INCModel1 commited on
Commit
0c4730a
·
verified ·
1 Parent(s): c6846d0

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +46 -0
README.md ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model:
4
+ - unsloth/DeepSeek-R1-BF16
5
+ ---
6
+ ## Model Details
7
+
8
+ This model card is for mxfp4 quantization of [unsloth/DeepSeek-R1-BF16](https://huggingface.co/unsloth/DeepSeek-R1-BF16) based on [intel/auto-round](https://github.com/intel/auto-round).
9
+ Please follow the license of the original model.
10
+
11
+ ## How to Use
12
+
13
+ The step-by-step README of quantization and evaluation can be found in [Intel Neural Compressor Examples](https://github.com/intel/neural-compressor/blob/master/examples/pytorch/nlp/huggingface_models/language-modeling/quantization/auto_round/deepseek/README.md).
14
+
15
+ ## Evaluate Results
16
+
17
+ | Task | backend | BF16 | MXFP4 |
18
+ |:-----------:|:-------:|:----------:|:----------:|
19
+ | hellaswag | vllm | 0.6903 | 0.6898 |
20
+ | piqa | vllm | 0.8319 | 0.8297 |
21
+ | mmlu | vllm | 0.8489 | 0.8426 |
22
+ | gsm8k | vllm | 0.9568 | 0.9553 |
23
+ | **average** | vllm | **0.8320** | **0.8294** |
24
+
25
+
26
+ ## Ethical Considerations and Limitations
27
+
28
+ The model can produce factually incorrect output, and should not be relied on to produce factually accurate information.
29
+ Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
30
+
31
+ Therefore, before deploying any applications of the model, developers should perform safety testing.
32
+
33
+ ## Caveats and Recommendations
34
+
35
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model.
36
+
37
+ Here are a couple of useful links to learn more about Intel's AI software:
38
+
39
+ - [Intel Neural Compressor](https://github.com/intel/neural-compressor)
40
+ - [AutoRound](https://github.com/intel/auto-round)
41
+
42
+ ## Disclaimer
43
+
44
+ The license on this model does not constitute legal advice.
45
+ We are not responsible for the actions of third parties who use this model.
46
+ Please consult an attorney before using this model for commercial purposes.