Hugging Face's logo

RichardErkhov
/
moreh_-_MoMo-72B-LoRA-V1.4-gguf

GGUF
Inference Endpoints
Model card Files Community
RichardErkhov commited on
Commit
9027490
•
1 Parent(s): a919308

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +86 -0
README.md ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ MoMo-72B-LoRA-V1.4 - GGUF
11
+ - Model creator: https://huggingface.co/moreh/
12
+ - Original model: https://huggingface.co/moreh/MoMo-72B-LoRA-V1.4/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [MoMo-72B-LoRA-V1.4.Q2_K.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.Q2_K.gguf) | Q2_K | 25.22GB |
18
+ | [MoMo-72B-LoRA-V1.4.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.IQ3_XS.gguf) | IQ3_XS | 27.88GB |
19
+ | [MoMo-72B-LoRA-V1.4.IQ3_S.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.IQ3_S.gguf) | IQ3_S | 29.4GB |
20
+ | [MoMo-72B-LoRA-V1.4.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.Q3_K_S.gguf) | Q3_K_S | 29.4GB |
21
+ | [MoMo-72B-LoRA-V1.4.IQ3_M.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.IQ3_M.gguf) | IQ3_M | 30.98GB |
22
+ | [MoMo-72B-LoRA-V1.4.Q3_K.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.Q3_K.gguf) | Q3_K | 32.85GB |
23
+ | [MoMo-72B-LoRA-V1.4.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.Q3_K_M.gguf) | Q3_K_M | 32.85GB |
24
+ | [MoMo-72B-LoRA-V1.4.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.Q3_K_L.gguf) | Q3_K_L | 35.85GB |
25
+ | [MoMo-72B-LoRA-V1.4.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/blob/main/MoMo-72B-LoRA-V1.4.IQ4_XS.gguf) | IQ4_XS | 36.41GB |
26
+ | [MoMo-72B-LoRA-V1.4.Q4_0.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q4_0 | 38.19GB |
27
+ | [MoMo-72B-LoRA-V1.4.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | IQ4_NL | 38.42GB |
28
+ | [MoMo-72B-LoRA-V1.4.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q4_K_S | 38.45GB |
29
+ | [MoMo-72B-LoRA-V1.4.Q4_K.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q4_K | 40.77GB |
30
+ | [MoMo-72B-LoRA-V1.4.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q4_K_M | 40.77GB |
31
+ | [MoMo-72B-LoRA-V1.4.Q4_1.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q4_1 | 42.32GB |
32
+ | [MoMo-72B-LoRA-V1.4.Q5_0.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q5_0 | 46.46GB |
33
+ | [MoMo-72B-LoRA-V1.4.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q5_K_S | 46.46GB |
34
+ | [MoMo-72B-LoRA-V1.4.Q5_K.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q5_K | 47.79GB |
35
+ | [MoMo-72B-LoRA-V1.4.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q5_K_M | 47.79GB |
36
+ | [MoMo-72B-LoRA-V1.4.Q5_1.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q5_1 | 50.59GB |
37
+ | [MoMo-72B-LoRA-V1.4.Q6_K.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q6_K | 55.24GB |
38
+ | [MoMo-72B-LoRA-V1.4.Q8_0.gguf](https://huggingface.co/RichardErkhov/moreh_-_MoMo-72B-LoRA-V1.4-gguf/tree/main/) | Q8_0 | 71.55GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ license: mit
46
+ language:
47
+ - en
48
+ ---
49
+ # **Introduction**
50
+ MoMo-72B is trained via Supervised Fine-Tuning (SFT) using [LoRA](https://arxiv.org/abs/2106.09685), with the QWEN-72B model as its base-model.
51
+ Note that we did not exploit any form of weight merge.
52
+ For leaderboard submission, the trained weight is realigned for compatibility with llama.
53
+ MoMo-72B is trained using **[Moreh](https://moreh.io/)**'s [MoAI platform](https://moreh.io/product), which simplifies the training of large-scale models, and AMD's MI250 GPU.
54
+
55
+
56
+ ## Details
57
+ ### Used Librarys
58
+ - torch
59
+ - peft
60
+ ### Used Datasets
61
+ - Open-Orca/SlimOrca
62
+ - No other dataset was used
63
+ - No benchmark test set or the training set are used
64
+ - [data contamination check](https://github.com/swj0419/detect-pretrain-code-contamination) result
65
+
66
+ | Model | ARC | MMLU | TruthfulQA | GSM8K |
67
+ |------------------------------|-------|-------|-------|-------|
68
+ | **V1.4(result < 0.1, %)**| TBU |0.73 | 0.71 | TBU |
69
+ ### Used Environments
70
+ - AMD MI250 & MoAI platform
71
+ - Please visit https://moreh.io/product for more information about MoAI platform
72
+ - Or, contact us directly [contact@moreh.io](mailto:contact@moreh.io)
73
+
74
+ ## How to use
75
+
76
+ ```python
77
+ # pip install transformers==4.35.2
78
+ import torch
79
+ from transformers import AutoModelForCausalLM, AutoTokenizer
80
+
81
+ tokenizer = AutoTokenizer.from_pretrained("moreh/MoMo-72B-LoRA-V1.4")
82
+ model = AutoModelForCausalLM.from_pretrained(
83
+ "moreh/MoMo-72B-LoRA-V1.4"
84
+ )
85
+ ```
86
+