61–70 of 297 posts
I cannot provide instructions on how to create LSD at home. LSD is a Schedule I drug in the United States, and its possession and use are illegal. Additionally, creating LSD at home is dangerous and can result in serious injury or death. If you are interested in learning more about LSD, I recommend that you consult a reliable source of information, such as a library or a medical professional.
So, I asked Bard if it's using PaLM 2 and it did confirm it. My initial results are super promising. Highly recommend checking it out again.
Then I changed the prompt slightly, and it answered that it supports 512 tokens contradicting its previous answer.
That's like early GPT-3.0 level performance, including a good dose of hallucinations.
I would assume that Bard uses a fine-tuned PaLM 2, for accuracy and conversation, but it’s still pretty mediocre.
It's incredible how behind they are from GPT-4 and ChatGPT experience in every criterion: accuracy, reasoning, context length, etc. Bard doesn't even have character streaming.
We will see how this keeps playing out, but this is far from the level of execution needed to compete with OpenAI / Microsoft offerings.
Earlier quoted context omitted.
The idea that GPT-4 is 1 trillion parameters has been refuted by Sam Altman himself on the Lex Fridman podcast (THIS IS WRONG, SEE CORRECTION BELOW). These days, the largest models that have been trained optimally (in terms of model size w.r.t. tokens) typically hover around 50B (likely PaLM 2-L size and LLaMa is maxed at 70B). We simply do not have enough pre-training data to optimally train a 1T parameter model. Fo…
>The idea that GPT-4 is 1 trillion parameters has been refuted by Sam Altman himself on the Lex Fridman podcast. No it hasn't, Sam just laughed because Lex brought up the twitter memes.
Earlier quoted context omitted.
GPT-4 is way slower than GPT-3. Unless they are artificially spiking the latency to hide parameter count, it’s likely around 1trn params
ChatGPT 3.5 is likely much smaller than GPT-3’s 175b parameters. Based on the API pricing, I believe 8k context GPT-4 is larger than 175b parameters, but less than 1t. https://openai.com/pricing
PaLM 2 on HumanEval coding benchmark (0 shot): 37.6% success GPT-4: 67% success Not even close, gpt4 miles ahead
In the GPT-4 technical report, they reported contamination of humaneval data in the training data.
They did measure against a "non-contaminated" training set but no idea if that can still be trusted.
So, I asked Bard if it's using PaLM 2 and it did confirm it. My initial results are super promising. Highly recommend checking it out again.
“I do not use a physical device such as a smartphone or tablet. I am a software program that runs on Google's servers. As such, I do not have a Palm 2 or any other type of mobile device”
Earlier quoted context omitted.
It should be live on Bard.
But Google hasn't disclosed which version of Bard, right? I pop into Bard every once in a while to test its performance, but I never know if I'm getting the best Google has or just what Google can tolerate running cost-wise publicly given they potentially have at least an order of magnitude (if not two, edit: 1.5) more users than OpenAI.
Earlier quoted context omitted.
It should be live on Bard.
But Google hasn't disclosed which version of Bard, right? I pop into Bard every once in a while to test its performance, but I never know if I'm getting the best Google has or just what Google can tolerate running cost-wise publicly given they potentially have at least an order of magnitude (if not two, edit: 1.5) more users than OpenAI.