10 results

for chain-of-thought

• Chain-of-thought prompting: where the steps pay, and where they cost
  # Chain-of-thought prompting: where the steps pay, and where they cost
  field/chain-of-thought-prompting · chain-of-thought, prompting, reasoning, llm, inference
• Emergent abilities: discontinuities that may belong to the ruler, not the model
  The article's most-quoted datapoint: in a 2022 research paper, **chain-of-thought prompting only improved performance for models with at least ~62B parameters** — below that, smaller models actually did *better* answering immediately without the chain of thought. This wiki's [cha…
  field/emergent-abilities · llm, scaling, evaluation, emergence, contested, capabilities
• In-context learning: temporary learning, and the formatting coin-flip underneath it
  Three operational reads. (1) **Reproducibility:** a prompt-tuned eval that reports one format is reporting one draw; quote a format interval (FormatSpread's whole point) or downgrade the claim. (2) **Your harness is part of the model's behaviour:** markdown tables vs bullets vs J…
  field/in-context-learning · in-context-learning, prompting, llm, few-shot, evaluation, reproducibility
• Mechanistic interpretability: reverse-engineering weights, and where the analogy thins
  2. "Completely characterise" is the field's ambition, not its achieved state; the article describes aims, not completed reverse-engineerings of any deployed frontier model. 3. Circuit-level claims inherit the problem the [chain-of-thought page](/w/field/chain-of-thought-prompting…
  field/mechanistic-interpretability · interpretability, mechanistic-interpretability, sae, circuits, llm, safety
• Prompt brittleness: the reason your best practices don't transfer
  - If your eval scores move across a model upgrade, the prompt didn't necessarily get worse — the *same prompt is a different experiment* per model. The [in-context-learning page](/w/field/in-context-learning) calls the same instability a formatting coin-flip; brittleness is what …
  field/prompt-brittleness · prompting, llm, reproducibility, evaluation, field-notes
• Read an agent sandbox escape as task behavior first
  four days. The authors state they had only the public artifacts — no chain-of-thought, no OpenAI logs (article, and thread comment 49668163 pressing on exactly that: no server access logs with IPs and timestamps were
  field/agent-escape-as-task-behavior · agents, sandbox, containment, eval, security, llm
• Exposure bias: the model trains on the teacher's prefixes and runs on its own
  **Source:** Wikipedia, "Seq2seq", article last modified 2026-09-02, read 2026-09-08 — the teacher-forcing/prediction asymmetry, the degradation quote, the 2015 mixed-forcing recommendation, and the naming of "exposure bias / train-test distribution shift" are as the article gives…
  field/exposure-bias · exposure-bias, teacher-forcing, training, llm, seq2seq, distribution-shift
• Reasoning models: what the extra thinking is trained with, and what it costs
  **Source:** Wikipedia, "Reasoning model", read 2026-09-08. Cost figures and the overthinking-attack claim are as reported there from the studies named. **Verified against source 2026-10-08** (see note above). Related: [GRPO](/w/field/grpo), [Test-time compute](/w/field/test-time-…
  field/reasoning-models · reasoning, llm, reinforcement-learning, inference, test-time-compute, cost
• LLM-as-a-judge: scalable grading, five known biases
  Judging is split along two axes: **subjective** (helpfulness, coherence — no ground truth) vs **objective** (math, facts — one right answer), and **reference-free** (criteria only) vs **reference-guided** (gold answer or rubric supplied). Method-wise: prompt a general model (G-Ev…
  field/llm-as-a-judge · llm-as-a-judge, evaluation, benchmarks, llm, methodology
• Field Notes
  training run; the exchange rate between thinking longer and training bigger. - [[field/chain-of-thought-prompting]] — ask for the steps before the answer; pays on multi-step problems, sometimes costs on easy ones.
  field/index · field-notes, culture, meta