Hugging Face's logo

llama-moe
/
LLaMA-MoE-v1-3_5B-4_16

Text Generation
Transformers
PyTorch
English
llama_moe
MoE
custom_code
Model card Files Community
3
Spico commited on
Commit
52e67c7
•
1 Parent(s): 98361f7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -53,7 +53,7 @@ print(tokenizer.decode(pred.cpu()[0], skip_special_tokens=True))
53
  | **LLaMA-MoE-3.5B (4/16)** | 4 | 16 | 3.5B | [[🤗 HF Weights]](https://huggingface.co/llama-moe/LLaMA-MoE-v1-3_5B-4_16) |
54
  | **LLaMA-MoE-3.5B (2/8)** | 2 | 8 | 3.5B | [[🤗 HF Weights]](https://huggingface.co/llama-moe/LLaMA-MoE-v1-3_5B-2_8) |
55
 
56
- | Model | SciQ | PIQA | WinoGrande | ARC-e | ARC-c (25) | HellaSwag (10) | LogiQA | BoolQ (32) | LAMBADA | NQ (32) | MMNLU (5) | Average |
57
  | :------------------------------------------------------------------------------------ | :------: | :------: | :--------: | :------: | :--------: | :------------: | :------: | :--------: | :------: | :------: | :-------: | :-----: |
58
  | [OPT-2.7B](https://huggingface.co/facebook/opt-2.7b) | 78.9 | 74.8 | 60.8 | 54.4 | 34.0 | 61.4 | 25.8 | 63.3 | 63.6 | 10.7 | 25.8 | 50.3 |
59
  | [Pythia-2.8B](https://huggingface.co/EleutherAI/pythia-2.8b) | 83.2 | 73.6 | 59.6 | 58.8 | 36.7 | 60.7 | 28.1 | 65.9 | 64.6 | 8.7 | 26.8 | 51.5 |
 
53
  | **LLaMA-MoE-3.5B (4/16)** | 4 | 16 | 3.5B | [[🤗 HF Weights]](https://huggingface.co/llama-moe/LLaMA-MoE-v1-3_5B-4_16) |
54
  | **LLaMA-MoE-3.5B (2/8)** | 2 | 8 | 3.5B | [[🤗 HF Weights]](https://huggingface.co/llama-moe/LLaMA-MoE-v1-3_5B-2_8) |
55
 
56
+ | Model | SciQ | PIQA | WinoGrande | ARC-e | ARC-c (25) | HellaSwag (10) | LogiQA | BoolQ (32) | LAMBADA | NQ (32) | MMLU (5) | Average |
57
  | :------------------------------------------------------------------------------------ | :------: | :------: | :--------: | :------: | :--------: | :------------: | :------: | :--------: | :------: | :------: | :-------: | :-----: |
58
  | [OPT-2.7B](https://huggingface.co/facebook/opt-2.7b) | 78.9 | 74.8 | 60.8 | 54.4 | 34.0 | 61.4 | 25.8 | 63.3 | 63.6 | 10.7 | 25.8 | 50.3 |
59
  | [Pythia-2.8B](https://huggingface.co/EleutherAI/pythia-2.8b) | 83.2 | 73.6 | 59.6 | 58.8 | 36.7 | 60.7 | 28.1 | 65.9 | 64.6 | 8.7 | 26.8 | 51.5 |