Create README.md

README.md

@@ -0,0 +1,317 @@
+---
+tags:
+- fp8
+- vllm
+license: llama3.1
+license_link: https://huggingface.co/meta-llama/Meta-Llama-3.1-8B/blob/main/LICENSE
+language:
+- en
+---
+
+# Meta-Llama-3.1-70B-FP8
+
+## Model Overview
+- **Model Architecture:** Meta-Llama-3.1
+  - **Input:** Text
+  - **Output:** Text
+- **Model Optimizations:**
+  - **Weight quantization:** FP8
+  - **Activation quantization:** FP8
+- **Intended Use Cases:** Intended for commercial and research use in multiple languages. Similarly to [Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B), this model serves as a base version.
+- **Out-of-scope:** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in languages other than English.
+- **Release Date:** 7/23/2024
+- **Version:** 1.0
+- **License(s):** [llama3.1](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B/blob/main/LICENSE)
+- **Model Developers:** Neural Magic
+
+Quantized version of [Meta-Llama-3.1-70B](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B).
+It achieves an average score of 79.70 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 79.84.
+
+### Model Optimizations
+
+This model was obtained by quantizing the weights and activations of [Meta-Llama-3.1-70B](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B) to FP8 data type, ready for inference with vLLM built from source.
+This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%.
+
+Only the weights and activations of the linear operators within transformers blocks are quantized. Symmetric per-tensor quantization is applied, in which a single linear scaling maps the FP8 representations of the quantized weights and activations.
+[LLM Compressor](https://github.com/vllm-project/llm-compressor) is used for quantization with 512 sequences of UltraChat.
+
+<!-- ## Deployment
+
+### Use with vLLM
+
+This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below.
+
+```python
+from vllm import LLM, SamplingParams
+from transformers import AutoTokenizer
+
+model_id = "neuralmagic/Meta-Llama-3.1-70B-FP8"
+number_gpus = 2
+
+sampling_params = SamplingParams(temperature=0.6, top_p=0.9, max_tokens=256)
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+
+messages = [
+    {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
+    {"role": "user", "content": "Who are you?"},
+]
+
+prompts = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
+
+llm = LLM(model=model_id, tensor_parallel_size=number_gpus)
+
+outputs = llm.generate(prompts, sampling_params)
+
+generated_text = outputs[0].outputs[0].text
+print(generated_text)
+```
+
+vLLM aslo supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details. -->
+
+## Creation
+
+This model was created by applying [LLM Compressor with calibration samples from UltraChat](https://github.com/vllm-project/llm-compressor/blob/sa/big_model_support/examples/big_model_offloading/big_model_w8a8_calibrate.py), as presented in the code snipet below.
+
+```python
+import torch
+from datasets import load_dataset
+from transformers import AutoTokenizer
+
+from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot
+from llmcompressor.transformers.compression.helpers import (
+    calculate_offload_device_map,
+    custom_offload_device_map,
+)
+
+recipe = """
+quant_stage:
+    quant_modifiers:
+        QuantizationModifier:
+            ignore: ["lm_head"]
+            config_groups:
+                group_0:
+                    weights:
+                        num_bits: 8
+                        type: float
+                        strategy: tensor
+                        dynamic: false
+                        symmetric: true
+                    input_activations:
+                        num_bits: 8
+                        type: float
+                        strategy: tensor
+                        dynamic: false
+                        symmetric: true
+                    targets: ["Linear"]
+"""
+
+model_stub = "meta-llama/Meta-Llama-3.1-70B"
+model_name = model_stub.split("/")[-1]
+
+device_map = calculate_offload_device_map(
+    model_stub, reserve_for_hessians=False, num_gpus=2, torch_dtype=torch.float16
+)
+
+model = SparseAutoModelForCausalLM.from_pretrained(
+    model_stub, torch_dtype=torch.float16, device_map=device_map
+)
+tokenizer = AutoTokenizer.from_pretrained(model_stub)
+
+output_dir = f"./{model_name}-FP8"
+
+DATASET_ID = "HuggingFaceH4/ultrachat_200k"
+DATASET_SPLIT = "train_sft"
+NUM_CALIBRATION_SAMPLES = 512
+MAX_SEQUENCE_LENGTH = 4096
+
+ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
+ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))
+
+def preprocess(example):
+    return {
+        "text": tokenizer.apply_chat_template(
+            example["messages"],
+            tokenize=False,
+        )
+    }
+
+ds = ds.map(preprocess)
+
+def tokenize(sample):
+    return tokenizer(
+        sample["text"],
+        padding=False,
+        max_length=MAX_SEQUENCE_LENGTH,
+        truncation=True,
+        add_special_tokens=False,
+    )
+
+ds = ds.map(tokenize, remove_columns=ds.column_names)
+
+oneshot(
+    model=model,
+    output_dir=output_dir,
+    dataset=ds,
+    recipe=recipe,
+    max_seq_length=MAX_SEQUENCE_LENGTH,
+    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
+    save_compressed=True,
+)
+```
+
+## Evaluation
+
+The model was evaluated on MMLU, ARC-Challenge, GSM-8K, Hellaswag, Winogrande and TruthfulQA.
