Adding Evaluation Results
#3
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -1,11 +1,11 @@
|
|
1 |
---
|
2 |
-
license: apache-2.0
|
3 |
language:
|
4 |
- en
|
5 |
-
|
6 |
datasets:
|
7 |
- Skylion007/openwebtext
|
8 |
- Locutusque/TM-DATA
|
|
|
9 |
inference:
|
10 |
parameters:
|
11 |
do_sample: true
|
@@ -15,22 +15,130 @@ inference:
|
|
15 |
max_new_tokens: 250
|
16 |
repetition_penalty: 1.16
|
17 |
widget:
|
18 |
-
- text:
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
---
|
35 |
# Training
|
36 |
This model was trained on two datasets, shown in this model page.
|
@@ -41,4 +149,17 @@ Training took approximately 500 GPU hours on a single Titan V.
|
|
41 |
You can look at the training metrics here:
|
42 |
https://wandb.ai/locutusque/TinyMistral-V2/runs/g0rvw6wc
|
43 |
|
44 |
-
🔥 This model performed excellently on TruthfulQA, outperforming models more than 720x its size. These models include: mistralai/Mixtral-8x7B-v0.1, tiiuae/falcon-180B, berkeley-nest/Starling-LM-7B-alpha, upstage/SOLAR-10.7B-v1.0, and more. 🔥
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
|
|
2 |
language:
|
3 |
- en
|
4 |
+
license: apache-2.0
|
5 |
datasets:
|
6 |
- Skylion007/openwebtext
|
7 |
- Locutusque/TM-DATA
|
8 |
+
pipeline_tag: text-generation
|
9 |
inference:
|
10 |
parameters:
|
11 |
do_sample: true
|
|
|
15 |
max_new_tokens: 250
|
16 |
repetition_penalty: 1.16
|
17 |
widget:
|
18 |
+
- text: 'TITLE: Dirichlet density QUESTION [5 upvotes]: How to solve the following
|
19 |
+
exercise: Let $q$ be prime. Show that the set of primes p for which $p \equiv
|
20 |
+
1\pmod q$ and $2^{(p-1)/q} \equiv 1 \pmod p$ has Dirichlet density $\dfrac{1}{q(q-1)}$.
|
21 |
+
I want to show that $X^q-2$ (mod $p$) has a solution and $q$ divides $p-1$ , these
|
22 |
+
two conditions are simultaneonusly satisfied iff p splits completely in $\Bbb{Q}(\zeta_q,2^{\frac{1}{q}})$.
|
23 |
+
$\zeta_q $ is primitive $q^{th}$ root of unity. If this is proved the I can conclude
|
24 |
+
the result by Chebotarev density theorem. REPLY [2 votes]:'
|
25 |
+
- text: An emerging clinical approach to treat substance abuse disorders involves
|
26 |
+
a form of cognitive-behavioral therapy whereby addicts learn to reduce their reactivity
|
27 |
+
to drug-paired stimuli through cue-exposure or extinction training. It is, however,
|
28 |
+
- text: '\begin{document} \begin{frontmatter} \author{Mahouton Norbert Hounkonnou\corref{cor1}${}^1$}
|
29 |
+
\cortext[cor1]{norbert.hounkonnou@cipma.uac.bj} \author{Sama Arjika\corref{cor2}${}^1$}
|
30 |
+
\cortext[cor2]{rjksama2008@gmail.com} \author{ Won Sang Chung\corref{cor3}${}^2$
|
31 |
+
} \cortext[cor3]{mimip4444@hanmail.net} \title{\bf New families of $q$ and $(q;p)-$Hermite
|
32 |
+
polynomials } \address{${}^1$International Chair of Mathematical Physics and Applications
|
33 |
+
\\ (ICMPA-UNESCO Chair), University of Abomey-Calavi,\\ 072 B. P.: 50 Cotonou,
|
34 |
+
Republic of Benin,\\ ${}^2$Department of Physics and Research Institute of Natural
|
35 |
+
Science, \\ College of Natural Science, \\ Gyeongsang National University, Jinju
|
36 |
+
660-701, Korea } \begin{abstract} In this paper, we construct a new family of
|
37 |
+
$q-$Hermite polynomials denoted by $H_n(x,s|q).$ Main properties and relations
|
38 |
+
are established and'
|
39 |
+
model-index:
|
40 |
+
- name: TinyMistral-248M-v2
|
41 |
+
results:
|
42 |
+
- task:
|
43 |
+
type: text-generation
|
44 |
+
name: Text Generation
|
45 |
+
dataset:
|
46 |
+
name: AI2 Reasoning Challenge (25-Shot)
|
47 |
+
type: ai2_arc
|
48 |
+
config: ARC-Challenge
|
49 |
+
split: test
