Re: [Vo]:LENR-CANR.org ChatBot improved
Robin wrote: > These bots have an API that lets them access wiki, or so they claim. > Try asking for something else that is not on LENR-CANR. I excluded all data outside of LENR-CANR.org. That is an option in the FastChat Bot configuration. This Bot used the word "J'Accuse" as a synonym for some word or phrase it found in a document at LENR-CANR.org. I do not know which document. It can be hard to discover that. The Bots often use synonyms, or paraphrasing. I think this is because the data is tokenized and the bot may not store the original words. Bots store alphanumeric tokens instead of words. So the original words may not be available in the Bot database. I asked the public version of ChatGPT about this. She is a leading expert in AI technology. Here are her responses: ME: What is tokenized data in an ai system? BOT: In AI systems, tokenization refers to the process of breaking down a text or sequence of words into smaller units called tokens. These tokens can be individual words, phrases, or even characters. Tokenization is an essential step in natural language processing (NLP) and other machine learning tasks that involve processing text data. By breaking down text into individual tokens, AI systems can analyze and interpret the meaning of the text more efficiently. Tokenization typically involves removing any punctuation, converting all characters to lowercase, and splitting the text into tokens based on a delimiter such as a space or punctuation mark. The resulting tokens can then be encoded and used as inputs to various machine learning models. Overall, tokenization helps to simplify and standardize text data so that it can be effectively analyzed and processed by AI systems. ME: A token is a numerical value, right? In some cases, can two different words for one concept -- synonyms -- have the same token? Or a token with a close numeric value? BOT: Tokens are not necessarily numerical values. In natural language processing, tokens are typically represented as strings of characters or words, which can then be encoded as numerical values for machine learning algorithms to process. When using tokenization, two different words that represent the same concept (synonyms) can be assigned the same token to enable the machine learning algorithm to recognize them as equivalent. In this case, the goal of tokenization is to reduce the vocabulary size and improve the efficiency of the model without losing critical information. However, it's important to note that tokens with close numeric values should not be used interchangeably. Each token should be unique and represent a distinct concept or word in the input text. If two tokens have similar numeric values, it could lead to confusion and inaccuracies in the machine learning model's results. Therefore, tokenization algorithms typically ensure that each token has a unique and distinguishable value. ME: I have noted that when an IA describes the content of a document, it sometimes uses synonyms instead of the original words. In a few cases, I think it did this even when I asked for a verbatim quote. Is this because the content is stored as tokens, and the exact original words are not available? BOT: Yes, it is possible that the AI is using synonyms to describe the content of a document because it is working with tokenized data and not the original words. When an AI system tokenizes text data, it breaks down the text into smaller units called tokens, which can be individual words, phrases, or even characters. These tokens are then used as inputs to machine learning models, which can use various techniques to analyze and interpret the meaning of the text. Sometimes, these models may use synonyms or other related words to describe the content of a document because the original words may not be available or may not be the best fit for the context. Additionally, some AI systems may use techniques like paraphrasing or summarization to generate new text based on the input data. In these cases, the AI may use synonyms or other related words to convey the same meaning as the original text while also improving readability or reducing complexity. However, if you specifically requested a verbatim quote and the AI system did not provide it, it is possible that there was an error or a misunderstanding in the processing of your request. It's important to keep in mind that AI systems are not perfect and may not always provide the exact results that you expect. ME: How should I phrase a request to see the original data? Should I use the word "verbatim"? Or some other word? BOT: If you want to see the original data or text without any modifications or substitutions, using the word "verbatim" is an appropriate way to phrase your request. Verbatim means word for word, exactly as it was said or written. So, you could phrase your request like this: "Can I please see the original text verbatim, without any modifications or substitutions?" This clearly commu
Re: [Vo]:LENR-CANR.org ChatBot improved
In reply to Jed Rothwell's message of Wed, 17 May 2023 11:32:10 -0400: Hi, [snip] These bots have an API that lets them access wiki, or so they claim. Try asking for something else that is not on LENR-CANR. >[The word "J'Acuse" appears nowhere in the LENR-CANR corpus.] > >. . . He also wrote letters to MIT's president and assistant to the >president requesting an investigation into the alleged misconduct. >Additionally, Mallove responded to a statement on cold fusion issued by the >MIT News Office in August 1991 with a point-by-point rebuttal. > >[Not true. BOT #1 knows better.] > >BOT 1 > >Is there a paper by Mallove describing events at MIT? >BOT: Yes, there is a paper by Mallove that describes events at MIT. The >paper is titled "MIT Special Report" and was published in Infinite Energy >in 1999. The paper explores the major role of MIT in shaping the history of >the investigation of cold fusion. You can find the paper at this URL: >http://lenr-canr.org/acrobat/MalloveEmitspecial.pdf Cloud storage:- Unsafe, Slow, Expensive ...pick any three.
