Merged Model Autotrain compatible Base model:finetune:microsoft/... Base model:microsoft/phi-2 Conversational Dataset:argilla/distilabel-int... Dataset:argilla/distilabel-mat... Dataset:huggingfaceh4/ultrafee... Dataset:teknium/openhermes-2.5 Endpoints compatible Instruct Phi Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/rasyosef/phi-2-instruct-v0.1
Phi 2 Instruct V0.1 Benchmarks
MMLU Pro: 13.85
GPQA: 3.24
MUSR: 5.04
BBH: 26.36
IFEval: 36.81 vs 88 (so35)-58.2%
LLME Score: 0.27653
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Phi 2 Instruct V0.1 Parameters and Internals
| LLM Name | Phi 2 Instruct V0.1 |
| Repository 🤗 | https://huggingface.co/rasyosef/phi-2-instruct-v0.1 |
| Base Model(s) | |
| Merged Model | Yes |
| Model Size | 2.8b |
| Required VRAM | 5.6 GB |
| Updated | 2024-12-21 |
| Maintainer | rasyosef |
| Model Type | phi |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | PhiForCausalLM |
| License | mit |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.42.4 |
| Tokenizer Class | CodeGenTokenizer |
| Padding Token | <|im_end|> |
| Vocabulary Size | 50297 |
| Torch Data Type | float16 |
Best Alternatives to Phi 2 Instruct V0.1
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Phi 2 Instruct Apo | 2K / 5.6 GB | 33 | 0 |
| EEVE Korean Instruct 2.8B V1.0 | 2K / 5.7 GB | 3551 | 20 |
| Att Model | 2K / 5.7 GB | 9 | 0 |
| Eeve2.8 Base | 2K / 5.7 GB | 5 | 0 |
| Eeve2.8 Ko | 2K / 5.7 GB | 18 | 0 |
| ... Instruct 2.8B V1.0 20240430 2 | 2K / 2.9 GB | 12 | 0 |
| Phi 2 Code Instruct | 2K / 5.6 GB | 20 | 4 |
| Dolphin 2 6 Phi 2 | 0K / 5.6 GB | 658 | 192 |
| Phi 2 Evol Instruct Chinese | 0K / 5.6 GB | 0 | 4 |
| Phi 2 Dolly Instruction Polish | 0K / 5.6 GB | 0 | 2 |
Rank the Phi 2 Instruct V0.1 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? 40013 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217