|
50 |
+
args:
|
51 |
+
num_few_shot: 25
|
52 |
+
metrics:
|
53 |
+
- type: acc_norm
|
54 |
+
value: 21.25
|
55 |
+
name: normalized accuracy
|
56 |
+
source:
|
57 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/TinyMistral-248M-v2
|
58 |
+
name: Open LLM Leaderboard
|
59 |
+
- task:
|
60 |
+
type: text-generation
|
61 |
+
name: Text Generation
|
62 |
+
dataset:
|
63 |
+
name: HellaSwag (10-Shot)
|
64 |
+
type: hellaswag
|
65 |
+
split: validation
|
66 |
+
args:
|
67 |
+
num_few_shot: 10
|
68 |
+
metrics:
|
69 |
+
- type: acc_norm
|
70 |
+
value: 26.56
|
71 |
+
name: normalized accuracy
|
72 |
+
source:
|
73 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/TinyMistral-248M-v2
|
74 |
+
name: Open LLM Leaderboard
|
75 |
+
- task:
|
76 |
+
type: text-generation
|
77 |
+
name: Text Generation
|
78 |
+
dataset:
|
79 |
+
name: MMLU (5-Shot)
|
80 |
+
type: cais/mmlu
|
81 |
+
config: all
|
82 |
+
split: test
|
83 |
+
args:
|
84 |
+
num_few_shot: 5
|
85 |
+
metrics:
|
86 |
+
- type: acc
|
87 |
+
value: 23.39
|
88 |
+
name: accuracy
|
89 |
+
source:
|
90 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/TinyMistral-248M-v2
|
91 |
+
name: Open LLM Leaderboard
|
92 |
+
- task:
|
93 |
+
type: text-generation
|
94 |
+
name: Text Generation
|
95 |
+
dataset:
|
96 |
+
name: TruthfulQA (0-shot)
|
97 |
+
type: truthful_qa
|
98 |
+
config: multiple_choice
|
99 |
+
split: validation
|
100 |
+
args:
|
101 |
+
num_few_shot: 0
|
102 |
+
metrics:
|
103 |
+
- type: mc2
|
104 |
+
value: 49.6
|
105 |
+
source:
|
106 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/TinyMistral-248M-v2
|
107 |
+
name: Open LLM Leaderboard
|
108 |
+
- task:
|
109 |
+
type: text-generation
|
110 |
+
name: Text Generation
|
111 |
+
dataset:
|
112 |
+
name: Winogrande (5-shot)
|
113 |
+
type: winogrande
|
114 |
+
config: winogrande_xl
|
115 |
+
split: validation
|
116 |
+
args:
|
117 |
+
num_few_shot: 5
|
118 |
+
metrics:
|
119 |
+
- type: acc
|
120 |
+
value: 51.85
|
121 |
+
name: accuracy
|
122 |
+
source:
|
123 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/TinyMistral-248M-v2
|
124 |
+
name: Open LLM Leaderboard
|
125 |
+
- task:
|
126 |
+
type: text-generation
|
127 |
+
name: Text Generation
|
128 |
+
dataset:
|
129 |
+
name: GSM8k (5-shot)
|
130 |
+
type: gsm8k
|
131 |
+
config: main
|
132 |
+
split: test
|
133 |
+
args:
|
134 |
+
num_few_shot: 5
|
135 |
+
metrics:
|
136 |
+
- type: acc
|
137 |
+
value: 0.0
|
138 |
+
name: accuracy
|
139 |
+
source:
|
140 |
+
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Locutusque/TinyMistral-248M-v2
|
141 |
+
name: Open LLM Leaderboard
|
142 |
---
|
143 |
# Training
|
144 |
This model was trained on two datasets, shown in this model page.
|
|
|
149 |
You can look at the training metrics here:
|
150 |
https://wandb.ai/locutusque/TinyMistral-V2/runs/g0rvw6wc
|
151 |
|
152 |
+
🔥 This model performed excellently on TruthfulQA, outperforming models more than 720x its size. These models include: mistralai/Mixtral-8x7B-v0.1, tiiuae/falcon-180B, berkeley-nest/Starling-LM-7B-alpha, upstage/SOLAR-10.7B-v1.0, and more. 🔥
|
153 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
154 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Locutusque__TinyMistral-248M-v2)
|
155 |
+
|
156 |
+
| Metric |Value|
|
157 |
+
|---------------------------------|----:|
|
158 |
+
|Avg. |28.78|
|
159 |
+
|AI2 Reasoning Challenge (25-Shot)|21.25|
|
160 |
+
|HellaSwag (10-Shot) |26.56|
|
161 |
+
|MMLU (5-Shot) |23.39|
|
162 |
+
|TruthfulQA (0-shot) |49.60|
|
163 |
+
|Winogrande (5-shot) |51.85|
|
164 |
+
|GSM8k (5-shot) | 0.00|
|
165 |
+
|