bratao

11 ms·

121.
by bratao 3y ago
2.7B size with better performance than Mistral 7B is impressive!!
122.
by bratao 3y ago
Just one observation. This model seems "less" multilingual than LLAMA and Mistral. I always ask in Portuguese "What is the capital of Bahia?"/ "Qual a capital da Bahia?" and DeciLM answered wrong and in En
123.
Is Austin losing its luster?(techcrunch.com)
48 points by bratao 3y ago|73 comments
124.
by bratao 3y ago
And this uses Hyena, that can be considered a "previous generation" of Mamba. I think that this anwsers the question about the scalability of SSM and the transformer finally found an opponent.
125.
by bratao 3y ago
I don´t think so. There are few algorithms that cover the majority of use. For example in NLP a fast LLM (such as LLAMA) training/prediction and support for Hugging face transformers (BERT and such) would cover 95%+ of NLP use-cases th
126.
Unsloth: 80% faster, 50% less memory, 0% accuracy loss Llama finetuning(github.com)
132 points by bratao 3y ago|3 comments
127.
Together AI raises a $102.5M Series A(together.ai)
70 points by bratao 3y ago|24 comments
128.
Keras 3.0(keras.io)
224 points by bratao 3y ago|75 comments
129.
by bratao 3y ago
Do anyone know what Q* is?
130.
by bratao 3y ago
I fully agree that the work load should reach an optimal work/life balance. But how this is possible if someone is taking two extra full time jobs on top of it?
131.
by bratao 3y ago
I'm curious as to how this would work in practice. For instance, if an individual is able to complete their sprint while juggling two other full-time jobs, does this imply they should have taken on more tasks? This approach seems to on
132.
by bratao 3y ago
For software development, quantifying output or defining a finite set of tasks is almost impossible. Unlike traditional industries where tasks can be standardized and expected output can be clearly defined, software development is a differe
133.
The secretive rise of the 'overemployed'(businessinsider.com)
63 points by bratao 3y ago|100 comments
134.
Another Major Outage Across ChatGPT and API(status.openai.com)
73 points by bratao 3y ago|5 comments
135.
by bratao 3y ago
OpenAI ( https://fortune.com/2023/05/05/openai-ceo-sam-altman-remote-... ) is a example of a high performing company where remote work is an exception. That's why I say that there is no cake recipe that gu
136.
by bratao 3y ago
I feel like everyone is pointing out and screaming to AMD what they should do to get money, and they are still struggling to find it out.
137.
by bratao 3y ago
The GIL only protect the Interpreter. The only thing it may do is to make it infrequently. There are MANY threading bugs in actual Python code.
138.
by bratao 3y ago
In my opinion, Python needs to have an urgent answer to the use of multiple cores. AMD just launched a CPU with 96 cores. Today the use on multiple cores is through the use multiprocess, which has many limitations. I understand that the mul
139.
FBI says North Korea deployed thousands of IT workers to get remote jobs in US(businessinsider.com)
22 points by bratao 3y ago|7 comments
140.
Product Hunt cleans house with layoffs impacting 60% of staff(techcrunch.com)
40 points by bratao 3y ago|51 comments
141.
by bratao 3y ago
In my opinion They won't outperform the best LLMs for one problem: Quality of data. A hidden secret that everyone knows is that good LLMs use copyrighted data like books-3. OpenAI itself cites a dataset from books and does not go into
142.
by bratao 3y ago
Sorry about that. I´m not a native speaker and asked GPT-4 to: "Create a engaging reply for HackerNews talking that this is a great model, and I really hope that they release a 13B and 34B version. As those sizes are way more capable a
143.
by bratao 3y ago
Yeah +1 for VERY competitive. The vector capabilities of Vespa are incredible and the Text/ranking features are amazing. I don´t think any other product have those two sides so developed as them.
144.
Vespa.ai is spinning out of Yahoo as a separate company(blog.vespa.ai)
351 points by bratao 3y ago|134 comments
145.
by bratao 3y ago
+1 for Vespa. For me it is VERY resilient and production ready. It is such a dream compared to Elasticsearch, that we migrated from.
146.
by bratao 3y ago
HostGator
147.
AMD Hip SDK: Making CUDA Applications Run Across Consumer, Pro GPUs and APUs(phoronix.com)
22 points by bratao 3y ago|5 comments
148.
by bratao 3y ago
It´s missing the most feature-complete for vectors database, the vespa.ai. I highly recommend.
149.
HuggingFace Text Generation License No Longer Open-Source(github.com)
84 points by bratao 3y ago|55 comments
150.
by bratao 3y ago
Here someone from the team explain the change: https://twitter.com/jeffboudier/status/1684988169669005312 "HFOIL is not a true open source license: to sell a hosted or managed service built with TGI, we requi

More ›