Autotrain compatible Base model:bee-spoke-data/zeph... Base model:finetune:bee-spoke-... Conversational Dataset:huggingfaceh4/ultrafee... Endpoints compatible Generated from trainer Mistral Region:us Safetensors Tensorboard
Model Card on HF 🤗: https://huggingface.co/BEE-spoke-data/zephyr-220m-dpo-full
Zephyr 220M DPO Full Benchmarks
ARC: 25.43 vs 96.7 (so35)-73.7%
HellaSwag: 29.15 vs 95.3 (gpt4)-69.4%
MMLU: 26.43 vs 88.3 (so35)-70.1%
TruthfulQA: 43.44 vs 59 (gpt4)-26.4%
WinoGrande: 50.99 vs 87.5 (gpt4)-41.7%
GSM8K: 0.53 vs 96.4 (so35)-99.5%
LLME Score: 0.17847
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)
Zephyr 220M DPO Full Parameters and Internals
| LLM Name | Zephyr 220M DPO Full |
| Repository 🤗 | https://huggingface.co/BEE-spoke-data/zephyr-220m-dpo-full |
| Base Model(s) | |
| Model Size | 220m |
| Required VRAM | 0.4 GB |
| Updated | 2025-09-16 |
| Maintainer | BEE-spoke-data |
| Model Type | mistral |
| Model Files | |
| Model Architecture | MistralForCausalLM |
| License | apache-2.0 |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.37.0.dev0 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <unk> |
| Vocabulary Size | 32128 |
| Torch Data Type | bfloat16 |
Best Alternatives to Zephyr 220M DPO Full
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Zephyr 220M Sft Full | 2K / 0.4 GB | 732 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than BEE-spoke-data/zephyr-220m-dpo-full .
Rank the Zephyr 220M DPO Full 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? 51408 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124