feat: Review docs #749
Open
feat: Review docs #749
+89
−87
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jgerh
reviewed
Nov 3, 2025
jgerh
reviewed
Nov 3, 2025
jgerh
reviewed
Nov 4, 2025
| @@ -1,16 +1,16 @@ | |||
| # LLM Pre-Training with NeMo AutoModel | |||
|
|
|||
There was a problem hiding this comment.
LLM Pre-Training with NeMo Automodel
| # LLM Pre-Training with NeMo AutoModel | ||
|
|
||
| This guide covers **FineWeb** data preparation, **defining** a [NanoGPT‑style](https://github.com/KellerJordan/modded-nanogpt) model, and **launching and monitoring** a NeMo AutoModel pre‑training run. | ||
|
|
docs/guides/llm/pretraining.md
Outdated
Comment on lines
5
to
12
| In particular, it will show you how to: | ||
| In particular, it will show you how to: | ||
| 1. [Install NeMo AutoModel from git](#1-environment-setup). | ||
| 2. [Pre-process and tokenize the FineWeb dataset](#2-pre-process-the-fineweb-dataset). | ||
| 3. [Introduction to the NeMo AutoModel training workflow](#3-introduction-to-the-nemo-automodel-training-workflow). | ||
| 4. [Define your own model architecture](#4-define-your-own-model-architecture). | ||
| 5. [Inspect and adjust the YAML configuration](#5-inspect-and-adjust-the-yaml-configuration). | ||
| 6. [Launch training](#6-launch-training). | ||
| 7. [Monitoring and evaluation](#7-monitoring-and-evaluation). |
docs/guides/llm/pretraining.md
Outdated
|
|
||
| --- | ||
|
|
||
| ## 1. Environment setup |
There was a problem hiding this comment.
Suggested change
| ## Set Up Your Environment |
docs/model-coverage/overview.md
Outdated
| - Add custom implementations to extend support to models not yet covered upstream. | ||
| - Provide optimized, extended or faster implementations for specific models while retaining the same AutoModel interface. | ||
|
|
||
| ## Supported Hugging Face Auto classes |
There was a problem hiding this comment.
Suggested change
| ## Supported Hugging Face Auto Classes |
docs/model-coverage/overview.md
Outdated
| | `AutoModelForSequenceClassification`| Sequence Classification | WIP | Early support; interfaces may change. | | ||
|
|
||
|
|
||
| ## When a model listed on Hugging Face Hub may not be supported |
There was a problem hiding this comment.
Suggested change
| ## Troubleshooting Unsupported Models |
docs/model-coverage/overview.md
Outdated
|
|
||
| ## When a model listed on Hugging Face Hub may not be supported | ||
|
|
||
| Sometimes a model listed on the Hugging Face Hub may not support finetuning in NeMo AutoModel. |
There was a problem hiding this comment.
Suggested change
| Sometimes a model listed on the Hugging Face hub may not support fine-tuning in NeMo Automodel. |
docs/model-coverage/overview.md
Outdated
| | Model upper-bounds transformer version, requiring older version | — | Open a new GitHub issue. | | ||
| | Unsupported checkpoint format | OSError: `meta-llama/Llama-2-70b` does not appear to have a file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt or flax_model.msgpack. | Open a new GitHub issue or request from the model publisher to share a safetensors checkpoint. | | ||
|
|
||
| These cases typically stem from upstream packaging or dependency constraints and would surface similarly when using `transformers` directly. AutoModel mirrors the familiar load/finetune semantics. |
There was a problem hiding this comment.
Suggested change
| These cases typically stem from upstream packaging or dependency constraints. You would encounter the same issues when using `transformers` directly, as Automodel mirrors the familiar load and fine-tune semantics. |
docs/model-coverage/overview.md
Outdated
|
|
||
| If you encounter any issue, you can try: | ||
|
|
||
| - Upgrade to a NeMo AutoModel release that supports the required `transformers` version. |
There was a problem hiding this comment.
Suggested change
| - Upgrade to a NeMo Automodel release that supports the required `transformers` version. |
jgerh
reviewed
Nov 4, 2025
Co-authored-by: jgerh <163925524+jgerh@users.noreply.github.com> Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
Co-authored-by: jgerh <163925524+jgerh@users.noreply.github.com> Signed-off-by: Alexandros Koumparoulis <akoumparouli@nvidia.com>
akoumpa
force-pushed
the
akoumparouli/update_docs2
branch
from
2d86254 to
9d21f68
Compare
|
/ok to test 9d21f68 |
|
/ok to test d79426d |
|
/ok to test f874220 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Docs review for: