I am really enjoying this version of Cinder. More information coming. Training data similar to openhermes2.5 with some added math, STEM, and reasoning mostly from OpenOrca. As well as Cinder character specific data, a mix of RAG generated Q and A of world knowledge, STEM topics, and Cinder Character data. I suplimented the Cinder character with an abreviated Samantha dataset edited for Cinder and removed a lot of the negative responses. Model Overview Cinder is an AI chatbot tailored for engaging users in scientific and educational conversations, offering companionship, and sparking imaginative exploration.

image/png

Chat example from LM Studio:

image/png

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 58.86
AI2 Reasoning Challenge (25-Shot) 58.28
HellaSwag (10-Shot) 74.04
MMLU (5-Shot) 54.46
TruthfulQA (0-shot) 44.50
Winogrande (5-shot) 74.66
GSM8k (5-shot) 47.23

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 10.86
IFEval (0-Shot) 23.57
BBH (3-Shot) 22.45
MATH Lvl 5 (4-Shot) 0.00
GPQA (0-shot) 4.25
MuSR (0-shot) 1.97
MMLU-PRO (5-shot) 12.90
Downloads last month
161
Safetensors
Model size
2.78B params
Tensor type
FP16
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Evaluation results