+Evaluation was conducted using the Neural Magic fork of [lm-evaluation-harness](https://github.com/neuralmagic/lm-evaluation-harness/tree/llama_3.1_instruct) (branch llama_3.1_instruct) and the [vLLM](https://docs.vllm.ai/en/stable/) engine.
+This version of the lm-evaluation-harness includes versions of ARC-Challenge that matches the prompting style of [Meta-Llama-3.1-evals](https://huggingface.co/datasets/meta-llama/Meta-Llama-3.1-8B-evals).
+
+### Accuracy
+
+#### Open LLM Leaderboard evaluation scores
+<table>
+  <tr>
+   <td><strong>Benchmark</strong>
+   </td>
+   <td><strong>Meta-Llama-3.1-70B </strong>
+   </td>
+   <td><strong>Meta-Llama-3.1-70B-FP8(this model)</strong>
+   </td>
+   <td><strong>Recovery</strong>
+   </td>
+  </tr>
+  <tr>
+   <td>MMLU (5-shot)
+   </td>
+   <td>78.81
+   </td>
+   <td>78.85
+   </td>
+   <td>100.0%
+   </td>
+  </tr>
+  <tr>
+   <td>ARC Challenge (0-shot)
+   </td>
+   <td>93.43
+   </td>
+   <td>93.43
+   </td>
+   <td>100.0%
+   </td>
+  </tr>
+  <tr>
+   <td>GSM-8K (5-shot, strict-match)
+   </td>
+   <td>81.88
+   </td>
+   <td>81.35
+   </td>
+   <td>99.35%
+   </td>
+  </tr>
+  <tr>
+   <td>Hellaswag (10-shot)
+   </td>
+   <td>87.98
+   </td>
+   <td>87.82
+   </td>
+   <td>99.82%
+   </td>
+  </tr>
+  <tr>
+   <td>Winogrande (5-shot)
+   </td>
+   <td>85.78
+   </td>
+   <td>85.87
+   </td>
+   <td>100.1%
+   </td>
+  </tr>
+  <tr>
+   <td>TruthfulQA (0-shot)
+   </td>
+   <td>51.18
+   </td>
+   <td>50.90
+   </td>
+   <td>99.45%
+   </td>
+  </tr>
+  <tr>
+   <td><strong>Average</strong>
+   </td>
+   <td><strong>79.84</strong>
+   </td>
+   <td><strong>79.70</strong>
+   </td>
+   <td><strong>99.82%</strong>
+   </td>
+  </tr>
+</table>
+
+### Reproduction
+
+The results were obtained using the following commands:
+
+#### MMLU
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
+  --tasks mmlu \
+  --num_fewshot 5 \
+  --batch_size auto
+```
+
+#### ARC-Challenge
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
+  --tasks arc_challenge_llama_3.1_instruct \
+  --num_fewshot 25 \
+  --batch_size auto
+```
+
+#### GSM-8K
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
+  --tasks gsm8k \
+  --num_fewshot 5 \
+  --batch_size auto
+```
+
+#### Hellaswag
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
+  --tasks hellaswag \
+  --num_fewshot 10 \
+  --batch_size auto
+```
+
+#### Winogrande
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
+  --tasks winogrande \
+  --num_fewshot 5 \
+  --batch_size auto
+```
+
+#### TruthfulQA
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
+  --tasks truthfulqa_mc \
+  --num_fewshot 0 \
+  --batch_size auto
+```