davidshtian/Mistral-7B-Instruct-v0.2-neuron-1x2048-2-cores

Neuronx for mistralai/Mistral-7B-Instruct-v0.2 - Updated Mistral 7B Model on AWS Inferentia2

This model has been exported to the neuron format using specific input_shapes and compiler parameters detailed in the paragraphs below.

Please refer to the 🤗 optimum-neuron documentation for an explanation of these parameters.

Note: To compile the mistralai/Mistral-7B-Instruct-v0.2 on Inf2, you need to update the model config sliding_window (either file or model variable) from null to default 4096.

Usage with 🤗 optimum-neuron

>>> from optimum.neuron import pipeline

>>> p = pipeline('text-generation', 'davidshtian/Mistral-7B-Instruct-v0.2-neuron-1x2048-2-cores')
>>> p("My favorite place on earth is", max_new_tokens=64, do_sample=True, top_k=50)
[{'generated_text': 'My favorite place on earth is Hawaii,” she said, her voice bright and clear despite her quietness.
“That place, and the ocean. But it’s hard to ever live there permanently. The ocean is there and it calls to me, but
it’s big and vast and doesn’t allow me a lot of freedom.”'}]

This repository contains tags specific to versions of neuronx. When using with 🤗 optimum-neuron, use the repo revision specific to the version of neuronx you are using, to load the right serialized checkpoints.

Arguments passed during export

input_shapes

{
  "batch_size": 1,
  "sequence_length": 2048,
}

compiler_args

{
  "auto_cast_type": "bf16",
  "num_cores": 2,
}