Text Generation
Transformers
Safetensors
English
mistral
axolotl
Generated from Trainer
Mistral
instruct
finetune
chatml
gpt4
synthetic data
science
physics
chemistry
biology
math
conversational
Eval Results
Inference Endpoints
text-generation-inference
Weyaxi commited on
Commit
bca7a9f
1 Parent(s): 2a159df

correct open llm leaderboard in model card

Browse files
Files changed (1) hide show
  1. README.md +17 -20
README.md CHANGED
@@ -71,7 +71,7 @@ model-index:
71
  num_few_shot: 25
72
  metrics:
73
  - type: acc_norm
74
- value: 61.52
75
  name: normalized accuracy
76
  source:
77
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
@@ -87,7 +87,7 @@ model-index:
87
  num_few_shot: 10
88
  metrics:
89
  - type: acc_norm
90
- value: 80.91
91
  name: normalized accuracy
92
  source:
93
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
@@ -104,7 +104,7 @@ model-index:
104
  num_few_shot: 5
105
  metrics:
106
  - type: acc
107
- value: 62.02
108
  name: accuracy
109
  source:
110
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
@@ -121,7 +121,7 @@ model-index:
121
  num_few_shot: 0
122
  metrics:
123
  - type: mc2
124
- value: 51.24
125
  source:
126
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
127
  name: Open LLM Leaderboard
@@ -137,7 +137,7 @@ model-index:
137
  num_few_shot: 5
138
  metrics:
139
  - type: acc
140
- value: 78.53
141
  name: accuracy
142
  source:
143
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
@@ -154,7 +154,7 @@ model-index:
154
  num_few_shot: 5
155
  metrics:
156
  - type: acc
157
- value: 58.61
158
  name: accuracy
159
  source:
160
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
@@ -366,7 +366,17 @@ model.generate(**gen_input)
366
  - https://huggingface.co/bartowski/Einstein-v6-7B-exl2
367
 
368
  # 🎯 [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
 
369
 
 
 
 
 
 
 
 
 
 
370
 
371
  # 🤖 Additional information about training
372
 
@@ -394,17 +404,4 @@ Thanks to all open source AI community.
394
 
395
  If you would like to support me:
396
 
397
- [☕ Buy Me a Coffee](https://www.buymeacoffee.com/weyaxi)
398
- # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
399
- Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Weyaxi__Einstein-v6-7B)
400
-
401
- | Metric |Value|
402
- |---------------------------------|----:|
403
- |Avg. |65.47|
404
- |AI2 Reasoning Challenge (25-Shot)|61.52|
405
- |HellaSwag (10-Shot) |80.91|
406
- |MMLU (5-Shot) |62.02|
407
- |TruthfulQA (0-shot) |51.24|
408
- |Winogrande (5-shot) |78.53|
409
- |GSM8k (5-shot) |58.61|
410
-
 
71
  num_few_shot: 25
72
  metrics:
73
  - type: acc_norm
74
+ value: 63.57
75
  name: normalized accuracy
76
  source:
77
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
 
87
  num_few_shot: 10
88
  metrics:
89
  - type: acc_norm
90
+ value: 82.76
91
  name: normalized accuracy
92
  source:
93
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
 
104
  num_few_shot: 5
105
  metrics:
106
  - type: acc
107
+ value: 62.23
108
  name: accuracy
109
  source:
110
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
 
121
  num_few_shot: 0
122
  metrics:
123
  - type: mc2
124
+ value: 52.02
125
  source:
126
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
127
  name: Open LLM Leaderboard
 
137
  num_few_shot: 5
138
  metrics:
139
  - type: acc
140
+ value: 78.61
141
  name: accuracy
142
  source:
143
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
 
154
  num_few_shot: 5
155
  metrics:
156
  - type: acc
157
+ value: 63.53
158
  name: accuracy
159
  source:
160
  url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Weyaxi/Einstein-v6-7B
 
366
  - https://huggingface.co/bartowski/Einstein-v6-7B-exl2
367
 
368
  # 🎯 [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
369
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Weyaxi__Einstein-v6-7B)
370
 
371
+ | Metric |Value|
372
+ |---------------------------------|----:|
373
+ |Avg. |67.12|
374
+ |AI2 Reasoning Challenge (25-Shot)|63.57|
375
+ |HellaSwag (10-Shot) |82.76|
376
+ |MMLU (5-Shot) |62.23|
377
+ |TruthfulQA (0-shot) |52.02|
378
+ |Winogrande (5-shot) |78.61|
379
+ |GSM8k (5-shot) |63.53|
380
 
381
  # 🤖 Additional information about training
382
 
 
404
 
405
  If you would like to support me:
406
 
407
+ [☕ Buy Me a Coffee](https://www.buymeacoffee.com/weyaxi)