Adapter Base model:adapter:microsoft/p... Base model:microsoft/phi-2 Dataset:gsm8k En Finetuned Generated from trainer Lora Model-index Peft Pytorch Region:us Safetensors Sft Tensorboard Trl
Phi 2 Basic Maths Benchmarks
ARC: 55.8-42.1%
HellaSwag: 71.15-25.3%
MMLU: 47.27-45.4%
TruthfulQA: 41.4-29.8%
WinoGrande: 75.3-13.9%
GSM8K: 30.71-66.6%
nn.n% — How the model compares to the GPT-4.
Phi 2 Basic Maths Parameters and Internals
| LLM Name | Phi 2 Basic Maths |
| Repository | Open on 🤗 |
| Base Model(s) | |
| Required VRAM | 0.8 GB |
| Updated | 2024-07-27 |
| Maintainer | Menouar |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Adapter |
| License | mit |
| Model Max Length | 2048 |
| Is Biased | none |
| Tokenizer Class | CodeGenTokenizer |
| Padding Token | <|endoftext|> |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | k_proj|dense|v_proj|fc2|fc1|q_proj |
| LoRA Alpha | 128 |
| LoRA Dropout | 0.05 |
| R Param | 256 |
Best Alternatives to Phi 2 Basic Maths
Best Alternatives |
HF Rank |
Context/RAM |
Downloads |
Likes |
|---|---|---|---|---|
| Tot Llama Update | 0.3 | 0K / 0 GB | 1536 | 0 |
| CodeHAWK | 0.3 | 0K / 0 GB | 9 | 0 |
| Zephyr Phi 1 5 Sft Qlora | 0.3 | 0K / 0 GB | 5 | 0 |
| Phi Openllm Lb Test | 0.3 | 0K / 0 GB | 5 | 0 |
| Hua V0.1 | 0.3 | 0K / 0 GB | 10 | 0 |
| Cap Iaa Lora Attn | 0.3 | 0K / 0.2 GB | 32 | 0 |
| ...uu Lora Pretrain V3.1719338529 | 0.2 | 0K / 0.6 GB | 28 | 0 |
| Absa 10 Domains Large | 0.2 | 0K / 0.1 GB | 19 | 0 |
| ...uu Lora Pretrain V3.1719339792 | 0.2 | 0K / 0.3 GB | 15 | 0 |
| Dreamtobenlpsama Mnlp M3 | 0.2 | 0K / 0 GB | 48 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than Menouar/phi-2-basic-maths .
Rank the Phi 2 Basic Maths Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 34446 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072501