Commit
•
104f750
1
Parent(s):
06cf5d0
Update README.md
Browse files
README.md
CHANGED
|
@@ -117,7 +117,7 @@ Alternatively, you can use NeMo Megatron training docker container with all depe
|
|
| 117 |
git clone https://github.com/NVIDIA/NeMo.git
|
| 118 |
cd NeMo/examples/nlp/language_modeling
|
| 119 |
git checkout v1.11.0
|
| 120 |
-
python megatron_gpt_eval.py gpt_model_file=
|
| 121 |
```
|
| 122 |
|
| 123 |
### Step 3: Send prompts to your model!
|
|
|
|
| 117 |
git clone https://github.com/NVIDIA/NeMo.git
|
| 118 |
cd NeMo/examples/nlp/language_modeling
|
| 119 |
git checkout v1.11.0
|
| 120 |
+
python megatron_gpt_eval.py gpt_model_file=nemo_2b_bf16_tp1.nemo server=True tensor_model_parallel_size=1 trainer.devices=1
|
| 121 |
```
|
| 122 |
|
| 123 |
### Step 3: Send prompts to your model!
|