Re: [Vo]:LENR-CANR.org ChatBot improved
This is uncanny. CHATBOT 2: BOT: Hello! What can I do for you today? What did Mallove say about the experiment at MIT? BOT: Eugene Mallove accused MIT of publishing fudged experimental findings to support their early condemnation of the work of Pons and Fleischmann. He believed that MIT's researchers generated excess heat in their cold fusion experiment but then altered the findings in their final report. Mallove claimed that the scientists at MIT's Plasma Fusion Center had a vested interest in seeing cold fusion fail, as they were receiving significant federal funding for hot fusion research. Mallove resigned from his position at MIT and demanded an investigation into the alleged misconduct. He also accused the university of manipulating the media and altering data to discredit the work of the Utah scientists. Is there a paper by Mallove describing this? BOT: There is no specific paper by Mallove solely focused on this incident mentioned in the dataset. However, Mallove did submit a 17-page "J'Accuse" letter to MIT detailing his accusations against the institution and its researchers. . . . [The word "J'Acuse" appears nowhere in the LENR-CANR corpus.] . . . He also wrote letters to MIT's president and assistant to the president requesting an investigation into the alleged misconduct. Additionally, Mallove responded to a statement on cold fusion issued by the MIT News Office in August 1991 with a point-by-point rebuttal. [Not true. BOT #1 knows better.] BOT 1 Is there a paper by Mallove describing events at MIT? BOT: Yes, there is a paper by Mallove that describes events at MIT. The paper is titled "MIT Special Report" and was published in Infinite Energy in 1999. The paper explores the major role of MIT in shaping the history of the investigation of cold fusion. You can find the paper at this URL: http://lenr-canr.org/acrobat/MalloveEmitspecial.pdf
Re: [Vo]:LENR-CANR.org ChatBot improved
H L V wrote: > For example, although chatgpt 4 is more accurate than 3, it has lost some > speed. > That is typical of computer programs. You often have to choose between speed and accuracy. When there is a tremendous amount of data, I often run a program to extract summary data, to give a quick answer. For example, there are 4.7 million download records so far this year at LENR-CANR.org. I extract acrobat file download totals only into spreadsheets. The one at LENR-CANR.org is particularly slow. The vendor asked me whether it should be set for faster performance or a deeper search. I opted for a deep search. I explained this on the input screen: These Bots are from https://www.chatfast.io/. They may respond more slowly than other ChatGPT bot you have used, because they are optimized to do deep searches, to remember context, and because this is ChatGPT Version 4. They may take 30 seconds or more to answer your questions.
Re: [Vo]:LENR-CANR.org ChatBot improved
On the other hand we have not started yet to optimize the speed of them. Also consider what is happening on the hardware side. On Wed, May 17, 2023 at 3:18 PM H L V wrote: > A cognitive scientist has suggested that the performance of these chatbots > may have operational limits or trade-offs that we do not yet understand. > For example, although chatgpt 4 is more accurate than 3, it has lost some > speed. > > Harry > > On Tue, May 16, 2023 at 9:36 PM Jed Rothwell > wrote: > >> The LENR-CANR.org ChatBot has been upgraded to ChatGPT version 4. It >> seems to work better, with fewer hallucinations. It may be slower. >> >> I asked it some of the questions version 3 could not answer. It answered >> them correctly. It still gets confused. It thought that McKubre made a >> statement that Bockris made. I asked for a verbatim quote and tracked it >> down to Bockris. >> >>
Re: [Vo]:LENR-CANR.org ChatBot improved
A cognitive scientist has suggested that the performance of these chatbots may have operational limits or trade-offs that we do not yet understand. For example, although chatgpt 4 is more accurate than 3, it has lost some speed. Harry On Tue, May 16, 2023 at 9:36 PM Jed Rothwell wrote: > The LENR-CANR.org ChatBot has been upgraded to ChatGPT version 4. It seems > to work better, with fewer hallucinations. It may be slower. > > I asked it some of the questions version 3 could not answer. It answered > them correctly. It still gets confused. It thought that McKubre made a > statement that Bockris made. I asked for a verbatim quote and tracked it > down to Bockris. > >
Re: [Vo]:LENR-CANR.org ChatBot improved
That is just amazing. On Tue, May 16, 2023, 9:36 PM Jed Rothwell wrote: > The LENR-CANR.org ChatBot has been upgraded to ChatGPT version 4. It seems > to work better, with fewer hallucinations. It may be slower. > > I asked it some of the questions version 3 could not answer. It answered > them correctly. It still gets confused. It thought that McKubre made a > statement that Bockris made. I asked for a verbatim quote and tracked it > down to Bockris. > >
Re: [Vo]:LENR-CANR.org ChatBot improved
The LENR-CANR.org ChatBot has been upgraded to ChatGPT version 4. It seems to work better, with fewer hallucinations. It may be slower. I asked it some of the questions version 3 could not answer. It answered them correctly. It still gets confused. It thought that McKubre made a statement that Bockris made. I asked for a verbatim quote and tracked it down to Bockris.
