[doc] update docs/model_cookbook.md

[alexsu52]

Changes:

1. Added --max-decode-steps=128. Without max-decode-steps specified, the command failed with the following error:

Traceback (most recent call last): File "<frozen runpy>", line 198, in _run_module_as_main File "<frozen runpy>", line 88, in _run_code File "/home/asuslov/projects/shark-ai/sharktank/sharktank/examples/paged_llm_v1.py", line 136, in <module> main() File "/home/asuslov/projects/shark-ai/sharktank/sharktank/examples/paged_llm_v1.py", line 130, in main results = decoder.greedy_decode(token_ids, args.max_decode_steps) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/asuslov/projects/shark-ai/sharktank/sharktank/utils/llm_utils.py", line 575, in greedy_decode page_ids = [ ^ File "/home/asuslov/projects/shark-ai/sharktank/sharktank/utils/llm_utils.py", line 576, in <listcomp> self._runner.allocate(token_count=len(req) + steps) for req in requests ~~~~~~~~~^~~~~~~ TypeError: unsupported operand type(s) for +: 'int' and 'NoneType'

2. Added --add-start-token cli argument. mistral-7b-v0.1 requires start token, without it, the model cyclically returns the promt.

:: Result: ['.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a llama.\n\nWrite a story about a']

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

❌ Patch coverage is 0% with 2 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (main@3fe7392). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
sharktank/sharktank/examples/paged_llm_v1.py	0.00%	1 Missing ⚠️
sharktank/sharktank/utils/cli.py	0.00%	1 Missing ⚠️
❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2623   +/-   ##
=======================================
  Coverage        ?   77.54%           
=======================================
  Files           ?      264           
  Lines           ?    25193           
  Branches        ?        0           
=======================================
  Hits            ?    19537           
  Misses          ?     5656           
  Partials        ?        0           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.