LLM-FineTuning-Large-Language-Models

1
{
2
 "cells": [
3
  {
4
   "cell_type": "markdown",
5
   "metadata": {},
6
   "source": [
7
    "### Checkout my [Twitter(@rohanpaul_ai)](https://twitter.com/rohanpaul_ai) for daily LLM bits"
8
   ]
9
  },
10
  {
11
   "cell_type": "markdown",
12
   "metadata": {},
13
   "source": [
14
    "## LlaMa-2-FineTuning on any of your custom dataset"
15
   ]
16
  },
17
  {
18
   "cell_type": "code",
19
   "execution_count": null,
20
   "metadata": {},
21
   "outputs": [],
22
   "source": [
23
    "!pip install trl[peft]"
24
   ]
25
  },
26
  {
27
   "cell_type": "markdown",
28
   "metadata": {},
29
   "source": [
30
    "# From source. For running the examples in the repository\n",
31
    "# Clone the repository and install it with pip:"
32
   ]
33
  },
34
  {
35
   "cell_type": "code",
36
   "execution_count": null,
37
   "metadata": {},
38
   "outputs": [],
39
   "source": [
40
    "!git clone https://github.com/lvwerra/trl.git\n",
41
    "\n",
42
    "!cd trl/\n",
43
    "\n",
44
    "! pip install ."
45
   ]
46
  },
47
  {
48
   "cell_type": "markdown",
49
   "metadata": {},
50
   "source": [
51
    "```py\n",
52
    "!python examples/scripts/sft_trainer.py \\\n",
53
    "  --model_name meta-llama/Llama-2-7b-hf \\\n",
54
    "  --dataset_name timdettmers/openassistant-guanaco \\\n",
55
    "  --load_in_4bit \\\n",
56
    "  --use_peft \\\n",
57
    "  --batch_size 4 \\\n",
58
    "  --gradient_accumulation_steps 2\n",
59
    "\n",
60
    "```"
61
   ]
62
  }
63
 ],
64
 "metadata": {
65
  "language_info": {
66
   "name": "python"
67
  },
68
  "orig_nbformat": 4
69
 },
70
 "nbformat": 4,
71
 "nbformat_minor": 2
72
}
73