Adapter Base model:abacusai/smaug-72b-... Base model:finetune:abacusai/s... Dataset:neuralnovel/neural-dpo Finetuned Lora Model-index Region:us Safetensors
Model Card on HF 🤗: https://huggingface.co/ConvexAI/Luminex-72B-v0.1
Luminex 72B V0.1 Benchmarks
ARC: 43.43 vs 96.7 (so35)-55.1%
HellaSwag: 86.66 vs 95.3 (gpt4)-9.1%
MMLU: 73.36 vs 88.3 (so35)-16.9%
TruthfulQA: 41.85 vs 59 (gpt4)-29.1%
WinoGrande: 76.16 vs 87.5 (gpt4)-13%
GSM8K: 75.44 vs 96.4 (so35)-21.7%
LLME Score: 0.17931
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)
Luminex 72B V0.1 Parameters and Internals
| Model Type |
| |
| Additional Notes |
|
| LLM Name | Luminex 72B V0.1 |
| Repository 🤗 | https://huggingface.co/ConvexAI/Luminex-72B-v0.1 |
| Base Model(s) | |
| Model Size | 72b |
| Required VRAM | 0.1 GB |
| Updated | 2025-02-22 |
| Maintainer | ConvexAI |
| Model Files | |
| Model Architecture | Adapter |
| License | other |
| Model Max Length | 4096 |
| Is Biased | none |
| Tokenizer Class | GPT2Tokenizer |
| Padding Token | <|im_end|> |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | q_proj|v_proj |
| LoRA Alpha | 16 |
| LoRA Dropout | 0.1 |
| R Param | 8 |
| Errors | replace |
Best Alternatives to Luminex 72B V0.1
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...s V2.5 Qwen2 72B QLoRA Adapter | 0K / 0 GB | 5 | 4 |
| Qwen2 37B Pruned | 0K / 0.8 GB | 7 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than ConvexAI/Luminex-72B-v0.1 .
Rank the Luminex 72B 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? 43470 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227