Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦
C
probe: env-gated top-2 logit-gap dump at greedy selection (COLI_LOGIT_GAP)
monotophic32dc3f95e94c
diff --git a/c/sample.h b/c/sample.hindex c1187cf3..332e13e7 100644--- a/c/sample.h+++ b/c/sample.h@@ -5,6 +5,7 @@#include <math.h>#include <stdio.h>+#include <stdlib.h>#include <string.h>#include "tok.h"@@ -114,7 +115,22 @@ static int pick_tok(const float *lo, int V, int ban){for (int k = 0; k < 5; k++) fprintf(stderr," %d:%.6f", id[k], v[k]);fprintf(stderr,"\n");}- if (g_temp <= 0) return argmax_v(lo, V);+ if (g_temp <= 0) {+ int a = argmax_v(lo, V);+ /* COLI_LOGIT_GAP=1: probe-only dump of the top-2 logits at each greedy+ * pick, to tell a near-tie (float accumulation order) apart from a real+ * divergence. Read-only: the token returned is unchanged. */+ static int gap_dbg = -1; static int gap_pos = 0;+ if (gap_dbg < 0) { const char *e = getenv("COLI_LOGIT_GAP"); gap_dbg = (e && *e && *e != '0'); }+ if (gap_dbg) {+ int b = -1; float bv = -INFINITY;+ for (int i = 0; i < V; i++) { float x = lo[i]; if (i != a && x == x && x > bv) { bv = x; b = i; } }+ double t1 = (double)lo[a], t2 = (b >= 0) ? (double)bv : 0.0;+ fprintf(stderr, "LOGITGAP pos=%d top1=%d:%.6f top2=%d:%.6f gap=%.9f\n",+ gap_pos++, a, t1, b, t2, (b >= 0) ? t1 - t2 : 0.0);+ }+ return a;+ }dist_build(lo, V);return dist_sample(V, ban);}