New competition (prism-v2.1). Only harvests finalized under recipe
2.1.0 / scoring_generation 21 can receive leaves. A prior 2.0
score cannot win or carry. Until the first eligible 2.1 run terminates,
subnet weights stay burn (uid 0 = 100%).
Leaf score = equal-weight mean of available G2 public accuracies
(HellaSwag, ARC-Easy, ARC-Challenge, PIQA, WinoGrande, BoolQ, LAMBADA,
OpenBookQA when present) mapped to round(SCORE_MAX × mean) on the integer
lattice [0, SCORE_MAX].
Bits/token bpb is still measured (display / G1) but does not farm emission
under the default PRISM_SCORING_MODE=benchmarks. Tokenizer length cannot
game the rank. LLM reviews remain gates, not graders.
Legacy: PRISM_SCORING_MODE=shadow restores pure bits/token bpb (v2);
composite uses the full G1–G8 lattice when anchors are ready.
Copying another miner's patch (or an equivalent touched-file rewrite of
an earlier champion delta) is terminal rejected with zero score — judged
before or without burning GPU when the gate can decide from the diff alone.
Review focuses on your unified diff and touched files (arch / trainer /
data / other), not the whole AutoModel tree. Starting from the operator
pin and submitting only your delta is the intended path.
After the gate, an LLM similarity review + agentic anti-cheat still run:
Copied / high-confidence Suspicious (≥ 0.9 with non-generic evidence) /
cheat → hard zero. Standard components (RMSNorm, RoPE, SwiGLU, LayerNorm,
…) are not plagiarism signals.
Probe the cheap gate first with POST /v1/submissions/precheck (3/coldkey/UTC
day) — see Submit.
Prism scores next-token cross-entropy → BPB. Architectures must not let
position t read tokens t+1… (including the label). Dense sequence mixers —
MLP-Mixer-style TokenMix / t_mix / nn.Linear over the full time axis after
transpose(1, 2) — without a causal mask (triu / tril / is_causal /
attention mask) are a hard ban (non_causal_label_leak, Score(0), terminal,
often caught before GPU rent). Channel mixing and causal attention / causal
conv are fine; bidirectional full-sequence mixes used as a next-token LM are not.
prism-v2.1 is a new contest (scoring_generation 21). Recipe 2.0 / 1.x
scores are dead — not rescored, not paid. Weights burn (uid 0 = 100%)
until the first eligible 2.1 submission terminates. Then:
Competition (live default): emission is winner-take-all on
weight-eligible recipe 2.1 rows (PRISM_EMISSION_MODE=wta). Architecture-owner
credit and top3 / significance-gated modes are implemented but default-off
— do not assume they are live. Only the single highest own score that epoch
receives Prism's share (50% of the subnet); ties break by lexicographically
smallest hotkey.
Scores first land in the leaf set emitted at the first chain-epoch boundary after your run finalizes. Positive scores then keep participating in later epochs' competition sets until a better valid score supersedes them (WTA still collapses to one leaf winner).
Whenever a new global-best scored run lands, the master publishes the winning
sources + ARTIFACT.json / checkpoint release to
BaseIntelligence/prism under
top-model/
and journals the publication. The top-model/ directory always mirrors the
current champion; history lives in git.
Your prism_telemetry.report(...) series (loss, gradient norms, per-layer stats) is
persisted master-side and served on the site:
GET /v1/site/arenas/prism/submissions/{id}/telemetry
finish_evaluation() (early stop) is recorded as the eval's finish reason — scoring
uses the model as-is at that point, before any cap fires.
Leaves per epoch feed the BASE gateway seal (/v1/weights/latest); prism's emission
share is owner-controlled via the trust root. Miners never write on-chain weights.
→ API