---
base_model: unsloth/Phi-3-mini-4k-instruct-bnb-4bit
language:
- en
license: creativeml-openrail-m
tags:
- text-generation-inference
- transformers
- unsloth
- trl
- sft
---

# Note
- **This is an Experiment to generate Clinical Trial Synopsis. Will be making a better one Soon! Stay Updated**
- **LORA:** [ArvindSharma18/Phi-3-mini-4k-instruct-bnb-4bit-Clinical-Trail-Exp](https://huggingface.co/ArvindSharma18/Phi-3-mini-4k-instruct-bnb-4bit-Clinical-Trail-Exp)

# How to Use

**Note:** May Hallucinate on Trial attributes (I am planning to use this model as a foundational model for more downstream tasks I have in my pipeline) or Repeat (especially in case of Eligibility Criteria generation) in case of some trials. Working on making it more reliable.

```python
from unsloth import FastLanguageModel
import torch
max_seq_length = 4096
dtype = torch.float16
load_in_4bit = True

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "ArvindSharma18/Phi-3-mini-4k-instruct-bnb-4bit-Clinical-Trail-Merged-Exp", # "unsloth/mistral-7b" for 16bit loading
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit
)
FastLanguageModel.for_inference(model)
inputs = tokenizer(
[
    "Write Clinical Trial Summary for Effects of High-protein Milk Supplementation on Muscular Strength and Power, Body Composition, and Skeletal Muscle Regulatory Markers Following Heavy Resistance Training in Resistance-trained Men"
], return_tensors = "pt").to("cuda")
from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer, skip_prompt = True)
_ = model.generate(input_ids = inputs.input_ids, attention_mask = inputs.attention_mask,
                   streamer = text_streamer, max_new_tokens = 4096, do_sample=True)
```

# Uploaded  model

- **Developed by:** ArvindSharma18
- **Finetuned from model :** unsloth/Phi-3-mini-4k-instruct-bnb-4bit


[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)