L34 MODEL PAGE EXCERPT (offline fallback for Activity A3)
-------------------------------------------------------------------
This is what the model page at
https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3
and the pages it links to said on 5 and 7 November 2026, reduced to the
parts a provenance check would read. It is an excerpt written by the
course, not a copy of the page. Use it only if you cannot open the
page. Write "excerpt, 2026-11-07" in the date field of your sheet.
Anything not listed here was not on the pages we read.

PAGE HEADER
  Repository:  mistralai / Mistral-7B-Instruct-v0.3
  Tags shown:  text generation, safetensors, apache-2.0 (licence tag)
  Card title:  "Model Card for Mistral-7B-Instruct-v0.3"
  Signed:      "The Mistral AI Team"

DESCRIPTION (the card's own words, quoted)
  "The Mistral-7B-Instruct-v0.3 Large Language Model (LLM) is an
  instruct fine-tuned version of the Mistral-7B-v0.3."

  Changes listed compared with the previous version:
  - Extended vocabulary to 32768
  - Supports v3 Tokenizer
  - Supports function calling

LINKED PAGE: the base model, mistralai / Mistral-7B-v0.3
  Its card says the model "is a Mistral-7B-v0.2 with extended
  vocabulary." That card links one step further back in the same way.

LIMITATIONS (the card's own words, quoted)
  "The Mistral 7B Instruct model is a quick demonstration that the
  base model can be easily fine-tuned to achieve compelling
  performance. It does not have any moderation mechanisms. We're
  looking forward to engaging with the community on ways to make the
  model finely respect guardrails, allowing for deployment in
  environments requiring moderated outputs."

FILES TAB
  Each large file has its own page showing its size and a SHA256
  value. For consolidated.safetensors (14.5 GB) the page shows:
  76d5729be995fb6510c7f103c5b8825d3d82f8126ac949db6be367a638694d1f

OTHER SECTIONS ON THE CARD
  The card's headings, in order: Installation; Download (Chat,
  Instruct following, Function calling); Generate with transformers;
  Function calling with transformers; Limitations; The Mistral AI Team
  (a list of contributor names).

That is the whole of what the pages state. Before you mark anything
E, find the line above that establishes it. Before you mark anything
N, check the whole excerpt, including the headings for what is not
there.
