You can deploy this model and start earning money today!

RichardErkhov/
Vezora_-_Mistral-14b-Merge-Base-gguf

Name: Vezora_-_Mistral-14b-Merge-Base-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Mistral-14b-Merge-Base - GGUF

Model creator: https://huggingface.co/Vezora/
Original model: https://huggingface.co/Vezora/Mistral-14b-Merge-Base/

Name	Quant method	Size
Mistral-14b-Merge-Base.Q2_K.gguf	Q2_K	4.92GB
Mistral-14b-Merge-Base.IQ3_XS.gguf	IQ3_XS	5.47GB
Mistral-14b-Merge-Base.IQ3_S.gguf	IQ3_S	5.77GB
Mistral-14b-Merge-Base.Q3_K_S.gguf	Q3_K_S	5.74GB
Mistral-14b-Merge-Base.IQ3_M.gguf	IQ3_M	5.97GB
Mistral-14b-Merge-Base.Q3_K.gguf	Q3_K	6.4GB
Mistral-14b-Merge-Base.Q3_K_M.gguf	Q3_K_M	6.4GB
Mistral-14b-Merge-Base.Q3_K_L.gguf	Q3_K_L	6.97GB
Mistral-14b-Merge-Base.IQ4_XS.gguf	IQ4_XS	7.18GB
Mistral-14b-Merge-Base.Q4_0.gguf	Q4_0	7.48GB
Mistral-14b-Merge-Base.IQ4_NL.gguf	IQ4_NL	7.57GB
Mistral-14b-Merge-Base.Q4_K_S.gguf	Q4_K_S	7.54GB
Mistral-14b-Merge-Base.Q4_K.gguf	Q4_K	7.97GB
Mistral-14b-Merge-Base.Q4_K_M.gguf	Q4_K_M	7.97GB
Mistral-14b-Merge-Base.Q4_1.gguf	Q4_1	8.3GB
Mistral-14b-Merge-Base.Q5_0.gguf	Q5_0	9.12GB
Mistral-14b-Merge-Base.Q5_K_S.gguf	Q5_K_S	9.12GB
Mistral-14b-Merge-Base.Q5_K.gguf	Q5_K	9.37GB
Mistral-14b-Merge-Base.Q5_K_M.gguf	Q5_K_M	9.37GB
Mistral-14b-Merge-Base.Q5_1.gguf	Q5_1	9.94GB
Mistral-14b-Merge-Base.Q6_K.gguf	Q6_K	10.87GB
Mistral-14b-Merge-Base.Q8_0.gguf	Q8_0	14.07GB

Original model description:

license: apache-2.0

Contributors: Nicolas Mejia Petit

License

Mistral 14b: A New Base Model

The objective of this model is to serve as a fully open source new base model with 14.3 billion parameters. It has been enhanced with a LoRa adapter attached to all 62 layers of the merged model. The model is capable of generating outputs and responding accurately to inputs. However, it tends to over-respond with unasked questions when asked to process more than 512 tokens, which is its training limit using QLoRa.

With full fine-tuning, this model has the potential to deliver excellent performance.

Qlora adapter was trained on a modified dataset of airoboros-m-7b-3.1.2, using SFT. In the Alpaca Format.

Model Creation

The model was created by merging two models: Dolphin and Zephyr, along with Meta-math7b and Speechless code, to form a single model. The layers of these two models were stacked on top of each other to create this model.

Initially, the output from the model was pure jargon. To rectify this, a LoRa adapter was trained and merged across all layers.

Useful Resources

LoRa Adapter Merging (https://gist.github.com/ChrisHayduk/1a53463331f52dca205e55982baf9930)
Model Merging (MergeKit) (https://github.com/cg123/mergekit)

Source Models

Dolphin2.1-mistral-7b by Eric Hartford (https://huggingface.co/ehartford/dolphin-2.1-mistral-7b)
Zephyr-7b-beta by HuggingFace (https://huggingface.co/HuggingFaceH4/zephyr-7b-beta)
MetaMath-Mistral-7B by meta-math (https://huggingface.co/meta-math/MetaMath-Mistral-7B)
Speechless-code-mistral-7b-v1.0 (https://huggingface.co/uukuguy/speechless-code-mistral-7b-v1.0)
Airoboros-m-7b-3.1.2 (https://huggingface.co/jondurbin/airoboros-m-7b-3.1.2)

upcoming Mistral 30b

We currently have a Mistral model with 29 billion parameters(29.2B params) in development. At present, the model's output is not yet refined and may appear as jargon. If there is interest in the community for fine-tuning this model, we are open to uploading it in its current state. Otherwise, we plan to complete our training process before making it available. You can let us know with a post in this repo's discussion's!

RichardErkhov/Vezora_-_Mistral-14b-Merge-Base-gguf