Update README.md

README.md

@@ -4,14 +4,53 @@ tags:
 - vllm
 ---
 
+# Meta-Llama-3-8B-Instruct-FP8-KV
 
-Meta-Llama-3-8B-Instruct quantized to FP8 weights and activations using per-tensor quantization, ready for inference with vLLM >= 0.5.0. 
-This model checkpoint also includes experimental per-tensor scales for FP8 quantized KV Cache, accessed through the `--kv-cache-dtype fp8` argument in vLLM.
+## Model Overview
+Meta-Llama-3-8B-Instruct quantized to FP8 weights and activations using per-tensor quantization, ready for inference with vLLM >= 0.5.0.
+This model checkpoint also includes per-tensor scales for FP8 quantized KV Cache, accessed through the `--kv-cache-dtype fp8` argument in vLLM.
 
 ```python
 from vllm import LLM
-
-model = LLM(model="nm-testing/Meta-Llama-3-8B-Instruct-FP8-KV", kv_cache_dtype="fp8")
+model = LLM(model="neuralmagic/Meta-Llama-3-8B-Instruct-FP8-KV", kv_cache_dtype="fp8")
 result = model.generate("Hello, my name is")
 ```
 
+## Usage and Creation
+Produced using [AutoFP8 with calibration samples from ultrachat](https://github.com/neuralmagic/AutoFP8/blob/147fa4d9e1a90ef8a93f96fc7d9c33056ddc017a/example_dataset.py).
+
+```python
+from datasets import load_dataset
+from transformers import AutoTokenizer
+
+from auto_fp8 import AutoFP8ForCausalLM, BaseQuantizeConfig
+
+pretrained_model_dir = "meta-llama/Meta-Llama-3-8B-Instruct"
+quantized_model_dir = "Meta-Llama-3-8B-Instruct-FP8-KV"
+
+tokenizer = AutoTokenizer.from_pretrained(pretrained_model_dir, use_fast=True)
+tokenizer.pad_token = tokenizer.eos_token
+
+ds = load_dataset("mgoin/ultrachat_2k", split="train_sft")
+examples = [tokenizer.apply_chat_template(batch["messages"], tokenize=False) for batch in ds]
+examples = tokenizer(examples, padding=True, truncation=True, return_tensors="pt").to("cuda")
+
+quantize_config = BaseQuantizeConfig(
+    quant_method="fp8",
+    activation_scheme="static",
+    ignore_patterns=["re:.*lm_head"],
+    kv_cache_quant_targets=("k_proj", "v_proj"),
+)
+
+model = AutoFP8ForCausalLM.from_pretrained(pretrained_model_dir, quantize_config)
+model.quantize(examples)
+model.save_quantized(quantized_model_dir)
+```
+
+## Evaluation
+
+### Open LLM Leaderboard evaluation scores
+|                      | Meta-Llama-3-8B-Instruct | Meta-Llama-3-8B-Instruct-FP8 | Meta-Llama-3-8B-Instruct-FP8-KV<br>(this model) |
+| :------------------: | :----------------------: | :--------------------------: | :---------------------------------------------: |
+| gsm8k<br>5-shot      | 75.44                    | 74.37                        | 74.98                                           |
+