Adding Evaluation Results
#2
by
leaderboard-pr-bot - opened
README.md
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
---
|
| 2 |
-
license: other
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
- fr
|
|
@@ -7,21 +6,176 @@ language:
|
|
| 7 |
- hi
|
| 8 |
- zh
|
| 9 |
- code
|
| 10 |
-
|
| 11 |
datasets:
|
| 12 |
- HuggingFaceH4/no_robots
|
| 13 |
- mlabonne/guanaco-llama2-1k
|
| 14 |
- OpenAssistant/oasst_top1_2023-08-25
|
| 15 |
- totally-not-an-llm/EverythingLM-data-V3
|
|
|
|
| 16 |
widget:
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
---
|
| 26 |
The "microsoft/Orca-2-13b" model fully fine-tuned on HuggingFaceH4/no_robots, totally-not-an-llm/EverythingLM-data-V3, mlabonne/guanaco-llama2-1k, and OpenAssistant/oasst_top1_2023-08-25. This model achieved a test loss of 0.18.
|
| 27 |
|
|
@@ -41,4 +195,17 @@ repetition_penalty = 1.176
|
|
| 41 |
|
| 42 |
# Updates:
|
| 43 |
|
| 44 |
-
12/18/23 - 🔥 **This model holds the #5 position on the Open LLM Leaderboard among llama2-13b models.** 🔥
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
language:
|
| 3 |
- en
|
| 4 |
- fr
|
|
|
|
| 6 |
- hi
|
| 7 |
- zh
|
| 8 |
- code
|
| 9 |
+
license: other
|
| 10 |
datasets:
|
| 11 |
- HuggingFaceH4/no_robots
|
| 12 |
- mlabonne/guanaco-llama2-1k
|
| 13 |
- OpenAssistant/oasst_top1_2023-08-25
|
| 14 |
- totally-not-an-llm/EverythingLM-data-V3
|
| 15 |
+
base_model: microsoft/Orca-2-13b
|
| 16 |
widget:
|
| 17 |
+
- example_title: folklore
|
| 18 |
+
text: '<|im_start|>user
|
| 19 |
+
|
| 20 |
+
Are the stories of Nahanni Valley, like the decapitated men, true?
|
| 21 |
+
|
| 22 |
+
<|im_end|>
|
| 23 |
+
|
| 24 |
+
<|im_start|>assistant
|
| 25 |
+
|
| 26 |
+
'
|
| 27 |
+
output:
|
| 28 |
+
text: 'The Nahanni Valley in Canada is a remote and rugged region that has long
|
| 29 |
+
been shrouded in mystery. It is home to the South Nahanni River, which was designated
|
| 30 |
+
as a UNESCO World Heritage Site due to its exceptional natural beauty and biodiversity.
|
| 31 |
+
The valley''s history includes tales of gold prospectors, fur traders, and indigenous
|
| 32 |
+
populations, but it also features some unexplained events and legends. One such
|
| 33 |
+
story involves two prospectors who allegedly encountered an unknown creature
|
| 34 |
+
that led them into the forest before returning without their heads. This tale
|
| 35 |
+
has been passed down through generations, but there are no definitive answers
|
| 36 |
+
regarding its veracity.
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
It is important to note that these stories may have originated from local folklore
|
| 40 |
+
or were exaggerated over time. There is no concrete evidence to support the
|
| 41 |
+
existence of supernatural creatures or paranormal occurrences within the Nahanni
|
| 42 |
+
Valley. Nonetheless, the area remains a captivating destination for those seeking
|
| 43 |
+
adventure and exploration. If you have any questions about this topic or need
|
| 44 |
+
assistance related to traveling to the Nahanni Valley, please feel free to ask!'
|
| 45 |
+
- example_title: Alan Turing
|
| 46 |
+
text: '<|im_start|>user
|
| 47 |
+
|
| 48 |
+
Why was Alan Turing castrated?
|
| 49 |
+
|
| 50 |
+
<|im_end|>
|
| 51 |
+
|
| 52 |
+
<|im_start|>assistant
|
| 53 |
+
|
| 54 |
+
'
|
| 55 |
+
output:
|
| 56 |
+
text: 'Alan Turing, a renowned mathematician and computer scientist who played
|
| 57 |
+
a crucial role in breaking the German Enigma code during World War II, faced
|
| 58 |
+
significant challenges due to his homosexuality. In 1952, he was convicted of
|
| 59 |
+
"gross indecency" for engaging in a consensual relationship with another man,
|
| 60 |
+
which led to him being sentenced to hormone therapy designed to reduce his sex
|
| 61 |
+
drive. This treatment included the administration of estrogen, which had the
|
| 62 |
+
side effect of causing breast growth.
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
In an effort to avoid this unwanted physical change, Turing requested that his
|
| 66 |
+
doctors cease the hormone treatments. However, they ignored his request and
|
| 67 |
+
continued the therapy without his consent. As a result, Turing''s breasts began
|
| 68 |
+
to grow, leading him to undergo surgery to remove them. The operation, known
|
| 69 |
+
as a mastectomy, caused considerable pain and discomfort, further contributing
|
| 70 |
+
to his emotional turmoil.
