3n0chK4n enochkan

4 Steps in Running LLaMA-7B on a M1 MacBook

The large language models usability

The problem with large language models is that you can’t run these locally on your laptop. Thanks to Georgi Gerganov and his llama.cpp project, it is now possible to run Meta’s LLaMA on a single computer without a dedicated GPU.

Running LLaMA

There are multiple steps involved in running LLaMA locally on a M1 Mac after downloading the model weights.

How to install dlib v19.9 or newer (w/ python bindings) from github on macOS and Ubuntu

Pre-reqs:

Have Python 3 installed. On macOS, this could be installed from homebrew or even via standard Python 3.6 downloaded installer from https://www.python.org/download. On Linux, just use your package manager.
On macOS:
- Install XCode from the Mac App Store (or install the XCode command line utils).
- Have homebrew installed
On Linux:

"It has become appallingly obvious that our technology has exceeded our humanity." -Albert Einstein

##Intro

Technology is constantly evolving. For better or for worse, technology has moved forward faster than anything in history. With new devices coming out every few years and our human needs craving for bigger and better things, it make me wonder if technology has moved on in all aspects of life. For example, say the medical field. The medical field is one of the oldest fields in history and having something so new like technology can pose a problem especially when it comes to human life.

The rapid growth of technology has given many fields a chance to grow. However, the medical field has no really embraced its techy roots. Humans are skeptical when it comes to machines for fear that it can be used for the wrong reasons or that the machines lack a conscious. Regardless, it has not stopped dev

	# Clone llama.cpp
	git clone https://github.com/ggerganov/llama.cpp.git
	cd llama.cpp

	# Build it
	make clean
	LLAMA_METAL=1 make

	# Download model
	export MODEL=llama-2-13b-chat.ggmlv3.q4_0.bin

	##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
	## Created by: Hang Zhang, Rutgers University, Email: [email protected]
	## Modified by Thomas Wolf, HuggingFace Inc., Email: [email protected]
	## Copyright (c) 2017-2018
	##
	## This source code is licensed under the MIT-style license found in the
	## LICENSE file in the root directory of this source tree
	##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

	"""Encoding Data Parallel"""

	#!/usr/bin/python
	# -- coding: utf-8 --

	from pyspark import SparkContext
	from pyspark.sql import SparkSession
	from pyspark.sql.functions import *
	from pyspark.sql.types import StringType
	from pyspark import SQLContext
	from itertools import islice
	from pyspark.sql.functions import col

	from __future__ import print_function
	import imageio
	from PIL import Image
	import numpy as np
	import keras

	from keras.layers import Input, Dense, Conv2D, MaxPooling2D, AveragePooling2D, ZeroPadding2D, Dropout, Flatten, Concatenate, Reshape, Activation
	from keras.models import Model
	from keras.regularizers import l2
	from keras.optimizers import SGD

	""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
	import numpy as np
	import cPickle as pickle
	import gym

	# hyperparameters
	H = 200 # number of hidden layer neurons
	batch_size = 10 # every how many episodes to do a param update?
	learning_rate = 1e-4
	gamma = 0.99 # discount factor for reward

	import tensorflow as tf
	import numpy as np
	import os

	import zconfig
	import utils

	class DenoisingAutoencoder(object):

	""" Implementation of Denoising Autoencoders using TensorFlow.