Merge t18-dropout into main

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

# Conflicts:
#	README.md
#	crates/xtrain-autodiff/tests/autograd.rs
#	crates/xtrain-model/src/model.rs
#	crates/xtrain-train/src/bin/train.rs
#	crates/xtrain-train/src/train_loop.rs
#	docs/evolution.md

crates/xtrain-train/src/bin/train.rs

@@ -113,6 +113,10 @@ fn main() {
     let val_tokens: usize = flag(&args, "--val-tokens", 0);
     let eval_every: usize = flag(&args, "--eval-every", 0);
     let eval_batches: usize = flag(&args, "--eval-batches", 64);
+    // Dropout (Phase T18): residual-path dropout prob, active at training time
+    // only (inverted scaling), identity at eval/sampling/export. Default 0 = off
+    // (forward graph bit-identical to the no-dropout path).
+    let dropout: f32 = flag(&args, "--dropout", 0.0f32);
     // bf16 mixed precision (Phase T12): fp32 master weights, bf16 linears +
     // activations. Opt-in; default fp32 reproduces v0–v4 numerics.
     let bf16 = args.iter().any(|a| a == "--bf16");
@@ -156,7 +160,8 @@ fn main() {
         (corpus, None)
     };
 
-    let cfg = Config::from_arch(vocab, n_heads, head_dim, n_layers, ffn);
+    let mut cfg = Config::from_arch(vocab, n_heads, head_dim, n_layers, ffn);
+    cfg.dropout = dropout;
     println!(
         "model: dim {} layers {} heads {} head_dim {} ffn {} → core {:.3}M params \
          (+ embed/lm {:.2}M = {:.2}M total)",
@@ -194,6 +199,9 @@ fn main() {
         model = model.with_flash(true);
         println!("flash-attention: ON (fused SDPA kernel, no materialized scores)");
     }
+    if dropout > 0.0 {
+        println!("dropout: ON (p={dropout}, residual-path, train-only inverted scaling)");
+    }
 
     // Eval-only mode: load a checkpoint and score it on the held-out val set, then
     // exit. Used to put an EXISTING model (e.g. v0) and a new one on the same

crates/xtrain-train/src/train_loop.rs

@@ -92,6 +92,10 @@ pub fn train(
         // is scaled by 1/accum before backward (the tape SUM-accumulates the scaled
         // grads). `accum == 1` skips the scale entirely → bit-identical to pre-T16.
         let mut step_loss_sum = 0.0f32;
+        // Training mode → dropout active (T18; no-op when cfg.dropout == 0). Set
+        // each step so it is restored after a periodic eval flips to eval mode.
+        // Each micro-step's forward bumps the per-step seed → fresh masks.
+        model.train();
         for _ in 0..accum {
             let mut inputs = Vec::with_capacity(cfg.batch_size);
             let mut targets_v = Vec::with_capacity(cfg.batch_size);
@@ -190,6 +194,8 @@ pub fn eval_loss(
     if valid.len() <= seq + 1 {
         return f32::NAN;
     }
+    // Eval mode → dropout is identity (T18).
+    model.eval();
     let n_win = (valid.len() - 1) / seq; // disjoint windows that fit
     let batches = batches.max(1).min(n_win.max(1));
     let stride = (n_win / batches).max(1);