Commit
26a22e6
1 Parent(s): 9e5e050

Adding Evaluation Results (#1)

Browse files

- Adding Evaluation Results (cfcd3aa10cabe146c5afea01d15d0c17a80a0257)


Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>

Files changed (1) hide show
  1. README.md +118 -2
README.md CHANGED
@@ -1,6 +1,109 @@
1
  ---
2
- library_name: transformers
3
  license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  ---
5
 
6
  # Model Card for devingulliver/llama-pile-350b
@@ -194,4 +297,17 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
194
 
195
  ## Model Card Contact
196
 
197
- [More Information Needed]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
 
2
  license: apache-2.0
3
+ library_name: transformers
4
+ model-index:
5
+ - name: llama-pile-350b
6
+ results:
7
+ - task:
8
+ type: text-generation
9
+ name: Text Generation
10
+ dataset:
11
+ name: AI2 Reasoning Challenge (25-Shot)
12
+ type: ai2_arc
13
+ config: ARC-Challenge
14
+ split: test
15
+ args:
16
+ num_few_shot: 25
17
+ metrics:
18
+ - type: acc_norm
19
+ value: 33.19
20
+ name: normalized accuracy
21
+ source:
22
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=devingulliver/llama-pile-350b
23
+ name: Open LLM Leaderboard
24
+ - task:
25
+ type: text-generation
26
+ name: Text Generation
27
+ dataset:
28
+ name: HellaSwag (10-Shot)
29
+ type: hellaswag
30
+ split: validation
31
+ args:
32
+ num_few_shot: 10
33
+ metrics:
34
+ - type: acc_norm
35
+ value: 56.6
36
+ name: normalized accuracy
37
+ source:
38
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=devingulliver/llama-pile-350b
39
+ name: Open LLM Leaderboard
40
+ - task:
41
+ type: text-generation
42
+ name: Text Generation
43
+ dataset:
44
+ name: MMLU (5-Shot)
45
+ type: cais/mmlu
46
+ config: all
47
+ split: test
48
+ args:
49
+ num_few_shot: 5
50
+ metrics:
51
+ - type: acc
52
+ value: 24.66
53
+ name: accuracy
54
+ source:
55
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=devingulliver/llama-pile-350b
56
+ name: Open LLM Leaderboard
57
+ - task:
58
+ type: text-generation
59
+ name: Text Generation
60
+ dataset:
61
+ name: TruthfulQA (0-shot)
62
+ type: truthful_qa
63
+ config: multiple_choice
64
+ split: validation
65
+ args:
66
+ num_few_shot: 0
67
+ metrics:
68
+ - type: mc2
69
+ value: 36.28
70
+ source:
71
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=devingulliver/llama-pile-350b
72
+ name: Open LLM Leaderboard
73
+ - task:
74
+ type: text-generation
75
+ name: Text Generation
76
+ dataset:
77
+ name: Winogrande (5-shot)
78
+ type: winogrande
79
+ config: winogrande_xl
80
+ split: validation
81
+ args:
82
+ num_few_shot: 5
83
+ metrics:
84
+ - type: acc
85
+ value: 58.48
86
+ name: accuracy
87
+ source:
88
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=devingulliver/llama-pile-350b
89
+ name: Open LLM Leaderboard
90
+ - task:
91
+ type: text-generation
92
+ name: Text Generation
93
+ dataset:
94
+ name: GSM8k (5-shot)
95
+ type: gsm8k
96
+ config: main
97
+ split: test
98
+ args:
99
+ num_few_shot: 5
100
+ metrics:
101
+ - type: acc
102
+ value: 0.76
103
+ name: accuracy
104
+ source:
105
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=devingulliver/llama-pile-350b
106
+ name: Open LLM Leaderboard
107
  ---
108
 
109
  # Model Card for devingulliver/llama-pile-350b
 
297
 
298
  ## Model Card Contact
299
 
300
+ [More Information Needed]
301
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
302
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_devingulliver__llama-pile-350b)
303
+
304
+ | Metric |Value|
305
+ |---------------------------------|----:|
306
+ |Avg. |35.00|
307
+ |AI2 Reasoning Challenge (25-Shot)|33.19|
308
+ |HellaSwag (10-Shot) |56.60|
309
+ |MMLU (5-Shot) |24.66|
310
+ |TruthfulQA (0-shot) |36.28|
311
+ |Winogrande (5-shot) |58.48|
312
+ |GSM8k (5-shot) | 0.76|
313
+