Commit
•
3fc77d3
1
Parent(s):
4814575
Adding Evaluation Results (#1)
Browse files- Adding Evaluation Results (d4540048eab11a55d6e3381ed8d41d34277bd896)
Co-authored-by: Open LLM Leaderboard PR Bot <[email protected]>
README.md
CHANGED
@@ -1,83 +1,194 @@
|
|
1 |
---
|
2 |
-
library_name: transformers
|
3 |
-
license: mit
|
4 |
language:
|
5 |
- en
|
6 |
- vi
|
7 |
-
|
8 |
-
|
9 |
tags:
|
10 |
- ghost
|
11 |
-
|
12 |
-
|
13 |
-
results:
|
14 |
-
- task:
|
15 |
-
type: text-generation
|
16 |
-
dataset:
|
17 |
-
type: vmlu_v1.5
|
18 |
-
name: VMLU
|
19 |
-
metrics:
|
20 |
-
- name: Average
|
21 |
-
type: avg
|
22 |
-
value: 36.06
|
23 |
-
verified: true
|
24 |
-
- name: STEM
|
25 |
-
type: stem
|
26 |
-
value: 33.54
|
27 |
-
verified: true
|
28 |
-
- name: Social science
|
29 |
-
type: ss
|
30 |
-
value: 38.74
|
31 |
-
verified: true
|
32 |
-
- name: Humanities
|
33 |
-
type: hm
|
34 |
-
value: 37.15
|
35 |
-
verified: true
|
36 |
-
- name: Other
|
37 |
-
type: ot
|
38 |
-
value: 36.78
|
39 |
-
verified: true
|
40 |
-
- task:
|
41 |
-
type: text-generation
|
42 |
-
dataset:
|
43 |
-
type: open_llm_leaderboard
|
44 |
-
name: Open LLM Leaderboard
|
45 |
-
source:
|
46 |
-
name: Open LLM Leaderboard
|
47 |
-
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
48 |
-
metrics:
|
49 |
-
- name: Average
|
50 |
-
type: avg
|
51 |
-
value: 56.89
|
52 |
-
verified: true
|
53 |
-
- name: ARC
|
54 |
-
type: arc
|
55 |
-
value: 53.07
|
56 |
-
verified: true
|
57 |
-
- name: HellaSwag
|
58 |
-
type: hs
|
59 |
-
value: 77.93
|
60 |
-
verified: true
|
61 |
-
- name: HellaSwag
|
62 |
-
type: hs
|
63 |
-
value: 77.93
|
64 |
-
verified: true
|
65 |
-
- name: MMLU
|
66 |
-
type: mmlu
|
67 |
-
value: 55.09
|
68 |
-
verified: true
|
69 |
-
- name: Winogrande
|
70 |
-
type: wg
|
71 |
-
value: 73.72
|
72 |
-
verified: true
|
73 |
-
- name: GSM8K
|
74 |
-
type: gsm8k
|
75 |
-
value: 33.74
|
76 |
-
verified: true
|
77 |
widget:
|
78 |
-
|
79 |
-
|
80 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
81 |
---
|
82 |
|
83 |
# Model Card for Model ID
|
@@ -291,4 +402,17 @@ Many thanks for
|
|
291 |
|
292 |
## Model Card Contact
|
293 |
|
294 |
-
**Lam H** ([email protected])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
|
|
|
|
2 |
language:
|
3 |
- en
|
4 |
- vi
|
5 |
+
license: mit
|
6 |
+
library_name: transformers
|
7 |
tags:
|
8 |
- ghost
|
9 |
+
pipeline_tag: text-generation
|
10 |
+
base_model: HuggingFaceH4/zephyr-7b-beta
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
widget:
|
12 |
+
- text: '<|system|>
|
13 |
+
|
14 |
+
You are a helpful assistant.</s>
|
15 |
+
|
16 |
+
<|user|>
|
17 |
+
|
18 |
+
Thông tin về Peristernia despecta</s>
|
19 |
+
|
20 |
+
<|assistant|>
|
21 |
+
|
22 |
+
'
|
23 |
+
output:
|
24 |
+
text: Peristernia despecta là một loài ốc biển, là động vật thân mềm chân bụng
|
25 |
+
sống ở biển trong họ Fasciolariidae.
