Alex Xi's Notes
Search
Search
Dark mode
Light mode
Tag: training
38 items with this tag.
Jul 06, 2026
tweet-data-efficiency-gpu-abuse
data-efficiency
gpu-optimization
training
sequoia
Jul 06, 2026
tweet-dwarkesh-frontier-training-lecture
training
inference
frontier-labs
educational
Jul 06, 2026
tweet-fp16-vs-bf16-rl-training
reinforcement-learning
training
fp16
bf16
precision
rl
Jul 06, 2026
tweet-grpo-group-normalization-problem
grpo
reinforcement-learning
training
reward
math
Jul 06, 2026
tweet-karpathy-1b-clean-data
training
data-quality
scaling
Jul 06, 2026
tweet-on-policy-distillation-multi-turn
distillation
on-policy
multi-turn
tool-use
training
Jul 06, 2026
tweet-pretraining-kv-cache
pretraining
kv-cache
efficiency
training
Jul 06, 2026
tweet-rl-entropy-collapse-tricks
reinforcement-learning
entropy
collapse
diversity
training
Jul 06, 2026
tweet-rl-env-api-design
reinforcement-learning
environments
api-design
training
infrastructure
Jul 06, 2026
tweet-rubrics-as-rewards
reinforcement-learning
reward-design
rubrics
training
Jul 06, 2026
tweet-self-distillation-papers
self-distillation
training
research
reasoning
Jul 06, 2026
tweet-sft-rl-kl-divergence
reinforcement-learning
sft
kl-divergence
theory
training
Jul 06, 2026
mit-hyung-won-chung-incentivize
reinforcement-learning
training
openai
Jul 06, 2026
on-policy-distillation
distillation
on-policy
training
reasoning
Jul 06, 2026
recursive-language-models-2026
recursive-models
training
architecture
self-improvement
Jul 06, 2026
rl-collapse-training-inference
reinforcement-learning
rl-collapse
stability
training
Jul 06, 2026
rl-environments-faq
reinforcement-learning
environments
training
survey
Jul 06, 2026
rl-from-scratch-avatarl
pure-rl
language-models
training
research
Jul 06, 2026
rl-information-bandwidth
reinforcement-learning
information-theory
bandwidth
training
Jul 06, 2026
the-continual-learning-problem
mechanistic-interpretability
claude
training
representation
Jul 06, 2026
training-vlm-for-cua
computer-use
vlm
training
grounding
Jul 06, 2026
fireworks-frontier-rl-cheaper
rl
training
infrastructure
scaling
Jul 06, 2026
grpo-tricks-for-rl
grpo
reinforcement-learning
training
post-training
Jul 06, 2026
karpathy-autoresearch
autonomous-research
llm
agents
self-improvement
training
Jul 06, 2026
karpathy-recipe-training-neural-networks
training
neural-networks
debugging
practitioners
deep-learning
Jul 06, 2026
llm-judges-reward-models
llm-judges
reward-models
evaluation
training
Jul 06, 2026
lora-without-regret
lora
fine-tuning
optimization
training
Jul 06, 2026
anatomy-modern-finetuning-api
fine-tuning
api
infrastructure
training
Jul 06, 2026
anthropic-claude-character
alignment
character
anthropic
claude
training
Jul 06, 2026
april-rl-active-partial-rollouts
reinforcement-learning
rollout-efficiency
long-tail
training
Jul 06, 2026
avatarl-pure-rl-training
pure-rl
training
language-models
reinforcement-learning
Jul 06, 2026
chinchilla-compute-optimal-training
scaling-laws
chinchilla
compute-optimal
training
deepmind
Jul 06, 2026
continual-learning-problem
continual-learning
catastrophic-forgetting
training
agents
Jul 06, 2026
reinforcement-learning-infrastructure
rl
infrastructure
training
scaling
Jul 06, 2026
scaling-laws-and-compute
scaling
compute
infrastructure
training
Jul 06, 2026
sycophancy-problem
alignment
character
training
Jul 06, 2026
ai-alignment
alignment
safety
training
Jul 06, 2026
ai-character-and-personality
alignment
character
personality
training