Information needed
PersonalizationV4PG-Personalization-Amazon2023PersonalizationV3PersonalizationV3PersonalizationPersonalization_Databasefiltered_personalization_prompt_responsePersonalizationV2synth-persona-vectors
Persona Vectors
Mean activation vectors extracted from synthetic persona prompts.
This card is generated from the parquet files currently uploaded to the Hub.
Counts exclude the baseline_assistant row.
Available Vectors
Model
Config
Variants
Question set
QA filter
google/gemma-2-9b-it
google__gemma-2-9b-it__answer_mean
biography: 100 personas, 50 QA/persona; templated: 1000 personas, 50 QA/persona
train_test_split(n_train=50)
all
google/gemma-3-27b-it… See the full description on the dataset page: https://huggingface.co/datasets/implicit-personalization/synth-persona-vectors.personalization_prompt_response_euruspersonalization-agentic-rag-10docs-generationspersonalization_promptresponseTaeseo06-personalization-datasetpersonalization_prompt_responsepersonalization_prompt_response_oasst_pythia_1bmapo-dog-personalizationpersonalization_promptspersonalization-state-mini-experiment
Personalization State: Маленький экспериментальный датасет для формата промпта с интеграцией памяти и рекурсивности контекста
Разработано: Altterra
Лицензия: apache-2.0
Язык: ru
НЕ РЕКОМЕНДУЕТСЯ ДЛЯ ИСПОЛЬЗОВАНИЯ
Формат промпта:
# Input
<ShortTermMemory>
<VisualRepresentation>
[Визуальная информация и/или JSON-описание окружения]
</VisualRepresentation>
<PhonologicalRepresentation>
<External>
[Слова и фразы, поступающие из окружающего мира]… See the full description on the dataset page: https://huggingface.co/datasets/Altterra/personalization-state-mini-experiment.enron_personalization_dev
Dataset Card for "enron_personalization_dev"
More Information needed
personalization-agentic-rag-5docs-generationsenron_personalization_dev_uncapped
Enron Personalization — Dev (Uncapped)
DescriptionPer-user Enron email text with original case & punctuation, no word cap.Each user’s data is split 50/50 by sentence order into train and validation.
Users: 45
Split policy: first half of a user's sentences → train, second half → validation
Casing/punctuation: preserved
Build time: 2026-01-18 21:09 UTC
Columns
text: string (one sentence per row)
employee_id: string (derived from filename)
Notes
This dataset replaces the earlier… See the full description on the dataset page: https://huggingface.co/datasets/figmtu/enron_personalization_dev_uncapped.personalization_prompt_response_gemma_2bPersonalization_Benchpersonalization-agentic-rag-3docs-generationsshort_personalization_promptspersonalization_prompt_response_testpersonalization_prompt_response_mistralpersonalization_prompt_response_oasst_pythia_7b