Update README.md

README.md

@@ -1,21 +1,17 @@
 ---
 library_name: peft
+license: apache-2.0
+datasets:
+- neuralwork/fashion-style-instruct
+language:
+- en
 ---
-## Training procedure
+## Style-Instruct Mistral 7B
+Mistral 7B instruct fine-tuned on the [neuralwork/fashion-style-instruct]() dataset with LoRA and 4bit quantization. See the blog [post]() and Github [repository](https://github.com/neuralwork/instruct-finetune-mistral)
+for training details. This model is trained with body type / personal style descriptions as input, target events (e.g. casual date, business meeting) as context and outfit combination suggestions as output.
 
 
-The following `bitsandbytes` quantization config was used during training:
-- quant_method: bitsandbytes
-- load_in_8bit: False
-- load_in_4bit: True
-- llm_int8_threshold: 6.0
-- llm_int8_skip_modules: None
-- llm_int8_enable_fp32_cpu_offload: False
-- llm_int8_has_fp16_weight: False
-- bnb_4bit_quant_type: fp4
-- bnb_4bit_use_double_quant: False
-- bnb_4bit_compute_dtype: float32
-### Framework versions
-
-
-- PEFT 0.4.0
+## Usage
+This repo contains the LoRA parameters of the fine-tuned Mistral 7B model. To perform inference, load and use the model as follows:
+```
+```