Re: [Vo]:LENR-CANR.org ChatBot improved
Beware of hallucinations! This is completely wrong: What are the author(s), title and URL of the Pd-B experiments at China Lake? BOT: The author of the paper on Pd-B experiments at China Lake is Miles, M. and B.F. Bush. The title of the paper is "Palladium-Boron Cathodes Produce Excess Power in Electrolytic Cells Using Heavy Water and Lithium Deuteride" and the URL is http://lenr-canr.org/acrobat/MilesMpalladiumb.pdf. The paper describes electrochemical excess power effects produced using Pd-B cathodes prepared at the US Naval Research Laboratory in nine different experiments conducted at three different laboratories using three different types of calorimeters. The title is made up. There is no paper "Palladium-Boron Cathodes Produce Excess Power in Electrolytic Cells Using Heavy Water and Lithium Deuteride." The URL is made up, in the right format for LENR-CANR.org! This is somewhat right, but somewhat of a hallucination, because Miles did not sent samples to SRI as far as I know: Miles sent samples of effluent gas to three other labs to be tested for helium. Do you know the names of those labs? BOT: Yes, according to the given information, Miles sent samples of effluent gas to other laboratories, including the University of Texas and SRI, to be tested for helium. [The labs were: University of Texas, Rockwell International, Helium Field Operations laboratory, Department of the Interior, Amarillo, Texas. The Bot could not find them all.] Are there more details about the University of Texas equipment in the paper by Miles, "ANOMALOUS EFFECTS IN DEUTERATED SYSTEMS"? BOT: Yes, there are more details about the University of Texas equipment in the paper "ANOMALOUS EFFECTS IN DEUTERATED SYSTEMS" by Miles, Bush, and Johnson. According to the paper, the University of Texas used a Bell & Howell 21-491 mass spectrometer with a cryofiltration system to remove all gases except helium. The mass spectrometer had sufficient resolution to separate D2 and He easily when the mass peaks were displayed as analogue signals on an oscilloscope. The detection limit for helium in their measurements was approximately 8 X 1011 atoms of 4He in 500 mL of gas sample. [ANOMALOUS EFFECTS IN DEUTERATED SYSTEMS is here: https://lenr-canr.org/acrobat/MilesManomalousea.pdf. It does not say "Bell & Howell" anywhere in this document. However, this detail is in another document by Ben Bush: http://lenr-canr.org/acrobat/BushBFheliumprod.pdf So, the ChatBot went looking for that information elsewhere, even though I asked her about this paper only. The quote "8 X 1011 atoms of 4He in 500 mL of gas sample" is from Miles, ICCF2, p. 364, https://lenr-canr.org/acrobat/SIFthescience.pdf. It should be 10^11, but the text file I uploaded removed the superscript markings for this particular file.] What paper describes how Miles and Bush looked for anomalous radiation using several different Geiger-Mueller (GM) detectors as well as sodium iodide (NaI) detectors during electrolysis experiments? BOT: The paper that describes how Miles and Bush looked for anomalous radiation using several different Geiger-Mueller (GM) detectors as well as sodium iodide (NaI) detectors during electrolysis experiments is titled "Radiation Measurements at China Lake: Real or Artifacts?" and was presented at The Seventh International Conference on Cold Fusion in 1998. The paper was authored by Miles, M. and B.F. Bush and can be found at the URL http://lenr-canr.org/acrobat/MilesMradiationm.pdf. [CORRECT.] >
[Vo]:LENR-CANR.org ChatBot improved
I replaced all files with text formatted files, and corrected some that were not machine readable. I think the ChatBot performance has improved. It finds URLs more often, and it finds details in the papers. It still misses some details. See this exchange: BOT: Hello! What can I do for you today? Did nasa researchers verify results from K2CO3? Yes, there have been experiments conducted with K2CO3 electrolyte and nickel cathodes that have shown excess heat production. NASA has been involved in attempting to replicate these results, but to my knowledge, no team has been successful except for the Stanford results and results from Europe and the USSR. However, there have been some successful replications