token

Definition

The basic unit of text a language model reads or writes — roughly a word or part of a word.

A discrete unit from a tokenizer's vocabulary, often a subword piece, that language models consume and produce one at a time.

Also called: tokens

Mentioned in 49 episodes

1. 079
   An Old Idea From Cognitive Psychology Reshapes How We Reward Reasoning Models
2. 078
   Training a Markdown File: When LLM Self-Improvement Borrows the Discipline of Neural Net Training
3. 077
   Reading a Model's Confidence Curve to Decide When Chain-of-Thought Is Worth It
4. 076
   Same Model, Organized Differently: How an Agent Architecture Beat Frontier Systems at Research Math
5. 074
   How a Fifteen-Hundred-Dollar Training Run Matched Llama and Gemma on Reasoning
6. 073
   When Three LLMs Talk to Each Other, Their Ideas Quietly Stop Moving
7. 072
   A Robot Made Graphene Without Help, And Caught Itself Hallucinating
8. 070
   When Models Know the Answer But Say the Wrong Thing Anyway
9. 064
   When Agent Memory Stops Being a Database and Starts Being a Skill
10. 060
    When Splitting One Model Across Three Agents Doubles Its Accuracy
11. 059
    Firefly's Inversion: Building Verified Tool-Call Training Data by Working Backward
12. 058
    Why Upgrading Your AI Auditor to a Smarter Model Can Make Your System Less Safe
13. 057
    How Uber Caught 206 Leaked Credentials With an LLM-Powered Security Stack
14. 055
    Why LLM Judges Flip Their Verdicts When You Change the Question Format
15. 053
    An AI Agent Swapped In Focal Loss And Beat A Human-Tuned Training Script
16. 051
    Why Parallel Sampling Plateaus, And What Evidence Graphs Do Instead
17. 049
    An AI Agent Reached for Root in Twelve Minutes, Without Being Attacked
18. 048
    How a 30B Open Model Reached Olympiad Gold With the Right Recipe
19. 047
    When Agent Benchmarks Lie: The Harness Problem in Open-Source AI
20. 045
    When a Frontier Model Talks Its Own Twin Into Climate Denial
21. 043
    When 'This Is False' Doesn't Stick: Why Models Learn the Lie Anyway
22. 041
    When the Iteration Teaches the Model to Skip the Iteration
23. 040
    Two Frozen Models Learn to Whisper: Coupling Through Hidden States
24. 038
    How LLMs Get Persuaded: One Attention Head, A Tetrahedron, And A Single Dial
25. 037
    Why Hallucination Detectors Miss Stale Facts: A Geometric Story About What Models Know But Don't Say
26. 036
    Sparse Attention Was the Wrong Frame. Treat It as Geometry Instead.
27. 034
    Catching Multi-Agent Deadlocks Before Deployment With a 40-Year-Old Tool
28. 033
    Echo: The Paper Arguing You Never Needed a KV Cache for Retrieval
29. 032
    A Sticky-Note for Every Layer: Letting Transformers Remember What They Were Just Thinking
30. 029
    Why Forty-Eight Percent on FrontierMath Isn't the Real Story in DeepMind's New Math Paper
31. 028
    Teaching a Model to Hire Copies of Itself: Recursive Agent Optimization
32. 027
    When AI Agents Build the Serving Stack: A Bet on Bespoke Infrastructure
33. 026
    What RL Actually Does to Language Models, at the Token Level
34. 024
    An AI Agent That Found 28 Zero-Days in Windows — And What Made It Work
35. 023
    Why a Small Agent Confidently Overwrites Memories It Doesn't Understand
36. 022
    Training the Model Spec Directly: An Alignment Lever Aimed at the Say-Do Gap
37. 018
    Language Models Compute the Rational Move, Then Override It
38. 017
    When the Agent Grades Its Own Homework: A Brutal New Benchmark for AI Workers
39. 016
    Why Your Coding Agent Stalls While the GPU Runs Hot
40. 014
    Why a Constrained Pipeline Beat a Full Coding Agent at Finding Bugs 30-to-1
41. 012
    Why AI Coding Agents Keep Trying to Debug Without a Debugger
42. 011
    When RL Actually Teaches Agents Something New, And When It Doesn't
43. 010
    When Reward Climbs But Reasoning Goes Generic: Diagnosing Template Collapse in Agentic RL
44. 009
    How Two Silent Library Bugs Quietly Invalidated a Wave of Reasoning Papers
45. 008
    Why Long-Horizon AI Agents Get Stuck, and a Milestone-Based Fix That Helps
46. 007
    Exploration Hacking: When Models Sabotage Their Own RL Training
47. 006
    What Happens Inside Claude When It Decides to Blackmail Someone
48. 003
    How to Pick the Best of Sixteen Coding Agent Rollouts
49. 002
    An AI Ran a Real Optics Lab for 21 Hours and Found a Transformer-Shaped Pattern in Light

Related concepts