referring out to updated models
Browse files
README.md
CHANGED
|
@@ -10,6 +10,13 @@ tags:
|
|
| 10 |
- inferentia2
|
| 11 |
- neuron
|
| 12 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
# Neuronx model for Mistral
|
| 14 |
|
| 15 |
This repository contains [AWS Inferentia2](https://aws.amazon.com/ec2/instance-types/inf2/) and [`neuronx`](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/) compatible checkpoints for [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1).
|
|
|
|
| 10 |
- inferentia2
|
| 11 |
- neuron
|
| 12 |
---
|
| 13 |
+
# Please read
|
| 14 |
+
This repository was based on the transformer implementation of Mistral before Optimum-neuron included support.
|
| 15 |
+
|
| 16 |
+
Consider using an Optimum based repository such as [this](https://huggingface.co/aws-neuron/Mistral-7B-Instruct-v0.1-neuron-1x2048-2-cores/tree/main).
|
| 17 |
+
|
| 18 |
+
This is especially important if you are changing any paramters that require a recompile because Optimum-neuron will let you take advantage of the compilation cache.
|
| 19 |
+
|
| 20 |
# Neuronx model for Mistral
|
| 21 |
|
| 22 |
This repository contains [AWS Inferentia2](https://aws.amazon.com/ec2/instance-types/inf2/) and [`neuronx`](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/) compatible checkpoints for [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1).
|