thomwolf HF staff commited on
Commit
9d018f5
β€’
1 Parent(s): 0c6f487

explain better

Browse files
Files changed (1) hide show
  1. README.md +10 -1
README.md CHANGED
@@ -6,9 +6,11 @@ library_name: nanotron
6
 
7
  Modeling code for Mistral to use with [Nanotron](https://github.com/huggingface/nanotron/)
8
 
 
 
9
  ## πŸš€ Quickstart
10
 
11
- ```python
12
  # Generate a config file
13
  python config_tiny_mistral.py
14
 
@@ -17,6 +19,13 @@ export CUDA_DEVICE_MAX_CONNECTIONS=1 # important for some distributed operations
17
  torchrun --nproc_per_node=8 run_train.py --config-file config_tiny_mistral.yaml
18
  ```
19
 
 
 
 
 
 
 
 
20
  ## πŸš€ Use your custom model
21
 
22
  - Update the `MistralConfig` class in `config_tiny_mistral.py` to match your model's configuration
 
6
 
7
  Modeling code for Mistral to use with [Nanotron](https://github.com/huggingface/nanotron/)
8
 
9
+ Also contains converted pretrained weights for Mistral-7B-0.1: https://huggingface.co/mistralai/Mistral-7B-v0.1
10
+
11
  ## πŸš€ Quickstart
12
 
13
+ ```bash
14
  # Generate a config file
15
  python config_tiny_mistral.py
16
 
 
19
  torchrun --nproc_per_node=8 run_train.py --config-file config_tiny_mistral.yaml
20
  ```
21
 
22
+ ## πŸš€ Run generation with pretrained Mistral-7B-0.1
23
+
24
+ ```bash
25
+ export CUDA_DEVICE_MAX_CONNECTIONS=1
26
+ torchrun --nproc_per_node=1 run_generate.py --ckpt-path ./pretrained/Mistral-7B-v0.1
27
+ ```
28
+
29
  ## πŸš€ Use your custom model
30
 
31
  - Update the `MistralConfig` class in `config_tiny_mistral.py` to match your model's configuration