BEncoderRT commited on
Commit
4bc9f2a
·
verified ·
1 Parent(s): 4a1f78e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -0
README.md CHANGED
@@ -12,8 +12,16 @@ tags:
12
  - Instruction-Tuning
13
  - peft
14
  ---
 
 
 
 
 
 
15
  # QLoRA Instruction Tuning on Pythia-1B
16
 
 
 
17
  This repository provides a **Hugging Face–compatible LoRA adapter** trained via **QLoRA (4-bit quantization + LoRA adapters)** on the **EleutherAI Pythia-1B-deduped** base model.
18
 
19
  The project focuses on **producing and publishing a reusable LoRA adapter** using a modern, memory-efficient instruction-tuning pipeline built with Hugging Face Transformers, PEFT, and BitsAndBytes. It is designed for **learning, experimentation, and small-GPU environments (e.g. Colab)**.
 
12
  - Instruction-Tuning
13
  - peft
14
  ---
15
+
16
+ “Predict the next token”
17
+ not
18
+ “Obey the instruction”
19
+
20
+
21
  # QLoRA Instruction Tuning on Pythia-1B
22
 
23
+
24
+
25
  This repository provides a **Hugging Face–compatible LoRA adapter** trained via **QLoRA (4-bit quantization + LoRA adapters)** on the **EleutherAI Pythia-1B-deduped** base model.
26
 
27
  The project focuses on **producing and publishing a reusable LoRA adapter** using a modern, memory-efficient instruction-tuning pipeline built with Hugging Face Transformers, PEFT, and BitsAndBytes. It is designed for **learning, experimentation, and small-GPU environments (e.g. Colab)**.