Lab 01 · Teaching simulation
Token-by-token generation
Inspect candidate tokens, change the temperature, and compare two sequences generated with the same pseudo-random seed.
Local lab
Token-by-token generation
Inspect a prepared distribution, then sample a reproducible sequence. Temperature is only one part of a real generation pipeline.
Controls
Observe how several plausible continuations receive different probabilities.
1,0
At 1: the prepared distribution is preserved. A real system may apply other filters and settings.Same seed + same settings = same sequence in this simulation.
Current distribution
Current contextThe cat drinks
- milk68 %
- broth20 %
- juice12 %
Normalized total: 100 %. Individual percentages may differ slightly because of rounding.
Ready. Choose a scenario, then generate one token or start autoplay.
Compare two generations
Same scenario and seed: A uses a low temperature of 0.4; B uses your current setting (1).
Generation A · T 0.4
Not generated yet.
Generation B · T 1
Not generated yet.
What this simulation does not show
- Candidates and probabilities are written in advance, not calculated.
- There is no neural network, attention mechanism, or real tokenizer.
- A real system may apply top-k, top-p, penalties, and additional stop rules.
- A token probability is not the probability that a claim is true.
- The seed is reproducible only inside this lab.