Discover gists

Is telling a model to "not hallucinate" absurd?

Can you tell an LLM "don't hallucinate" and expect it to work? my gut reaction was "oh this is so silly" but upon some reflection, it really isn't. There is actually no reason why it shouldn't work, especially if it was preference-fine-tuned on instructions with "don't hallucinate" in them, and if it a recent commercial model, it likely was.

What does an LLM need in order to follow an instruction? It needs two things:

an ability to perform then task. Something in its parameters/mechanism should be indicative of the task objective, in a way that can be influenced. (In our case, it should "know" when it hallucinates, and/or should be able to change or adapt its behavior to reduce the chance of hallucinations.)
an ability to ground the instruction: the model should be able to associate the requested behavior with its parameters/mechanisms. (In our case, the model should associate "don't hallucinate" with the behavior related to 1).

puppeteer dependeces in ubuntu 18.04 (Bionic)

error while loading shared libraries: libnss3.so: cannot open shared object file: No such file or directory

sudo apt-get install libnss3

error while loading shared libraries: libXss.so.1: cannot open shared object file: No such file or directory

sudo apt-get install libxss1

	#!/bin/bash

	# SPDX-FileCopyrightText: © 2024 Ryan Carsten Schmidt <https://github.com/ryandesign>
	# SPDX-License-Identifier: MIT
	#
	# Adds ext4 driver to OpenCore Legacy Patcher so that a Linux partition can be selected.
	# Based on instructions by rikerjoe at:
	# https://tinkerdifferent.com/threads/dual-boot-linux-mint-21-2-and-macos-with-opencore-legacy-patcher-and-opencore-bootpicker.3115/

	set -u

	Latency Comparison Numbers (~2012)
	----------------------------------
	L1 cache reference 0.5 ns
	Branch mispredict 5 ns
	L2 cache reference 7 ns 14x L1 cache
	Mutex lock/unlock 25 ns
	Main memory reference 100 ns 20x L2 cache, 200x L1 cache
	Compress 1K bytes with Zippy 3,000 ns 3 us
	Send 1K bytes over 1 Gbps network 10,000 ns 10 us
	Read 4K randomly from SSD* 150,000 ns 150 us ~1GB/sec SSD

	# Create EKS Cluster

	eksctl create cluster --name=observability \
	--region=us-east-1 \
	--zones=us-east-1a,us-east-1b \
	--without-nodegroup

	eksctl utils associate-iam-oidc-provider \
	--region us-east-1 \
	--cluster observability \

	;; Challenge: find all the bugs in this small functional compiler
	#lang racket

	(provide (all-defined-out))

	(define verbose-mode (make-parameter #t))
	(define i-am-a-mac (make-parameter #f))

	;;
	;; Our compiler will have several layers

	{
	"Create public function": {
	"prefix": "pubf",
	"body": [
	"public function $1()",
	"{",
	"\t$2",
	"}",
	],
	"description": "Create public function"

	Registered Name: https://zhile.io
	License Key: 48891cf209c6d32bf4

	import SwiftUI

	struct ContentView: View {
	typealias Item = (Color, CGFloat)

	@State private var columnCount = 3

	@State private var isWaterfall = true

	@State var items: [Item] = [

	// *****************************************************************************************
	// *****************************************************************************************
	// *****************************************************************************************
	// **************************************PC SA Main Script******************************
	// *****************************************************************************************
	// *****************************************************************************************
	// *****************************************************************************************

	SCRIPT_NAME MAIN //NEW MAIN