Tendril Foundation Public Data Sample, Attribution and License Notice
=====================================================================

This NOTICE accompanies the files tendril-foundation-sample.jsonl / .json / .csv
in this directory. Those files are a SMALL, FILTERED, PER-ROW LICENSE-TAGGED
DERIVATIVE blended from two license-clean public preference datasets. They are
NOT a verbatim copy of either source. Each row was selected from Tendril's
committed seed buffer, re-serialized into a clean public schema, and re-emitted
with its OWN provenance (source dataset, license, source row id).

PROVENANCE IS PER ROW. Every row carries its own `provenance.source_dataset`,
`provenance.license`, and `provenance.source_row_id`. Apply the license named on
a given row to that row.

----------------------------------------------------------------------
SOURCE A , LAION-AI / Open-Assistant (oasst2)
----------------------------------------------------------------------
  Dataset:  OpenAssistant/oasst2
  Project:  LAION-AI/Open-Assistant
  License:  Apache License, Version 2.0 (Apache-2.0)
  Paper:    "OpenAssistant Conversations -- Democratizing Large Language
            Model Alignment", arXiv:2304.07327
  Row tag:  provenance.source_dataset = "OpenAssistant/oasst2",
            provenance.license = "Apache-2.0"

  Rows derived from this source are licensed under the Apache License,
  Version 2.0. A copy is available at:
      https://www.apache.org/licenses/LICENSE-2.0
  Per Apache-2.0 Section 4(d), this NOTICE preserves the attribution required
  for redistribution of derivative works. The rows tagged "OpenAssistant/oasst2"
  are a derivative of the oasst2 dataset, modified by filtering, pair extraction,
  and re-serialization.

----------------------------------------------------------------------
SOURCE B , Anthropic / hh-rlhf (helpful-base)
----------------------------------------------------------------------
  Dataset:  Anthropic/hh-rlhf  (subset: helpful-base)
  License:  MIT License
  Paper:    "Training a Helpful and Harmless Assistant with Reinforcement
            Learning from Human Feedback", arXiv:2204.05862
  Row tag:  provenance.source_dataset = "Anthropic/hh-rlhf",
            provenance.license = "MIT (via Anthropic/hh-rlhf)"

  Rows derived from this source are licensed under the MIT License. Only the
  helpful-base subset is used; the combined and red-team splits are excluded.

  MIT License (as applicable to the hh-rlhf-derived rows):

      Permission is hereby granted, free of charge, to any person obtaining a
      copy of this software and associated documentation files (the "Software"),
      to deal in the Software without restriction, including without limitation
      the rights to use, copy, modify, merge, publish, distribute, sublicense,
      and/or sell copies of the Software, and to permit persons to whom the
      Software is furnished to do so, subject to the following conditions:

      The above copyright notice and this permission notice shall be included in
      all copies or substantial portions of the Software.

      THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
      IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
      FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
      AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
      LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
      FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
      DEALINGS IN THE SOFTWARE.

----------------------------------------------------------------------
Summary
----------------------------------------------------------------------
This sample is a filtered, per-row license-tagged derivative blending
OpenAssistant/oasst2 (Apache-2.0) and Anthropic/hh-rlhf helpful-base (MIT). Each
row carries its own source and license; apply the corresponding terms on a
per-row basis. The sample is in English and demonstrates data format and
provenance discipline only; it is not underserved-language data. Retain this
NOTICE with any redistribution.