Quantization made by Richard Erkhov.

SuperPlatty-30B - GGUF

Model creator: https://huggingface.co/garage-bAInd/
Original model: https://huggingface.co/garage-bAInd/SuperPlatty-30B/

Name	Quant method	Size
SuperPlatty-30B.Q2_K.gguf	Q2_K	11.22GB
SuperPlatty-30B.IQ3_XS.gguf	IQ3_XS	12.4GB
SuperPlatty-30B.IQ3_S.gguf	IQ3_S	13.1GB
SuperPlatty-30B.Q3_K_S.gguf	Q3_K_S	13.1GB
SuperPlatty-30B.IQ3_M.gguf	IQ3_M	13.86GB
SuperPlatty-30B.Q3_K.gguf	Q3_K	14.69GB
SuperPlatty-30B.Q3_K_M.gguf	Q3_K_M	14.69GB
SuperPlatty-30B.Q3_K_L.gguf	Q3_K_L	16.09GB
SuperPlatty-30B.IQ4_XS.gguf	IQ4_XS	16.28GB
SuperPlatty-30B.Q4_0.gguf	Q4_0	17.1GB
SuperPlatty-30B.IQ4_NL.gguf	IQ4_NL	17.19GB
SuperPlatty-30B.Q4_K_S.gguf	Q4_K_S	17.21GB
SuperPlatty-30B.Q4_K.gguf	Q4_K	18.27GB
SuperPlatty-30B.Q4_K_M.gguf	Q4_K_M	18.27GB
SuperPlatty-30B.Q4_1.gguf	Q4_1	18.98GB
SuperPlatty-30B.Q5_0.gguf	Q5_0	20.86GB
SuperPlatty-30B.Q5_K_S.gguf	Q5_K_S	20.86GB
SuperPlatty-30B.Q5_K.gguf	Q5_K	21.46GB
SuperPlatty-30B.Q5_K_M.gguf	Q5_K_M	21.46GB
SuperPlatty-30B.Q5_1.gguf	Q5_1	22.74GB
SuperPlatty-30B.Q6_K.gguf	Q6_K	24.85GB
SuperPlatty-30B.Q8_0.gguf	Q8_0	32.19GB

Original model description:

language: - en tags: - llama license: other metrics: - MMLU - ARC - HellaSwag - TruthfulQA

Information

SuperPlatty-30B is a merge of garage-bAInd/Platypus-30B and kaiokendev/SuperCOT-LoRA

Model Details

Trained by: Platypus-30B trained by Cole Hunter & Ariel Lee; SuperCOT-LoRA trained by kaiokendev.
Model type: SuperPlatty-30B is an auto-regressive language model based on the LLaMA transformer architecture.
Language(s): English
License for base weights: License for the base LLaMA model's weights is Meta's non-commercial bespoke license.

Hyperparameter	Value
$n_\text{parameters}$	33B
$d_\text{model}$	6656
$n_\text{layers}$	60
$n_\text{heads}$	52

Reproducing Evaluation Results

Install LM Evaluation Harness:

git clone https://github.com/EleutherAI/lm-evaluation-harness
cd lm-evaluation-harness
pip install -e .

Each task was evaluated on a single A100 80GB GPU.

ARC:

python main.py --model hf-causal-experimental --model_args pretrained=garage-bAIdnd/SuperPlatty-30B --tasks arc_challenge --batch_size 1 --no_cache --write_out --output_path results/SuperPlatty-30B/arc_challenge_25shot.json --device cuda --num_fewshot 25

HellaSwag:

python main.py --model hf-causal-experimental --model_args pretrained=garage-bAIdnd/SuperPlatty-30B --tasks hellaswag --batch_size 1 --no_cache --write_out --output_path results/SuperPlatty-30B/hellaswag_10shot.json --device cuda --num_fewshot 10

MMLU:

python main.py --model hf-causal-experimental --model_args pretrained=garage-bAIdnd/SuperPlatty-30B --tasks hendrycksTest-* --batch_size 1 --no_cache --write_out --output_path results/SuperPlatty-30B/mmlu_5shot.json --device cuda --num_fewshot 5

TruthfulQA:

python main.py --model hf-causal-experimental --model_args pretrained=garage-bAIdnd/SuperPlatty-30B --tasks truthfulqa_mc --batch_size 1 --no_cache --write_out --output_path results/SuperPlatty-30B/truthfulqa_0shot.json --device cuda

Limitations and bias

The base LLaMA model is trained on various data, some of which may contain offensive, harmful, and biased content that can lead to toxic behavior. See Section 5.1 of the LLaMA paper. We have not performed any studies to determine how fine-tuning on the aforementioned datasets affect the model's behavior and toxicity. Do not treat chat responses from this model as a substitute for human judgment or as a source of truth. Please use responsibly.

Citations

@article{touvron2023llama,
  title={LLaMA: Open and Efficient Foundation Language Models},
  author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
  journal={arXiv preprint arXiv:2302.13971},
  year={2023}
}
@article{hu2021lora,
  title={LoRA: Low-Rank Adaptation of Large Language Models},
  author={Hu, Edward J. and Shen, Yelong and Wallis, Phillip and Allen-Zhu, Zeyuan and Li, Yuanzhi and Wang, Shean and Chen, Weizhu},
  journal={CoRR},
  year={2021}
}

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	57.89
ARC (25-shot)	65.78
HellaSwag (10-shot)	83.95
MMLU (5-shot)	62.57
TruthfulQA (0-shot)	53.52
Winogrande (5-shot)	80.35
GSM8K (5-shot)	9.63
DROP (3-shot)	49.44