higgsfield

Hey there! Higgsfield AI. We have a massive GPU cluster and developed our own infrastructure to manage the cluster and train massive models. There's how it works: - You upload the dataset with preconfigured format into HuggingFaсe [1]. Choose your LLM (e.g. LLaMa 70B, Mistral 7B) - Place your submission into the queue - Wait for it to get trained. - Then you get your trained model there on HuggingFace. Essentially, why would we want to do it? We already have an experience with training big LLMs. We could achieve near-perfect infrastructure performance for training. Sometimes GPUs have just nothing to train. Thus we thought it would be cool if we could utilize our GPU cluster 100%. And give back to Open Source community (already built an e2e distributed training framework [2]). This is in an early stage, so you can expect some bugs. Any th

We have a massive GPU cluster and developed our own infrastructure to manage the cluster and train massive models. There's how it works: 1. You upload the dataset with preconfigured format into HuggingFaсe [1]. 2. Choose your LLM (e.g. LLaMa 70B, Mistral 7B) 3. Place your submission into the queue 4. Wait for it to get trained. 5. Then you get your trained model there on HuggingFace. Essentially, why would we want to do it? 1. We already have an experience with training big LLMs. 2. We could achieve near-perfect infrastructure performance for training. 3. Sometimes GPUs have just nothing to train. Thus we thought it would be cool if we could utilize our GPU cluster 100%. And give back to Open Source community (already built an e2e distributed training framework [2]). This is in an early stage, so you can expect some bugs. Any thoughts, op

I'm currently studying this topic (preparing an online course about it actually) and here are the best resources I found so far: Full Courses - From David Silver: http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching.html - From Yandex: https://yandexdataschool.com/edu-process/rl - From Sergey Levine: http://rail.eecs.berkeley.edu/deeprlcoursesp17/index.html Articles: - Q-learning on Taxi-v2, very good basic explanation: https://www.learndatasci.com/tutorials/reinforcement-q-learn... - Q-learning and DQN, goes a bit further: https://neuro.cs.ut.ee/demystifying-deep-reinforcement-learn... - "Pong from Pixels from Karpathy", introduces DQN and PG: https://karpathy.github.io/2016/05/31/rl/ Baseline implementations: - "RL-Adventures", super clean Pytorch implementations: https://github.com/higgsfield/RL-Adventure (DQN) and https://github.com/h