Token-by-token generation — deepseek-r1:8b
Writing the Script
The 261 tokens generated after the reasoning span ended, while the model produced the batch file text itself. Every branch is a token that received nonzero probability at that step; copper is the one the sampler drew.
token span Ollama's API labels message.reasoning
sampled — was also the top-probability token
sampled — fell outside the shown top-k despite lower probability
nonzero probability, not sampled
Built with vis-network from a real trace via Ollama's logprobs API: at every step the model's top candidate tokens and their real probabilities were captured, and the next token drawn by the server's own sampling -- not scripted.