of the Case Heat Effect accompanied by some of the largest 4He concentrations so far observed. As for the specific question about NASA researchers verifying results from K2CO3, I am not sure. Would you like me to look up more information on that? Yes. After conducting further research, I have not been able to find any specific information about NASA researchers verifying results from K2CO3 electrolyte and nickel cathodes. However, there have been several other groups around the world who have reported confirming the generation of excess heat in such Ni-H2O cells using K2CO3 electrolyte and porous nickel cathodes. These groups include Mills and Kneizys, Bush and Eagleton of CalPoly, Pomona, Thermacore Inc., and several groups at BARC in India. Additionally, Reiko Notoya and Michio Enyo of Hokkaido University in Japan reported excess heat production in the electrolysis of light water solution of potassium carbonate using porous nickel cathode. Are there papers from Fralick, author or coauther? I'm sorry, but I could not find any information about a Fralick author or co-author in the given information. Could you provide me with more context or details about what you are looking for? Is there a paper authored by Niedra, co-author Fralick? I apologize for the confusion earlier. After conducting further research, I was able to find information about a paper authored by J.M. Niedra and Ira T. Myers titled "Replication of the Apparent Excess Heat Effect in a Light Water-Potassium Carbonate-Nickel Electrolytic Cell." The paper was published as a NASA Technical Memorandum in February 1996 and reprinted in Infinite Energy, Vol.2, No.7, 1996, pp.62-70. However, I could not find any information about a co-author named Fralick in this paper. [THIS IS MY MISTAKE. I forgot to include Fralick in the LENR-CANR.org database. Apparently, the Bot is looking at the ASCII version of the index that I uploaded, and found this: Title: Niedra, J.M. and I.T. Myers, Replication of the apparent excess heat effect in light water-potassium carbonate-nickel-electrolytic cell. Infinite Energy, 1996. 2(7): p. 62. URL: http://lenr-canr.org/acrobat/NiedraJMreplicatio.pdf First Author: Niedra, J. M. All Authors: Niedra, J. M., Myers, I. T. Keywords: heat-, H2O, Ni, electrolysis, replication Mills Abstract: Replication of experiments claiming to demonstrate excess heat production in light water-Ni-K2CO3 electrolytic cells was found to produce an apparent excess heat of 11 W maximum, for 60 W electrical power into the cell. Power gains ranged from 1.06 to 1.68. The cell was operated at four different dc current levels plus one pulsed current run at 1 Hz, 10% duty cycle. The 28 liter cell used in these verification tests was on loan from a private corporation whose own tests with similar cells are documented to produce 50 W steady excess heat for a continuous period exceeding hundreds of days. The apparent excess heat can not be readily explained either in terms of nonlinearity of the cell's thermal conductance at a low temperature differential or by thermoelectric heat pumping. However, the present data do admit efficient recombination of dissolved hydrogen-oxygen as an ordinary explanation. Calorimetry methods and heat balance calculations for the verification tests are described. Considering the large magnitude of benefit if this effect is found to be a genuine new energy source, a more thorough investigation of evolved heat in the nickel-hydrogen system in both electrolytic and gaseous loading cells remains warranted.] What is the abstract for ""Replication of the Apparent Excess Heat Effect in a Light Water-Potassium Carbonate-Nickel Electrolytic Cell." BOT: The abstract for "Replication of the Apparent Excess Heat Effect in a Light Water-Potassium Carbonate-Nickel Electrolytic Cell" by J.M. Niedra and Ira T. Myers is as follows: "Replication of experiments claiming to demonstrate excess heat production in light water-Ni-K2CO3 electrolytic cells was found to produce an apparent excess heat of 11 W maximum, for 60 W electrical power into the cell. Power gains ranged from 1.06 to 1.68. The cell was operated at four different dc current levels plus one pulsed current run at 1 Hz, 10% duty cycle. The 28 liter cell used in these verification te
