abdullahalbaki/phi3-personal-al-v1

This model is a fine-tuned version of the Microsoft Phi-3-mini-4k-instruct architecture. It has been specifically trained using QLoRA (4-bit quantization) to provide personalized information about Alamin Sarkar. — Developed by: Abdullah Al Baki — Base Model: microsoft/Phi-3-mini-4k-instruct — Language: English — Fine-tuning Method: SFT (Supervised Fine-Tuning) with LoRA — Training Tool: Hugging Face transformers, peft, and bitsandbytes The model was trained on a custom dataset containing personal information, professional expertise, and biographical data of Alamin Sarkar. — Epochs: 50 — Learning Rate: 5e-4 — Quantization: 4-bit (NF4) — Target Modules: qkvproj — Training Hardware:** NVIDIA T4 GPU (Google Colab) You can use this model directly with the transformers library: «`python from transformers import AutoModelForCausalLM, AutoTokenizer tokenizer = AutoTokenizer.frompretrained(modelid) model = AutoModelForCausalLM.frompretrained(modelid, devicemap=»auto», trustremote_code=True) prompt = «nWho is Alamin Sarkar?nn» inputs = tokenizer(pro

Модальности:
Генерация текста

Области применения:
Диалог / чат

Задача: Генерация текста
Автор: abdullahalbaki
Теги: peft, phi3, lora, personalized, fine-tuned, conversational, custom_code, en
Лайков: 3  |  Загрузок: 0

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.