Fix quant doc 2 (#643)

Summary:
Generate step did not specify model name args. Names llama3. Fix that.

Author: kimishpatel

docs/quantization.md

@@ -69,13 +69,13 @@ python3 generate.py [--compile] llama3 --prompt "Hello, my name is" --quantize '
 ```
 python3 torchchat.py export llama3 --quantize '{"embedding": {"bitwidth": 4, "groupsize":32}, "linear:int4": {"groupsize" : 256}}' --output-dso-path llama3.dso
 
-python3 generate.py --dso-path llama3.dso  --prompt "Hello my name is"
+python3 generate.py llama3 --dso-path llama3.dso  --prompt "Hello my name is"
 ```
 ### ExecuTorch
 ```
 python3 torchchat.py export llama3 --dtype fp32 --quantize '{"embedding": {"bitwidth": 4, "groupsize":32}, "linear:a8w4dq": {"groupsize" : 256}}' --output-pte-path llama3.pte
 
-python3 generate.py --pte-path llama3.pte  --prompt "Hello my name is"
+python3 generate.py llama3 --pte-path llama3.pte  --prompt "Hello my name is"
 ```
 
 ## Model precision (dtype precision setting)