CapybaraHermes-2.5-Mistral-7B

This model is the launching partner of the capybara-dpo dataset build with ⚗️ distilabel. It's a preference tuned OpenHermes-2.5-Mistral-7B.

CapybaraHermes has been preference tuned with LoRA and TRL for 3 epochs using argilla's dpo mix 7k.

To test the impact on multi-turn performance we have used MTBench. We also include the Nous Benchmark results and Mistral-7B-Instruct-v0.2 for reference as it's a strong 7B model on MTBench:

Model	AGIEval	GPT4All	TruthfulQA	Bigbench	MTBench First Turn	MTBench Second Turn	Nous avg.	MTBench avg.
argilla/CapybaraHermes-2.5-Mistral-7B	43.8	73.35	57.07	42.44	8.24375	7.5625	54.16	7.903125
teknium/OpenHermes-2.5-Mistral-7B	42.75	72.99	52.99	40.94	8.25	7.2875	52.42	7.76875
Mistral-7B-Instruct-v0.2	38.5	71.64	66.82	42.29	7.8375	7.1	54.81	7.46875

The most interesting aspect in the context of the capybara-dpo dataset is the increased performance in MTBench Second Turn scores.

For the merge lovers, we also preference tuned Beagle14-7B with a mix of capybara-dpo and distilabel orca pairs using the same recipe as NeuralBeagle (see YALL - Yet Another LLM Leaderboard for reference):

Model	AGIEval	GPT4All	TruthfulQA	Bigbench	Average
DistilabelBeagle14-7B	45.29	76.92	71.66	48.78	60.66

Model Details

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

• Developed by: Argilla
• Shared by [optional]: Argilla
• Model type: 7B chat model
• Language(s) (NLP): English
• License: Same as OpenHermes
• Finetuned from model [optional]: OpenHermes-2.5-Mistral-7B

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.14
AI2 Reasoning Challenge (25-Shot)	65.78
HellaSwag (10-Shot)	85.45
MMLU (5-Shot)	63.13
TruthfulQA (0-shot)	56.91
Winogrande (5-shot)	78.30
GSM8k (5-shot)	59.29