CorticalStack_-_mistral-7b-tak-stack-dpo-gguf

RichardErkhov/CorticalStack_-_mistral-7b-tak-stack-dpo-gguf

CorticalStack_-_mistral-7b-tak-stack-dpo-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

Downloads · 30 days

278

13% of all-time downloads

All-time downloads

2.1K

Public

Repo size

82.2 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.gguf95.2 GB · 100%

At a glance

Access: Public
Created: Sep 9, 2024
Updated: Sep 9, 2024
SHA: 1ea66963

Model card

From the Hugging Face model README

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mistral-7b-tak-stack-dpo - GGUF

Model creator: https://huggingface.co/CorticalStack/
Original model: https://huggingface.co/CorticalStack/mistral-7b-tak-stack-dpo/

Name	Quant method	Size
mistral-7b-tak-stack-dpo.Q2_K.gguf	Q2_K	2.53GB
mistral-7b-tak-stack-dpo.IQ3_XS.gguf	IQ3_XS	2.81GB
mistral-7b-tak-stack-dpo.IQ3_S.gguf	IQ3_S	2.96GB
mistral-7b-tak-stack-dpo.Q3_K_S.gguf	Q3_K_S	2.95GB
mistral-7b-tak-stack-dpo.IQ3_M.gguf	IQ3_M	3.06GB
mistral-7b-tak-stack-dpo.Q3_K.gguf	Q3_K	3.28GB
mistral-7b-tak-stack-dpo.Q3_K_M.gguf	Q3_K_M	3.28GB
mistral-7b-tak-stack-dpo.Q3_K_L.gguf	Q3_K_L	3.56GB
mistral-7b-tak-stack-dpo.IQ4_XS.gguf	IQ4_XS	3.67GB
mistral-7b-tak-stack-dpo.Q4_0.gguf	Q4_0	3.83GB
mistral-7b-tak-stack-dpo.IQ4_NL.gguf	IQ4_NL	3.87GB
mistral-7b-tak-stack-dpo.Q4_K_S.gguf	Q4_K_S	3.86GB
mistral-7b-tak-stack-dpo.Q4_K.gguf	Q4_K	4.07GB
mistral-7b-tak-stack-dpo.Q4_K_M.gguf	Q4_K_M	4.07GB
mistral-7b-tak-stack-dpo.Q4_1.gguf	Q4_1	4.24GB
mistral-7b-tak-stack-dpo.Q5_0.gguf	Q5_0	4.65GB
mistral-7b-tak-stack-dpo.Q5_K_S.gguf	Q5_K_S	4.65GB
mistral-7b-tak-stack-dpo.Q5_K.gguf	Q5_K	4.78GB
mistral-7b-tak-stack-dpo.Q5_K_M.gguf	Q5_K_M	4.78GB
mistral-7b-tak-stack-dpo.Q5_1.gguf	Q5_1	5.07GB
mistral-7b-tak-stack-dpo.Q6_K.gguf	Q6_K	5.53GB
mistral-7b-tak-stack-dpo.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0 tags:

dpo base_model:
mistralai/Mistral-7B-v0.1 dataset:
CorticalStack/tak-stack-dpo

mistral-7b-tak-stack-dpo

mistral-7b-tak-stack-dpo is a DPO fine-tuned version of mistralai/Mistral-7B-v0.1 using the CorticalStack/tak-stack-dpo dataset.

LoRA

r: 32
LoRA alpha: 32
LoRA dropout: 0.05

Training arguments

Batch size: 4
Gradient accumulation steps: 4
Optimizer: paged_adamw_32bit
Max steps: 100
Learning rate: 5e-05
Learning rate scheduler type: cosine
Beta: 0.1
Max prompt length: 1024
Max length: 1536

Created: Sep 9, 2024
Updated: Sep 9, 2024