Commit
2923fee
1 Parent(s): 02ad156

Adding Evaluation Results (#1)

Browse files

- Adding Evaluation Results (198b9cba4aea9efab62dd9b78bc77c16ea51cdf0)


Co-authored-by: Open LLM Leaderboard PR Bot <[email protected]>

Files changed (1) hide show
  1. README.md +26 -19
README.md CHANGED
@@ -1,14 +1,18 @@
1
  ---
2
- inference: false
 
 
 
3
  tags:
4
  - merge
5
  - mergekit
6
  - lazymergekit
 
7
  base_model:
8
  - senseable/Westlake-7B
9
  - Guilherme34/Samantha-v2
10
  - uukuguy/speechless-mistral-six-in-one-7b
11
- license: apache-2.0
12
  model-index:
13
  - name: sethuiyer/Nandine-7b
14
  results:
@@ -27,8 +31,7 @@ model-index:
27
  value: 69.28
28
  name: normalized accuracy
29
  source:
30
- url: >-
31
- https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
32
  name: Open LLM Leaderboard
33
  - task:
34
  type: text-generation
@@ -44,8 +47,7 @@ model-index:
44
  value: 87.01
45
  name: normalized accuracy
46
  source:
47
- url: >-
48
- https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
49
  name: Open LLM Leaderboard
50
  - task:
51
  type: text-generation
@@ -62,8 +64,7 @@ model-index:
62
  value: 64.83
63
  name: accuracy
64
  source:
65
- url: >-
66
- https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
67
  name: Open LLM Leaderboard
68
  - task:
69
  type: text-generation
@@ -79,8 +80,7 @@ model-index:
79
  - type: mc2
80
  value: 62.1
81
  source:
82
- url: >-
83
- https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
84
  name: Open LLM Leaderboard
85
  - task:
86
  type: text-generation
@@ -97,8 +97,7 @@ model-index:
97
  value: 83.19
98
  name: accuracy
99
  source:
100
- url: >-
101
- https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
102
  name: Open LLM Leaderboard
103
  - task:
104
  type: text-generation
@@ -115,13 +114,8 @@ model-index:
115
  value: 62.4
116
  name: accuracy
117
  source:
118
- url: >-
119
- https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
120
  name: Open LLM Leaderboard
121
- language:
122
- - en
123
- library_name: transformers
124
- pipeline_tag: text-generation
125
  ---
126
 
127
  # Nandine-7b
@@ -225,4 +219,17 @@ GGUF files are available at [Nandine-7b-GGUF](https://huggingface.co/sethuiyer/N
225
  ## Ollama
226
  Nandine is now available on Ollama. You can use it by running the command ```ollama run stuehieyr/nandine``` in your
227
  terminal. If you have limited computing resources, check out this [video](https://www.youtube.com/watch?v=Qa1h7ygwQq8) to learn how to run it on
228
- a Google Colab backend.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
+ license: apache-2.0
5
+ library_name: transformers
6
  tags:
7
  - merge
8
  - mergekit
9
  - lazymergekit
10
+ inference: false
11
  base_model:
12
  - senseable/Westlake-7B
13
  - Guilherme34/Samantha-v2
14
  - uukuguy/speechless-mistral-six-in-one-7b
15
+ pipeline_tag: text-generation
16
  model-index:
17
  - name: sethuiyer/Nandine-7b
18
  results:
 
31
  value: 69.28
32
  name: normalized accuracy
33
  source:
34
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
 
35
  name: Open LLM Leaderboard
36
  - task:
37
  type: text-generation
 
47
  value: 87.01
48
  name: normalized accuracy
49
  source:
50
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
 
51
  name: Open LLM Leaderboard
52
  - task:
53
  type: text-generation
 
64
  value: 64.83
65
  name: accuracy
66
  source:
67
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
 
68
  name: Open LLM Leaderboard
69
  - task:
70
  type: text-generation
 
80
  - type: mc2
81
  value: 62.1
82
  source:
83
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
 
84
  name: Open LLM Leaderboard
85
  - task:
86
  type: text-generation
 
97
  value: 83.19
98
  name: accuracy
99
  source:
100
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
 
101
  name: Open LLM Leaderboard
102
  - task:
103
  type: text-generation
 
114
  value: 62.4
115
  name: accuracy
116
  source:
117
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=sethuiyer/Nandine-7b
 
118
  name: Open LLM Leaderboard
 
 
 
 
119
  ---
120
 
121
  # Nandine-7b
 
219
  ## Ollama
220
  Nandine is now available on Ollama. You can use it by running the command ```ollama run stuehieyr/nandine``` in your
221
  terminal. If you have limited computing resources, check out this [video](https://www.youtube.com/watch?v=Qa1h7ygwQq8) to learn how to run it on
222
+ a Google Colab backend.
223
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
224
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_sethuiyer__Nandine-7b)
225
+
226
+ | Metric |Value|
227
+ |---------------------------------|----:|
228
+ |Avg. |71.47|
229
+ |AI2 Reasoning Challenge (25-Shot)|69.28|
230
+ |HellaSwag (10-Shot) |87.01|
231
+ |MMLU (5-Shot) |64.83|
232
+ |TruthfulQA (0-shot) |62.10|
233
+ |Winogrande (5-shot) |83.19|
234
+ |GSM8k (5-shot) |62.40|
235
+