zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-06-26 18:51:18 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

change file name for consistency, update readme.

Browse Source
This commit is contained in:
Awni Hannun 2023-12-14 08:34:24 -08:00
parent 840c0c36c2
commit 3d2a23184a
2 changed files with 12 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
phi2/README.md
View File

@ -1,8 +1,9 @@
# Phi-2 # Phi-2
Phi-2 is a 2.7B parameter model released by Microsoft[^1] and trained on a mixture Phi-2 is a 2.7B parameter model released by Microsoft[^1] and trained on a mixture
of GPT-4 outputs and clean web-text. Its performance rivals of GPT-4 outputs and clean web-text. Its performance rivals much larger models.
much, much stronger models.
Phi-2 efficiently runs on an Apple silicon device with 8 GB memory in 16-bit precision.
## Setup ## Setup
@ -12,14 +13,14 @@ Download and convert the model:
python convert.py python convert.py
``` ```
which will make a file `weights.npz`. This will make the `weights.npz` file which MLX can read.
## Generate ## Generate
To generate text with the default prompt: To generate text with the default prompt:
```sh ```sh
python model.py python phi2.py
``` ```
Should give the output: Should give the output:
@ -40,7 +41,13 @@ Answer: Logic in mathematics is like a compass in navigation. It helps
To use your own prompt: To use your own prompt:
```sh ```sh
python model.py --prompt <your prompt here> --max_tokens <max_token> python phi2.py --prompt <your prompt here> --max_tokens <max_tokens_to_generate>
```
To see a list of options run:
```sh
python phi2.py --help
``` ```
[^1]: For more details on the model see the [blog post]( [^1]: For more details on the model see the [blog post](

0
phi2/model.py → phi2/phi2.py
View File