Hugging Face's logo

OpenLLM-Ro
/
RoLlama2-7b-Instruct-2024-05-14

Text Generation
Transformers
Safetensors
Romanian
llama
Eval Results
text-generation-inference
Model card Files Community
mihaimasala commited on
Commit
3e6abc3
·
verified ·
1 Parent(s): 3616d0d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -7
README.md CHANGED
@@ -75,18 +75,20 @@ print(tokenizer.decode(outputs[0]))
75
  | Model | Average | ARC | MMLU |Winogrande|HellaSwag | GSM8k |TruthfulQA|
76
  |--------------------|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|
77
  | Llama-2-7b-chat | 38.03 | 37.95 | 27.22 | 59.29 | 57.22 | 2.53 | 44.00 |
78
- |RoLlama2-7b-Instruct|**45.71**|**43.66**|**39.70**|**70.34** | 57.36 |**18.78**| 44.44 |
79
- |*RoLlama2-7b-Chat* | *43.82* | *41.92* | *37.29* | *66.68* | ***57.91***| *13.47* | ***45.65***|
80
 
81
-
 
82
 
83
  ## MT-Bench
84
 
 
85
  | Model | Average | 1st turn | 2nd turn |
86
  |--------------------|:--------:|:--------:|:--------:|
87
  | Llama-2-7b-chat | 1.21 | 1.68 | 0.74 |
88
- |RoLlama2-7b-Instruct| **3.70**|**4.74**| 2.66 |
89
- |*RoLlama2-7b-Chat* | *TBC* | *TBC* | *TBC* |
90
 
91
 
92
 
@@ -95,8 +97,8 @@ print(tokenizer.decode(outputs[0]))
95
  | Model | Link |
96
  |--------------------|:--------:|
97
  |RoLlama2-7b-Base | [link](https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Base) |
98
- |RoLlama2-7b-Instruct| [link](https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Instruct) |
99
- |*RoLlama2-7b-Chat* | [link](https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Chat) |
100
 
101
 
102
  <!--
 
75
  | Model | Average | ARC | MMLU |Winogrande|HellaSwag | GSM8k |TruthfulQA|
76
  |--------------------|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|
77
  | Llama-2-7b-chat | 38.03 | 37.95 | 27.22 | 59.29 | 57.22 | 2.53 | 44.00 |
78
+ |*RoLlama2-7b-Instruct*|***45.71***|***43.66***|***39.70***|***70.34*** | *57.36* |***18.78***| *44.44* |
79
+ |RoLlama2-7b-Chat | 43.82 | 41.92 | 37.29 | 66.68 | **57.91**| 13.47 | **45.65**|
80
 
81
+
82
+
83
 
84
  ## MT-Bench
85
 
86
+
87
  | Model | Average | 1st turn | 2nd turn |
88
  |--------------------|:--------:|:--------:|:--------:|
89
  | Llama-2-7b-chat | 1.21 | 1.68 | 0.74 |
90
+ |*RoLlama2-7b-Instruct*| ***3.70***|***4.74***| *2.66* |
91
+ |RoLlama2-7b-Chat | TBC | TBC | TBC |
92
 
93
 
94
 
 
97
  | Model | Link |
98
  |--------------------|:--------:|
99
  |RoLlama2-7b-Base | [link](https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Base) |
100
+ |*RoLlama2-7b-Instruct*| [link](https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Instruct) |
101
+ |RoLlama2-7b-Chat | [link](https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Chat) |
102
 
103
 
104
  <!--