Loaded baked cauliflower casserole pairs roasted cauliflower with crispy bacon, cheddar, and a rich, creamy sauce. It’s an easy low-carb side that stays creamy, cheesy, and never watery. #cauliflower# #lowcarb# #casserole# #sidedishes# via @bakeitwithlove

America, pick a side! 🇺🇸🍟🧅 🍟 French Fries are the KING of side dishes. vs. 🧅 Onion Rings are the KING of side dishes. You can only keep ONE forever. Which one are you choosing? 😂

In a typical 1950s diner, a standard burger patty weighed just 1.6 ounces and a normal soda filled a 7-ounce glass. By modern standards, a 1950s adult was eating a kid's meal. And it shows in our modern culture. Over the past seven decades, American portion sizes have undergone severe inflation, drastically changing what people consider a "normal" serving. Take the hamburger, the quintessential mid-century staple. The entire sandwich—bun, meat, and condiments—was small enough to be eaten with one hand. Today, a standard fast-food burger is typically 4 ounces (a quarter pound) before cooking, while casual dining restaurants routinely serve half-pound (8-ounce) or even 12-ounce burgers. The standard 1950s burger would look like a modern slider. Side dishes followed the same trajectory. A typical order of French fries in the 1950s was roughly 2.4 ounces, which is slightly smaller than a modern fast-food "small" fry. Today, large fry portions often push 5.5 to 6 ounces. Beverages experienced the most dramatic expansion. When 12-ounce cups were introduced to replace those 7-ounce fountain glasses, they were explicitly marketed as "large." Now, a standard medium fountain drink sits around 21 ounces, and large sizes regularly exceed 30 or even 40 ounces—volumes that would have been physically impossible to serve in the standard glassware of the era. Even sit-down restaurant meals were substantially smaller. A typical steakhouse cut in the 1950s was 6 to 8 ounces, served alongside a modest vegetable side. Today, 12, 16, and 24-ounce cuts are the norm. This shift was driven largely by marketing and the belief more food makes you feel wealthy. Addtionally, restaurants realized they could increase profits by slightly raising prices and dramatically increasing portions, offering perceived "value" to customers. The plates themselves had to be redesigned to accommodate the extra food, growing from a standard 9-inch diameter in the 1950s to the 11- or 12-inch plates common today.

📺 Xiaomi Is Livestreaming a Production RL Run — Burn Rate Included @XiaomiMiMo is training its MiMo-V2.6 in public — not a demo, a live post-training dashboard where every step's reward, loss and benchmark movement is watchable in real time. One day in, the meter reads roughly $1.2M spent and 60B tokens consumed. As Zhihu contributor Kitt在进化 puts it: this isn't a training site, it's a money-burning site — about ¥4,000 a minute. His bigger point: this is rare, valuable data. Almost nobody publishes what frontier-scale RL actually costs. 1️⃣ What the dashboard actually shows Every training step exposes the internals most labs keep private. His reading guide: 🔹 Loss family: training loss, entropy (are answers diversifying or collapsing into one mode), gradient norm, and train/inference KL divergence. 🔹 Reward: a mix of test-case scoring and rubric-based reward. 🔹 dynsam (dynamic sampling): avg@n performance over repeated tries, plus pass-rate buckets — including samples never solved and always solved. 🔹 DeepSWE v1.1 as the headline probe (mini-swe-agent, avg@3): Pro at 63.7, Flash at 60.7 at the time of writing. 2️⃣ The cost ledger, out in the open From the dashboard he extracts the numbers the industry usually guesses at: 🔹 MiMo-V2.6 Pro: roughly $36 per million tokens of training. 🔹 MiMo-V2.6 Flash: roughly $8 per million tokens. 🔹 His rule of thumb: equivalent inference runs 30-50x cheaper than these training figures. 3️⃣ What's actually being trained The sample distribution is also public: about 1,500 prompts per step, ~70% of them coding tasks. The remaining third is split across general, visual, cybersecurity and chat tasks — which makes this, in effect, an agentic-coding-centric RL run with side dishes. 4️⃣ Why it's worth watching He notes a university lab livestreamed a training run days earlier, but at nowhere near this scale or frontier relevance. For anyone who wants to learn how production RL behaves — rewards, entropy, dynamic sampling, benchmarks moving step by step — this is a rare open classroom. His only complaint, half-joking: a livestream this good deserves a comment section. 🔗 Full Reading: #Xiaomi# #MiMo# #ReinforcementLearning# #PostTraining# #LLM# #OpenScience# #AI#