File size: 4,742 Bytes
39d2dce
c628b81
 
77acdb8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39d2dce
113b76d
e5dd511
40538e3
a7e19b1
e5dd511
40538e3
c628b81
113b76d
 
 
 
c628b81
113b76d
 
c628b81
60c71ea
 
 
 
e5b09fa
113b76d
 
 
 
 
 
c628b81
113b76d
 
 
 
c628b81
e5dd511
113b76d
e5dd511
113b76d
77acdb8
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
---
language:
- en
license: mit
model-index:
- name: MoMo-70B-LoRA-V1.4
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 69.2
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-LoRA-V1.4
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 85.07
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-LoRA-V1.4
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 77.12
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-LoRA-V1.4
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 62.66
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-LoRA-V1.4
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 83.74
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-LoRA-V1.4
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 70.2
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=moreh/MoMo-70B-LoRA-V1.4
      name: Open LLM Leaderboard
---
# **Introduction**
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.  
Note that we did not exploit any form of weight merge.  
For leaderboard submission, the trained weight is realigned for compatibility with llama.  
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.


## Details
### Used Librarys
- torch
- peft
### Used Datasets
- Open-Orca/SlimOrca
- No other dataset was used
- No benchmark test set or the training set are used
  - [data contamination check](https://github.com/swj0419/detect-pretrain-code-contamination) result
    
| Model                        | ARC   | MMLU | TruthfulQA | GSM8K |
|------------------------------|-------|-------|-------|-------|
| **V1.4(result < 0.1, %)**| TBU |0.73 | 0.71 | TBU |
### Used Environments
- AMD MI250 & MoAI platform
- Please visit https://moreh.io/product for more information about MoAI platform
- Or, contact us directly [contact@moreh.io](mailto:contact@moreh.io)

## How to use

```python
# pip install transformers==4.35.2
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("moreh/MoMo-72B-LoRA-V1.4")
model = AutoModelForCausalLM.from_pretrained(
    "moreh/MoMo-72B-LoRA-V1.4"
)
```
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_moreh__MoMo-70B-LoRA-V1.4)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |74.67|
|AI2 Reasoning Challenge (25-Shot)|69.20|
|HellaSwag (10-Shot)              |85.07|
|MMLU (5-Shot)                    |77.12|
|TruthfulQA (0-shot)              |62.66|
|Winogrande (5-shot)              |83.74|
|GSM8k (5-shot)                   |70.20|