Hugging Face's logo

arcee-ai
/
Llama-3.1-SuperNova-Lite

Text Generation
Transformers
Safetensors
English
llama
conversational
Eval Results
text-generation-inference
Model card Files
xet
Community
16

add AIBOM

#16
by RiccardoDav - opened
base: refs/heads/main
←
from: refs/pr/16
Discussion Files changed
+140
-0
Files changed (1) hide show
  1. arcee-ai_Llama-3.1-SuperNova-Lite.json +140 -0
arcee-ai_Llama-3.1-SuperNova-Lite.json ADDED
@@ -0,0 +1,140 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bomFormat": "CycloneDX",
3
+ "specVersion": "1.6",
4
+ "serialNumber": "urn:uuid:d758fab9-5640-4246-840f-4eb257180280",
5
+ "version": 1,
6
+ "metadata": {
7
+ "timestamp": "2025-06-05T09:36:31.645371+00:00",
8
+ "component": {
9
+ "type": "machine-learning-model",
10
+ "bom-ref": "arcee-ai/Llama-3.1-SuperNova-Lite-1c3869d0-46e7-5d1b-ba9c-793f79901c6e",
11
+ "name": "arcee-ai/Llama-3.1-SuperNova-Lite",
12
+ "externalReferences": [
13
+ {
14
+ "url": "https://huggingface.co/arcee-ai/Llama-3.1-SuperNova-Lite",
15
+ "type": "documentation"
16
+ }
17
+ ],
18
+ "modelCard": {
19
+ "modelParameters": {
20
+ "task": "text-generation",
21
+ "architectureFamily": "llama",
22
+ "modelArchitecture": "LlamaForCausalLM",
23
+ "datasets": [
24
+ {
25
+ "ref": "arcee-ai/EvolKit-20k-62482fcc-6326-5541-84fe-23ff004cbf51"
26
+ }
27
+ ]
28
+ },
29
+ "properties": [
30
+ {
31
+ "name": "library_name",
32
+ "value": "transformers"
33
+ },
34
+ {
35
+ "name": "base_model",
36
+ "value": "meta-llama/Meta-Llama-3.1-8B-Instruct"
37
+ }
38
+ ],
39
+ "quantitativeAnalysis": {
40
+ "performanceMetrics": [
41
+ {
42
+ "slice": "dataset: HuggingFaceH4/ifeval",
43
+ "type": "inst_level_strict_acc and prompt_level_strict_acc",
44
+ "value": 80.17
45
+ },
46
+ {
47
+ "slice": "dataset: BBH",
48
+ "type": "acc_norm",
49
+ "value": 31.57
50
+ },
51
+ {
52
+ "slice": "dataset: hendrycks/competition_math",
53
+ "type": "exact_match",
54
+ "value": 15.48
55
+ },
56
+ {
57
+ "slice": "dataset: Idavidrein/gpqa",
58
+ "type": "acc_norm",
59
+ "value": 7.49
60
+ },
61
+ {
62
+ "slice": "dataset: TAUR-Lab/MuSR",
63
+ "type": "acc_norm",
64
+ "value": 11.67
65
+ },
66
+ {
67
+ "slice": "dataset: TIGER-Lab/MMLU-Pro, split: test, config: main",
68
+ "type": "acc",
69
+ "value": 31.97
70
+ }
71
+ ]
72
+ }
73
+ },
74
+ "authors": [
75
+ {
76
+ "name": "arcee-ai"
77
+ }
78
+ ],
79
+ "licenses": [
80
+ {
81
+ "license": {
82
+ "name": "llama3"
83
+ }
84
+ }
85
+ ],
86
+ "description": "Llama-3.1-SuperNova-Lite is an 8B parameter model developed by Arcee.ai, based on the Llama-3.1-8B-Instruct architecture. It is a distilled version of the larger Llama-3.1-405B-Instruct model, leveraging offline logits extracted from the 405B parameter variant. This 8B variation of Llama-3.1-SuperNova maintains high performance while offering exceptional instruction-following capabilities and domain-specific adaptability.The model was trained using a state-of-the-art distillation pipeline and an instruction dataset generated with [EvolKit](https://github.com/arcee-ai/EvolKit), ensuring accuracy and efficiency across a wide range of tasks. For more information on its training, visit blog.arcee.ai.Llama-3.1-SuperNova-Lite excels in both benchmark performance and real-world applications, providing the power of large-scale models in a more compact, efficient form ideal for organizations seeking high performance with reduced resource requirements.",
87
+ "tags": [
88
+ "transformers",
89
+ "safetensors",
90
+ "llama",
91
+ "text-generation",
92
+ "conversational",
93
+ "en",
94
+ "dataset:arcee-ai/EvolKit-20k",
95
+ "base_model:meta-llama/Llama-3.1-8B-Instruct",
96
+ "base_model:finetune:meta-llama/Llama-3.1-8B-Instruct",
97
+ "license:llama3",
98
+ "model-index",
99
+ "autotrain_compatible",
100
+ "text-generation-inference",
101
+ "endpoints_compatible",
102
+ "region:us"
103
+ ]
104
+ }
105
+ },
106
+ "components": [
107
+ {
108
+ "type": "data",
109
+ "bom-ref": "arcee-ai/EvolKit-20k-62482fcc-6326-5541-84fe-23ff004cbf51",
110
+ "name": "arcee-ai/EvolKit-20k",
111
+ "data": [
112
+ {
113
+ "type": "dataset",
114
+ "bom-ref": "arcee-ai/EvolKit-20k-62482fcc-6326-5541-84fe-23ff004cbf51",
115
+ "name": "arcee-ai/EvolKit-20k",
116
+ "contents": {
117
+ "url": "https://huggingface.co/datasets/arcee-ai/EvolKit-20k",
118
+ "properties": [
119
+ {
120
+ "name": "license",
121
+ "value": "mit"
122
+ }
123
+ ]
124
+ },
125
+ "governance": {
126
+ "owners": [
127
+ {
128
+ "organization": {
129
+ "name": "arcee-ai",
130
+ "url": "https://huggingface.co/arcee-ai"
131
+ }
132
+ }
133
+ ]
134
+ },
135
+ "description": "\n\t\n\t\t\n\t\tEvolKit-20k\n\t\n\nThis is a subset of a larger dataset generated for the purpose of training our Llama-3.1-SuperNova model. It utilized our EvolKit repository: https://github.com/arcee-ai/EvolKit.\n"
136
+ }
137
+ ]
138
+ }
139
+ ]
140
+ }