Adapter Base model:thebloke/mistral-7b... Dataset:knkarthick/dialogsum En Finetuned Generated from trainer Gptq Instruct License:apache-2.0 Lora Peft Quantized Region:us Safetensors Summarization Tensorboard
Mistral Finetuned DialogSumm Benchmarks
LMSys ELO: 780 vs 1158 (GPT-4) -32.6%
HF Avg Score: 47.9 vs 86.1 (GPT-4) -44.4%
ARC: 62-35.6%
HellaSwag: 45.2-52.6%
MMLU: 43.6-49.6%
GSM8K: 25.1-72.7%
OpenCompass Avg: 44.8 vs 73.3 (GPT-4) -38.9%
Exam: 60.4-21.8%
Language: 12.5-79.8%
Knowledge: 42.5-42.2%
Understanding: 62-11.4%
Reasoning: 44-40.9%
nn.n% — How the model compares to the GPT-4.
Rank the Mistral Finetuned DialogSumm 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! 🌟
Best Alternatives to Mistral Finetuned DialogSumm
Best Alternatives |
HF Rank |
Context/RAM |
Downloads |
Likes |
|---|---|---|---|---|
| Term Qwen2 7 Json Lora | — | 0K / 0.1 GB | 30 | 0 |
| Futadom Mistral | — | 0K / 0.2 GB | 1 | 1 |
| ...ralic 7B 1 GPTQ LORA CAUSAL LM | — | 0K / 0.3 GB | 1 | 1 |
| Falcon 7B Instruct 4bit Lora | — | 0K / 0 GB | 0 | 1 |
| Zephyr 7B Beta Agent Instruct | — | 0K / 0.3 GB | 15 | 1 |
| Mistral Wikitable FV | — | 0K / 0.3 GB | 17 | 0 |
| Mistral Charttotext FV | — | 0K / 0.3 GB | 13 | 0 |
| Mistral Numericnlg FV | — | 0K / 0.3 GB | 8 | 0 |
| ...ral 7b WizardLMEvolInstruct70k | — | 0K / 0 GB | 1 | 11 |
| Thai Buffala Lora 7B V0.1 | — | 0K / 0 GB | 7 | 10 |
Mistral Finetuned DialogSumm Parameters and Internals
| LLM Name | Mistral Finetuned DialogSumm |
| Repository | Open on 🤗 |
| Base Model(s) | |
| Model Size | 7b |
| Required VRAM | 0 GB |
| Updated | 2024-07-05 |
| Maintainer | Villian7 |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| GPTQ Quantization | Yes |
| Quantization Type | gptq |
| Model Architecture | Adapter |
| License | apache-2.0 |
| Is Biased | none |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | </s> |
| PEFT Type | LORA |
| LoRA Model | Yes |
| PEFT Target Modules | q_proj|v_proj |
| LoRA Alpha | 16 |
| LoRA Dropout | 0.05 |
| R Param | 16 |
What open-source LLMs or SLMs are you in search of? 34431 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024042801