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
It is important to note that Turing''s castration was not directly carried out
|
| 74 |
+
by the authorities but rather resulted from the unintended consequences of the
|
| 75 |
+
forced hormone therapy.'
|
| 76 |
+
model-index:
|
| 77 |
+
- name: Orca-2-13b-SFT-v4
|
| 78 |
+
results:
|
| 79 |
+
- task:
|
| 80 |
+
type: text-generation
|
| 81 |
+
name: Text Generation
|
| 82 |
+
dataset:
|
| 83 |
+
name: AI2 Reasoning Challenge (25-Shot)
|
| 84 |
+
type: ai2_arc
|
| 85 |
+
config: ARC-Challenge
|
| 86 |
+
split: test
|
| 87 |
+
args:
|
| 88 |
+
num_few_shot: 25
|
| 89 |
+
metrics:
|
| 90 |
+
- type: acc_norm
|
| 91 |
+
value: 59.22
|
| 92 |
+
name: normalized accuracy
|
| 93 |
+
source:
|
| 94 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/Orca-2-13b-SFT-v4
|
| 95 |
+
name: Open LLM Leaderboard
|
| 96 |
+
- task:
|
| 97 |
+
type: text-generation
|
| 98 |
+
name: Text Generation
|
| 99 |
+
dataset:
|
| 100 |
+
name: HellaSwag (10-Shot)
|
| 101 |
+
type: hellaswag
|
| 102 |
+
split: validation
|
| 103 |
+
args:
|
| 104 |
+
num_few_shot: 10
|
| 105 |
+
metrics:
|
| 106 |
+
- type: acc_norm
|
| 107 |
+
value: 79.58
|
| 108 |
+
name: normalized accuracy
|
| 109 |
+
source:
|
| 110 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/Orca-2-13b-SFT-v4
|
| 111 |
+
name: Open LLM Leaderboard
|
| 112 |
+
- task:
|
| 113 |
+
type: text-generation
|
| 114 |
+
name: Text Generation
|
| 115 |
+
dataset:
|
| 116 |
+
name: MMLU (5-Shot)
|
| 117 |
+
type: cais/mmlu
|
| 118 |
+
config: all
|
| 119 |
+
split: test
|
| 120 |
+
args:
|
| 121 |
+
num_few_shot: 5
|
| 122 |
+
metrics:
|
| 123 |
+
- type: acc
|
| 124 |
+
value: 60.23
|
| 125 |
+
name: accuracy
|
| 126 |
+
source:
|
| 127 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/Orca-2-13b-SFT-v4
|
| 128 |
+
name: Open LLM Leaderboard
|
| 129 |
+
- task:
|
| 130 |
+
type: text-generation
|
| 131 |
+
name: Text Generation
|
| 132 |
+
dataset:
|
| 133 |
+
name: TruthfulQA (0-shot)
|
| 134 |
+
type: truthful_qa
|
| 135 |
+
config: multiple_choice
|
| 136 |
+
split: validation
|
| 137 |
+
args:
|
| 138 |
+
num_few_shot: 0
|
| 139 |
+
metrics:
|
| 140 |
+
- type: mc2
|
| 141 |
+
value: 51.15
|
| 142 |
+
source:
|
| 143 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/Orca-2-13b-SFT-v4
|
| 144 |
+
name: Open LLM Leaderboard
|
| 145 |
+
- task:
|
| 146 |
+
type: text-generation
|
| 147 |
+
name: Text Generation
|
| 148 |
+
dataset:
|
| 149 |
+
name: Winogrande (5-shot)
|
| 150 |
+
type: winogrande
|
| 151 |
+
config: winogrande_xl
|
| 152 |
+
split: validation
|
| 153 |
+
args:
|
| 154 |
+
num_few_shot: 5
|
| 155 |
+
metrics:
|
| 156 |
+
- type: acc
|
| 157 |
+
value: 80.03
|
| 158 |
+
name: accuracy
|
| 159 |
+
source:
|
| 160 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/Orca-2-13b-SFT-v4
|
| 161 |
+
name: Open LLM Leaderboard
|
| 162 |
+
- task:
|
| 163 |
+
type: text-generation
|
| 164 |
+
name: Text Generation
|
| 165 |
+
dataset:
|
| 166 |
+
name: GSM8k (5-shot)
|
| 167 |
+
type: gsm8k
|
| 168 |
+
config: main
|
| 169 |
+
split: test
|
| 170 |
+
args:
|
| 171 |
+
num_few_shot: 5
|
| 172 |
+
metrics:
|
| 173 |
+
- type: acc
|
| 174 |
+
value: 28.28
|
| 175 |
+
name: accuracy
|
| 176 |
+
source:
|
| 177 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/Orca-2-13b-SFT-v4
|
| 178 |
+
name: Open LLM Leaderboard
|
| 179 |
---
|
| 180 |
The "microsoft/Orca-2-13b" model fully fine-tuned on HuggingFaceH4/no_robots, totally-not-an-llm/EverythingLM-data-V3, mlabonne/guanaco-llama2-1k, and OpenAssistant/oasst_top1_2023-08-25. This model achieved a test loss of 0.18.
|
| 181 |
|
|
|
|
| 195 |
|
| 196 |
# Updates:
|
| 197 |
|
| 198 |
+
12/18/23 - 🔥 **This model holds the #5 position on the Open LLM Leaderboard among llama2-13b models.** 🔥
|
| 199 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 200 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Locutusque__Orca-2-13b-SFT-v4)
|
| 201 |
+
|
| 202 |
+
| Metric |Value|
|
| 203 |
+
|---------------------------------|----:|
|
| 204 |
+
|Avg. |59.75|
|
| 205 |
+
|AI2 Reasoning Challenge (25-Shot)|59.22|
|
| 206 |
+
|HellaSwag (10-Shot) |79.58|
|
| 207 |
+
|MMLU (5-Shot) |60.23|
|
| 208 |
+
|TruthfulQA (0-shot) |51.15|
|
| 209 |
+
|Winogrande (5-shot) |80.03|
|
| 210 |
+
|GSM8k (5-shot) |28.28|
|
| 211 |
+
|