I find Phi-3.5-mini and Qwen better for those tested one on low-end hardware. Llama I don't use, as it is non-free.
Jean Louis
JLouisBiz
AI & ML interests
- LLM for sales, marketing, promotion
- LLM for Website Revision System
- increasing quality of communication with customers
- helping clients access information faster
- saving people from financial troubles
Recent Activity
new activity
about 12 hours ago
microsoft/Phi-3.5-mini-instruct:tool calls
new activity
about 21 hours ago
llmware/bling-phi-3.5:12.5 tokens/second on 16 GB RAM with 4 GB VRAM GPU Nvidia GTX 1050 Ti
new activity
about 21 hours ago
llmware/bling-phi-3.5:Please confirm I am getting right model here
Organizations
JLouisBiz's activity
tool calls
2
#7 opened 4 months ago
by
stevensun1
12.5 tokens/second on 16 GB RAM with 4 GB VRAM GPU Nvidia GTX 1050 Ti
#5 opened about 21 hours ago
by
JLouisBiz
Please confirm I am getting right model here
1
#3 opened 2 days ago
by
JLouisBiz
Converting to GGUF failed, do you have a method?
1
#4 opened 2 days ago
by
JLouisBiz
replied to
AkimfromParis's
post
2 days ago
Hmm, you are so personal.
I am not supporting any kind of ideas, my unrelated comments may not be true, as you say, only accuracy.
I am supporting notion of providing accurate information in public statements.
Nikola Tesla wasn't born in Croatia, neither in Yugoslavia. Though the village Smiljan is now in Croatia. He was born in Austrian Empire. We cannot be using Croatian flag, or Yugoslavian flag to say he was born in some country where he wasn't born. Country is not perpetual and changes it's borders.
That is analogous example: https://en.wikipedia.org/wiki/Nikola_Tesla
Wikipedia keeps the same notion, country is different than territory. Conflating territory or attributing historical figures to modern countries doesn't make sense. It is not related.
You feel free to designate countries as you wish, but I don't consider it correct.
Actually, I am not getting you, why do you get upset each time?
LLAMA 2 Community License is NOT "Open Source", please remove deceptive statements
#13 opened 3 days ago
by
JLouisBiz
Download, and share by email or upload or copy paste into other window.
reacted to
AkimfromParis's
post with 👍
3 days ago
Post
1596
🇺🇸 🇨🇦 🇬🇧 Nobel Prize winners against USSR & Japanese AI pioneers ☭🇯🇵
🇩🇪 Prof. Jürgen Schmidhuber: “The #NobelPrize in Physics 2024 for Hopfield & Hinton turns out to be a Nobel Prize for plagiarism. They republished methodologies developed in #Ukraine and #Japan by Ivakhnenko and Amari in the 1960s & 1970s, as well as other techniques, without citing the original inventors.”
1965 - First Deep Learning - USSR ☭ (Ukraine 🇺🇦 now)
Ivakhnenko and Lapa introduced the first deep learning in deep MLPs that learn internal representations of input data.
1967/68 - Deep Learning by Stochastic Gradient Descent - Japan 🇯🇵
Shun-Ichi Amari trained MLPs with many layers in non-incremental end-to-end fashion from scratch by stochastic gradient descent (SGD).
1969 - Rectified linear unit - Japan 🇯🇵
In 1969, Kunihiko Fukushima introduced ReLU in the context of visual feature extraction in hierarchical neural networks.
1970 - Backpropagation - Finland 🇫🇮 😃
In 1970, Seppo Linnainmaa was the first the reverse mode of automatic differentiation, now known as backpropagation.
1972 - Recurrent Neural Network - Japan 🇯🇵
In 1972, Shun-Ichi Amari published a learning recurrent neural network based on Lenz-Ising model (Amari's net was later called the "Hopfield network". Hopfield republished in 1982, without citing Amari papers.)
1979 - First Convolutional neural network - Japan 🇯🇵
CNN architecture was introduced in 1979 by Kunihiko Fukushima, also known as Neocognitron.
https://people.idsia.ch/~juergen/deep-learning-history.html#AMH2
🇩🇪 Prof. Jürgen Schmidhuber: “The #NobelPrize in Physics 2024 for Hopfield & Hinton turns out to be a Nobel Prize for plagiarism. They republished methodologies developed in #Ukraine and #Japan by Ivakhnenko and Amari in the 1960s & 1970s, as well as other techniques, without citing the original inventors.”
1965 - First Deep Learning - USSR ☭ (Ukraine 🇺🇦 now)
Ivakhnenko and Lapa introduced the first deep learning in deep MLPs that learn internal representations of input data.
1967/68 - Deep Learning by Stochastic Gradient Descent - Japan 🇯🇵
Shun-Ichi Amari trained MLPs with many layers in non-incremental end-to-end fashion from scratch by stochastic gradient descent (SGD).
1969 - Rectified linear unit - Japan 🇯🇵
In 1969, Kunihiko Fukushima introduced ReLU in the context of visual feature extraction in hierarchical neural networks.
1970 - Backpropagation - Finland 🇫🇮 😃
In 1970, Seppo Linnainmaa was the first the reverse mode of automatic differentiation, now known as backpropagation.
1972 - Recurrent Neural Network - Japan 🇯🇵
In 1972, Shun-Ichi Amari published a learning recurrent neural network based on Lenz-Ising model (Amari's net was later called the "Hopfield network". Hopfield republished in 1982, without citing Amari papers.)
1979 - First Convolutional neural network - Japan 🇯🇵
CNN architecture was introduced in 1979 by Kunihiko Fukushima, also known as Neocognitron.
https://people.idsia.ch/~juergen/deep-learning-history.html#AMH2
-
- 11 replies
replied to
AkimfromParis's
post
3 days ago
I agree with you 100% that anthropomorphism of these tools is detrimental for our society.
I am sorry, the pun was not on the LLM, I was llm-izing the human. Not the other way around.
Kievan Rus is origin of all of the Russia. Research the subject. I do not know how is it relevant, as mixing history was the reason of my initial comment.
I appreciate your work, carry on, you may close this discussion.
That is not free software, as you forbid commercial use
19
#11 opened 11 days ago
by
JLouisBiz
Where do I get original uncensored model?
1
#1 opened 4 days ago
by
JLouisBiz
Why a proprietary license?
1
#1 opened 3 months ago
by
Regrin
Hallucinated reply if you prompt same prompt several times. Ollama and phi3.5
1
#25 opened about 2 months ago
by
sepihala
replied to
AkimfromParis's
post
5 days ago
Okay Ukraine existed 1965, okay, I got you.
I have Deja Vu, it's like talking to LLM with 0.5 B
replied to
AkimfromParis's
post
5 days ago
Sure, feel free to make jokes out of my comment.
I don't mind if you wish to strive for facts or play games with your post. Enjoy.
replied to
AkimfromParis's
post
5 days ago
Of course, I am not arguing on origins of person, though saying it was Ukraine in 1965, it wasn't. It was Soviet Union.
We talk of countries, not nationality. Country cannot become country back into the time.
Sorry, I do not know why you perceive this as impolite.
Let me give you example:
https://en.wikipedia.org/wiki/George_Washington
Was he born in United States or in British America?
How could he be born in United States if he was the founding father of United States?
Country is not territory, let's not conflate it.
Carry on with disputing plagiarism.
Rather work on Qwen, as it is free software, LLama is not.
奎恩是最好的！（Qwen shì zuì hǎo de!）