|
26 |
+
model-index:
|
27 |
+
- name: lamhieu/ghost-7b-v0.9.0
|
28 |
+
results:
|
29 |
+
- task:
|
30 |
+
type: text-generation
|
31 |
+
dataset:
|
32 |
+
name: VMLU
|
33 |
+
type: vmlu_v1.5
|
34 |
+
metrics:
|
35 |
+
- type: avg
|
36 |
+
value: 36.06
|
37 |
+
name: Average
|
38 |
+
verified: true
|
39 |
+
- type: stem
|
40 |
+
value: 33.54
|
41 |
+
name: STEM
|
42 |
+
verified: true
|
43 |
+
- type: ss
|
44 |
+
value: 38.74
|
45 |
+
name: Social science
|
46 |
+
verified: true
|
47 |
+
- type: hm
|
48 |
+
value: 37.15
|
49 |
+
name: Humanities
|
50 |
+
verified: true
|
51 |
+
- type: ot
|
52 |
+
value: 36.78
|
53 |
+
name: Other
|
54 |
+
verified: true
|
55 |
+
- task:
|
56 |
+
type: text-generation
|
57 |
+
dataset:
|
58 |
+
name: Open LLM Leaderboard
|
59 |
+
type: open_llm_leaderboard
|
60 |
+
metrics:
|
61 |
+
- type: avg
|
62 |
+
value: 56.89
|
63 |
+
name: Average
|
64 |
+
verified: true
|
65 |
+
- type: arc
|
66 |
+
value: 53.07
|
67 |
+
name: ARC
|
68 |
+
verified: true
|
69 |
+
- type: hs
|
70 |
+
value: 77.93
|
71 |
+
name: HellaSwag
|
72 |
+
verified: true
|
73 |
+
- type: hs
|
74 |
+
value: 77.93
|
75 |
+
name: HellaSwag
|
76 |
+
verified: true
|
77 |
+
- type: mmlu
|
78 |
+
value: 55.09
|
79 |
+
name: MMLU
|
80 |
+
verified: true
|
81 |
+
- type: wg
|
82 |
+
value: 73.72
|
83 |
+
name: Winogrande
|
84 |
+
verified: true
|
85 |
+
- type: gsm8k
|
86 |
+
value: 33.74
|
87 |
+
name: GSM8K
|
88 |
+
verified: true
|
89 |
+
source:
|
90 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
91 |
+
name: Open LLM Leaderboard
|
92 |
+
- task:
|
93 |
+
type: text-generation
|
94 |
+
name: Text Generation
|
95 |
+
dataset:
|
96 |
+
name: AI2 Reasoning Challenge (25-Shot)
|
97 |
+
type: ai2_arc
|
98 |
+
config: ARC-Challenge
|
99 |
+
split: test
|
100 |
+
args:
|
101 |
+
num_few_shot: 25
|
102 |
+
metrics:
|
103 |
+
- type: acc_norm
|
104 |
+
value: 53.07
|
105 |
+
name: normalized accuracy
|
106 |
+
source:
|
107 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
108 |
+
name: Open LLM Leaderboard
|
109 |
+
- task:
|
110 |
+
type: text-generation
|
111 |
+
name: Text Generation
|
112 |
+
dataset:
|
113 |
+
name: HellaSwag (10-Shot)
|
114 |
+
type: hellaswag
|
115 |
+
split: validation
|
116 |
+
args:
|
117 |
+
num_few_shot: 10
|
118 |
+
metrics:
|
119 |
+
- type: acc_norm
|
120 |
+
value: 77.93
|
121 |
+
name: normalized accuracy
|
122 |
+
source:
|
123 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
124 |
+
name: Open LLM Leaderboard
|
125 |
+
- task:
|
126 |
+
type: text-generation
|
127 |
+
name: Text Generation
|
128 |
+
dataset:
|
129 |
+
name: MMLU (5-Shot)
|
130 |
+
type: cais/mmlu
|
131 |
+
config: all
|
132 |
+
split: test
|
133 |
+
args:
|
134 |
+
num_few_shot: 5
|
135 |
+
metrics:
|
136 |
+
- type: acc
|
137 |
+
value: 55.09
|
138 |
+
name: accuracy
|
139 |
+
source:
|
140 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
141 |
+
name: Open LLM Leaderboard
|
142 |
+
- task:
|
143 |
+
type: text-generation
|
144 |
+
name: Text Generation
|
145 |
+
dataset:
|
146 |
+
name: TruthfulQA (0-shot)
|
147 |
+
type: truthful_qa
|
148 |
+
config: multiple_choice
|
149 |
+
split: validation
|
150 |
+
args:
|
151 |
+
num_few_shot: 0
|
152 |
+
metrics:
|
153 |
+
- type: mc2
|
154 |
+
value: 47.79
|
155 |
+
source:
|
156 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
157 |
+
name: Open LLM Leaderboard
|
158 |
+
- task:
|
159 |
+
type: text-generation
|
160 |
+
name: Text Generation
|
161 |
+
dataset:
|
162 |
+
name: Winogrande (5-shot)
|
163 |
+
type: winogrande
|
164 |
+
config: winogrande_xl
|
165 |
+
split: validation
|
166 |
+
args:
|
167 |
+
num_few_shot: 5
|
168 |
+
metrics:
|
169 |
+
- type: acc
|
170 |
+
value: 73.72
|
171 |
+
name: accuracy
|
172 |
+
source:
|
173 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
174 |
+
name: Open LLM Leaderboard
|
175 |
+
- task:
|
176 |
+
type: text-generation
|
177 |
+
name: Text Generation
|
178 |
+
dataset:
|
179 |
+
name: GSM8k (5-shot)
|
180 |
+
type: gsm8k
|
181 |
+
config: main
|
182 |
+
split: test
|
183 |
+
args:
|
184 |
+
num_few_shot: 5
|
185 |
+
metrics:
|
186 |
+
- type: acc
|
187 |
+
value: 33.74
|
188 |
+
name: accuracy
|
189 |
+
source:
|
190 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=lamhieu/ghost-7b-v0.9.0
|
191 |
+
name: Open LLM Leaderboard
|
192 |
---
|
193 |
|
194 |
# Model Card for Model ID
|
|
|
402 |
|
403 |
## Model Card Contact
|
404 |
|
405 |
+
**Lam H** ([email protected])
|
406 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
407 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_lamhieu__ghost-7b-v0.9.0)
|
408 |
+
|
409 |
+
| Metric |Value|
|
410 |
+
|---------------------------------|----:|
|
411 |
+
|Avg. |56.89|
|
412 |
+
|AI2 Reasoning Challenge (25-Shot)|53.07|
|
413 |
+
|HellaSwag (10-Shot) |77.93|
|
414 |
+
|MMLU (5-Shot) |55.09|
|
415 |
+
|TruthfulQA (0-shot) |47.79|
|
416 |
+
|Winogrande (5-shot) |73.72|
|
417 |
+
|GSM8k (5-shot) |33.74|
|
418 |
+
|