初始化项目，由ModelHub XC社区提供模型
Model: RedHatAI/SmolLM-135M-Instruct-quantized.w8a16 Source: Original Platform
This commit is contained in:
ModelHub XC
212
README.md
Normal file
212
README.md
Normal file
@@ -0,0 +1,212 @@
|
||||
---
|
||||
library_name: transformers
|
||||
license: apache-2.0
|
||||
language:
|
||||
- en
|
||||
pipeline_tag: text-generation
|
||||
tags:
|
||||
- int8
|
||||
- vllm
|
||||
base_model: HuggingFaceTB/SmolLM-135M-Instruct
|
||||
---
|
||||
|
||||
# SmolLM-135M-Instruct-quantized.w8a16
|
||||
|
||||
## Model Overview
|
||||
- **Model Architecture:** Llama
|
||||
- **Input:** Text
|
||||
- **Output:** Text
|
||||
- **Model Optimizations:**
|
||||
- **Weight quantization:** INT8
|
||||
- **Intended Use Cases:** Intended for commercial and research use in English. Similarly to [SmolLM-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-135M-Instruct), this models is intended for assistant-like chat.
|
||||
- **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:** 8/22/2024
|
||||
- **Version:** 1.0
|
||||
- **License(s):** [Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)
|
||||
- **Model Developers:** Neural Magic
|
||||
|
||||
Quantized version of [SmolLM-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-135M-Instruct).
|
||||
It achieves an average score of 31.98 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 31.66.
|
||||
|
||||
### Model Optimizations
|
||||
|
||||
This model was obtained by quantizing the weights of [SmolLM-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM-135M-Instruct) to INT8 data type.
|
||||
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 of the linear operators within transformers blocks are quantized. Symmetric per-channel quantization is applied, in which a linear scaling per output dimension maps the INT8 and floating point representations of the quantized weights.
|
||||
The [GPTQ](https://arxiv.org/abs/2210.17323) algorithm is applied for quantization, as implemented in the [llm-compressor](https://github.com/vllm-project/llm-compressor) library.
|
||||
GPTQ used a 1% damping factor and 1,024 sequences of 2,048 random tokens.
|
||||
|
||||
## 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/SmolLM-135M-Instruct-quantized.w8a16"
|
||||
|
||||
sampling_params = SamplingParams(temperature=0.6, top_p=0.92, max_tokens=100)
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
||||
|
||||
messages = [
|
||||
{"role": "user", "content": "List the steps to bake a chocolate cake from scratch."},
|
||||
]
|
||||
|
||||
prompts = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
||||
|
||||
llm = LLM(model=model_id)
|
||||
|
||||
outputs = llm.generate(prompts, sampling_params)
|
||||
|
||||
generated_text = outputs[0].outputs[0].text
|
||||
print(generated_text)
|
||||
```
|
||||
|
||||
vLLM also supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details.
|
||||
|
||||
## Creation
|
||||
|
||||
This model was created by using the [llm-compressor](https://github.com/vllm-project/llm-compressor) library as presented in the code snipet below.
|
||||
|
||||
```python
|
||||
from transformers import AutoTokenizer
|
||||
from datasets import Dataset
|
||||
from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot
|
||||
from llmcompressor.modifiers.quantization import GPTQModifier
|
||||
import random
|
||||
|
||||
model_id = "HuggingFaceTB/SmolLM-135M-Instruct"
|
||||
|
||||
num_samples = 1024
|
||||
max_seq_len = 2048
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
||||
|
||||
max_token_id = len(tokenizer.get_vocab()) - 1
|
||||
input_ids = [[random.randint(0, max_token_id) for _ in range(max_seq_len)] for _ in range(num_samples)]
|
||||
attention_mask = num_samples * [max_seq_len * [1]]
|
||||
ds = Dataset.from_dict({"input_ids": input_ids, "attention_mask": attention_mask})
|
||||
|
||||
recipe = GPTQModifier(
|
||||
targets="Linear",
|
||||
scheme="W8A16",
|
||||
ignore=["lm_head"],
|
||||
dampening_frac=0.01,
|
||||
)
|
||||
|
||||
model = SparseAutoModelForCausalLM.from_pretrained(
|
||||
model_id,
|
||||
device_map="auto",
|
||||
)
|
||||
|
||||
oneshot(
|
||||
model=model,
|
||||
dataset=ds,
|
||||
recipe=recipe,
|
||||
max_seq_length=max_seq_len,
|
||||
num_calibration_samples=num_samples,
|
||||
)
|
||||
model.save_pretrained("SmolLM-135M-Instruct-quantized.w8a16")
|
||||
```
|
||||
|
||||
## Evaluation
|
||||
|
||||
The model was evaluated on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) leaderboard tasks (version 1) with the [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness/tree/383bbd54bc621086e05aa1b030d8d4d5635b25e6) (commit 383bbd54bc621086e05aa1b030d8d4d5635b25e6) and the [vLLM](https://docs.vllm.ai/en/stable/) engine, using the following command:
|
||||
```
|
||||
lm_eval \
|
||||
--model vllm \
|
||||
--model_args pretrained="neuralmagic/SmolLM-135M-Instruct-quantized.w8a16",dtype=auto,gpu_memory_utilization=0.4,add_bos_token=True,max_model_len=4096 \
|
||||
--tasks openllm \
|
||||
--batch_size auto
|
||||
```
|
||||
|
||||
### Accuracy
|
||||
|
||||
#### Open LLM Leaderboard evaluation scores
|
||||
<table>
|
||||
<tr>
|
||||
<td><strong>Benchmark</strong>
|
||||
</td>
|
||||
<td><strong>SmolLM-135M-Instruct-quantized</strong>
|
||||
</td>
|
||||
<td><strong>SmolLM-135M-Instruct-quantized.w8a16 (this model)</strong>
|
||||
</td>
|
||||
<td><strong>Recovery</strong>
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>MMLU (5-shot)
|
||||
</td>
|
||||
<td>26.46
|
||||
</td>
|
||||
<td>26.80
|
||||
</td>
|
||||
<td>101.3%
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>ARC Challenge (25-shot)
|
||||
</td>
|
||||
<td>31.83
|
||||
</td>
|
||||
<td>31.74
|
||||
</td>
|
||||
<td>99.7%
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>GSM-8K (5-shot, strict-match)
|
||||
</td>
|
||||
<td>0.61
|
||||
</td>
|
||||
<td>0.68
|
||||
</td>
|
||||
<td>112.5%
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>Hellaswag (10-shot)
|
||||
</td>
|
||||
<td>40.50
|
||||
</td>
|
||||
<td>40.54
|
||||
</td>
|
||||
<td>100.1%
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>Winogrande (5-shot)
|
||||
</td>
|
||||
<td>52.41
|
||||
</td>
|
||||
<td>52.72
|
||||
</td>
|
||||
<td>100.6%
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>TruthfulQA (0-shot)
|
||||
</td>
|
||||
<td>39.38
|
||||
</td>
|
||||
<td>39.41
|
||||
</td>
|
||||
<td>100.1%
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td><strong>Average</strong>
|
||||
</td>
|
||||
<td><strong>31.86</strong>
|
||||
</td>
|
||||
<td><strong>31.98</strong>
|
||||
</td>
|
||||
<td><strong>100.4%</strong>
|
||||
</td>
|
||||
</tr>
|
||||
</table>
|
||||
Reference in New Issue
Block a user