Hugging Face's logo

RichardErkhov
/
IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf

GGUF
Inference Endpoints
conversational
Model card Files Community
RichardErkhov commited on
Commit
4745868
·
verified ·
1 Parent(s): 8f202ce

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +84 -0
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ sqft-phi-3-mini-4k-50-base - GGUF
11
+ - Model creator: https://huggingface.co/IntelLabs/
12
+ - Original model: https://huggingface.co/IntelLabs/sqft-phi-3-mini-4k-50-base/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [sqft-phi-3-mini-4k-50-base.Q2_K.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q2_K.gguf) | Q2_K | 1.32GB |
18
+ | [sqft-phi-3-mini-4k-50-base.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.IQ3_XS.gguf) | IQ3_XS | 1.51GB |
19
+ | [sqft-phi-3-mini-4k-50-base.IQ3_S.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.IQ3_S.gguf) | IQ3_S | 1.57GB |
20
+ | [sqft-phi-3-mini-4k-50-base.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q3_K_S.gguf) | Q3_K_S | 1.57GB |
21
+ | [sqft-phi-3-mini-4k-50-base.IQ3_M.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.IQ3_M.gguf) | IQ3_M | 1.73GB |
22
+ | [sqft-phi-3-mini-4k-50-base.Q3_K.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q3_K.gguf) | Q3_K | 1.82GB |
23
+ | [sqft-phi-3-mini-4k-50-base.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q3_K_M.gguf) | Q3_K_M | 1.82GB |
24
+ | [sqft-phi-3-mini-4k-50-base.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q3_K_L.gguf) | Q3_K_L | 1.94GB |
25
+ | [sqft-phi-3-mini-4k-50-base.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.IQ4_XS.gguf) | IQ4_XS | 1.93GB |
26
+ | [sqft-phi-3-mini-4k-50-base.Q4_0.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q4_0.gguf) | Q4_0 | 2.03GB |
27
+ | [sqft-phi-3-mini-4k-50-base.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.IQ4_NL.gguf) | IQ4_NL | 2.04GB |
28
+ | [sqft-phi-3-mini-4k-50-base.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q4_K_S.gguf) | Q4_K_S | 2.04GB |
29
+ | [sqft-phi-3-mini-4k-50-base.Q4_K.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q4_K.gguf) | Q4_K | 2.23GB |
30
+ | [sqft-phi-3-mini-4k-50-base.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q4_K_M.gguf) | Q4_K_M | 2.23GB |
31
+ | [sqft-phi-3-mini-4k-50-base.Q4_1.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q4_1.gguf) | Q4_1 | 2.24GB |
32
+ | [sqft-phi-3-mini-4k-50-base.Q5_0.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q5_0.gguf) | Q5_0 | 2.46GB |
33
+ | [sqft-phi-3-mini-4k-50-base.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q5_K_S.gguf) | Q5_K_S | 2.46GB |
34
+ | [sqft-phi-3-mini-4k-50-base.Q5_K.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q5_K.gguf) | Q5_K | 2.62GB |
35
+ | [sqft-phi-3-mini-4k-50-base.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q5_K_M.gguf) | Q5_K_M | 2.62GB |
36
+ | [sqft-phi-3-mini-4k-50-base.Q5_1.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q5_1.gguf) | Q5_1 | 2.68GB |
37
+ | [sqft-phi-3-mini-4k-50-base.Q6_K.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q6_K.gguf) | Q6_K | 2.92GB |
38
+ | [sqft-phi-3-mini-4k-50-base.Q8_0.gguf](https://huggingface.co/RichardErkhov/IntelLabs_-_sqft-phi-3-mini-4k-50-base-gguf/blob/main/sqft-phi-3-mini-4k-50-base.Q8_0.gguf) | Q8_0 | 3.78GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ language: en
46
+ license: apache-2.0
47
+ ---
48
+
49
+ # SQFT Base Model: sqft-phi-3-mini-4k-50-base
50
+
51
+ - Source Model: [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct)
52
+ - Sparse Method: [Wanda](https://github.com/locuslab/wanda)
53
+ - Sparsity: 50%
54
+ - Quantization: No
55
+
56
+ ## Model Sources
57
+
58
+ - **Repository:** [https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SQFT](https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SQFT)
59
+ - **Paper:** [SQFT: Low-cost Model Adaptation in Low-precision Sparse Foundation Models]()
60
+
61
+ ## How to get this model
62
+
63
+ Refer to the command in [SQFT/run_command/phi-3-mini-4k-instruct/sparse_quantization.sh#11](https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SQFT/run_command/phi-3-mini-4k-instruct/sparse_quantization.sh#11).
64
+
65
+ ## Citation
66
+
67
+ ```bash
68
+ @article{munoz2024sqft,
69
+ title = {SQFT: Low-cost Model Adaptation in Low-precision Sparse Foundation Models},
70
+ author={J. Pablo Munoz and Jinjie Yuan and Nilesh Jain},
71
+ journal={},
72
+ year={2024}
73
+ }
74
+ ```
75
+
76
+ ## Acknowledgement
77
+
78
+ Thanks to the work Wanda ([paper](https://arxiv.org/abs/2306.11695), [code](https://github.com/locuslab/wanda)), which provides a simple but effective pruning approach.
79
+
80
+ ## License
81
+
82
+ Apache-2.0
83
+
84
+