jaideepr97 commited on
Commit
30046b1
1 Parent(s): 136380c

add MMLU score

Browse files
Files changed (1) hide show
  1. README.md +11 -11
README.md CHANGED
@@ -19,17 +19,17 @@ base_model: ibm/granite-7b-base
19
 
20
  ### Performance
21
 
22
- | Model | Alignment | Base | Teacher | MTBench (Avg) * |
23
- | --- | --- | --- | --- | --- |
24
- | [Llama-2-13b-chat-hf](https://huggingface.co/meta-llama/Llama-2-13b-chat-hf) | RLHF | Llama-2-13b | Human Annotators | 6.65 |
25
- | [Orca-2-13b](https://huggingface.co/microsoft/Orca-2-13b) | Progressive Training | Llama-2-13b | GPT-4 | 6.15 |
26
- | [WizardLM-13B-V1.2](https://huggingface.co/WizardLM/WizardLM-13B-V1.2) | Evol-Instruct | Llama-2-13b | GPT-4 | 7.20 |
27
- | [Labradorite-13b](https://huggingface.co/ibm/labradorite-13b) | Large-scale Alignment for chatBots (LAB) | Llama-2-13b | Mixtral-8x7B-Instruct | 7.23 |
28
- | [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) | SFT | Mistral-7B-v0.1 | - | 6.84 |
29
- | [zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) | SFT/DPO | Mistral-7B-v0.1 | GPT-4 | 7.34 |
30
- | [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) | SFT | Mistral-7B-v0.1 | - | 7.6** |
31
- | [Merlinite-7b-lab](https://huggingface.co/instructlab/merlinite-7b-lab) | Large-scale Alignment for chatBots (LAB) | Mistral-7B-v0.1 | Mixtral-8x7B-Instruct | 7.66 |
32
- | Granite-7b-lab | Large-scale Alignment for chatBots (LAB) | Granite-7b-base| Mixtral-8x7B-Instruct | 6.69 |
33
 
34
  [*] Numbers for models other than Merlinite-7b-lab, Granite-7b-lab and [Labradorite-13b](https://huggingface.co/ibm/labradorite-13b) are taken from [lmsys/chatbot-arena-leaderboard](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard)
35
 
 
19
 
20
  ### Performance
21
 
22
+ | Model | Alignment | Base | Teacher | MTBench (Avg) * | MMLU(5-shot) |
23
+ | --- | --- | --- | --- | --- | --- |
24
+ | [Llama-2-13b-chat-hf](https://huggingface.co/meta-llama/Llama-2-13b-chat-hf) | RLHF | Llama-2-13b | Human Annotators | 6.65 |54.58 |
25
+ | [Orca-2-13b](https://huggingface.co/microsoft/Orca-2-13b) | Progressive Training | Llama-2-13b | GPT-4 | 6.15 | 60.37 * |
26
+ | [WizardLM-13B-V1.2](https://huggingface.co/WizardLM/WizardLM-13B-V1.2) | Evol-Instruct | Llama-2-13b | GPT-4 | 7.20 | 54.83 |
27
+ | [Labradorite-13b](https://huggingface.co/ibm/labradorite-13b) | Large-scale Alignment for chatBots (LAB) | Llama-2-13b | Mixtral-8x7B-Instruct | 7.23 | 58.89 |
28
+ | [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) | SFT | Mistral-7B-v0.1 | - | 6.84 | 60.37 |
29
+ | [zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) | SFT/DPO | Mistral-7B-v0.1 | GPT-4 | 7.34 | 61.07 |
30
+ | [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) | SFT | Mistral-7B-v0.1 | - | 7.6** | 60.78 |
31
+ | [Merlinite-7b-lab](https://huggingface.co/instructlab/merlinite-7b-lab) | Large-scale Alignment for chatBots (LAB) | Mistral-7B-v0.1 | Mixtral-8x7B-Instruct | 7.66 |64.88 |
32
+ | Granite-7b-lab | Large-scale Alignment for chatBots (LAB) | Granite-7b-base| Mixtral-8x7B-Instruct | 6.69 | 51.91 |
33
 
34
  [*] Numbers for models other than Merlinite-7b-lab, Granite-7b-lab and [Labradorite-13b](https://huggingface.co/ibm/labradorite-13b) are taken from [lmsys/chatbot-arena-leaderboard](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard)
35