diff --git a/hackable_diffusion/notebooks/dmmd_mnist_v2.ipynb b/hackable_diffusion/notebooks/dmmd_mnist_v2.ipynb
new file mode 100644
index 0000000..4e81c6d
--- /dev/null
+++ b/hackable_diffusion/notebooks/dmmd_mnist_v2.ipynb
@@ -0,0 +1,1063 @@
+{
+  "cells": [
+    {
+      "id": "63f8b627",
+      "cell_type": "markdown",
+      "source": [
+        "# DMMD on MNIST\n",
+        "\n",
+        "**Distribution Matching for Diffusion Model Distillation** ([arXiv:2405.06780](https://arxiv.org/abs/2405.06780))\n",
+        "\n",
+        "Trains a noise-conditional UNet critic $\\phi_\\theta$ to maximize the MMD² between clean and noisy data distributions, regularized by L2 + WGAN-GP gradient penalty. Samples are then generated by gradient flow on the learned witness function."
+      ],
+      "metadata": {
+        "id": "63f8b627"
+      }
+    },
+    {
+      "id": "5c10f802",
+      "cell_type": "code",
+      "source": [
+        "import os\n",
+        "os.environ['XLA_PYTHON_CLIENT_PREALLOCATE'] = 'false'  # MUST be before JAX import\n",
+        "\n",
+        "import functools\n",
+        "from etils import ecolab\n",
+        "import flax.linen as nn\n",
+        "import jax\n",
+        "import jax.numpy as jnp\n",
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "import optax\n",
+        "import tensorflow_datasets as tfds\n",
+        "import tqdm\n",
+        "\n",
+        "with ecolab.adhoc(\n",
+        "    reload=[\"hackable_diffusion\", \"kauldron\", \"lark\"],\n",
+        "    invalidate=False,\n",
+        "    cell_autoreload=True,\n",
+        "):\n",
+        "  from hackable_diffusion import hd\n",
+        "\n",
+        "gaussian = hd.corruption.gaussian\n",
+        "schedules = hd.corruption.schedules\n",
+        "\n",
+        "print(f\"Device: {jax.devices()[0]}\")"
+      ],
+      "metadata": {
+        "executionInfo": {
+          "elapsed": 24909,
+          "status": "ok",
+          "timestamp": 1784325385769,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "616651ea-55d7-43ec-f5e5-50445a43094b",
+        "id": "5c10f802"
+      },
+      "execution_count": 1,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Device: cuda:0\n"
+          ]
+        }
+      ]
+    },
+    {
+      "id": "aecb8a6a",
+      "cell_type": "markdown",
+      "source": [
+        "## Data & noise process"
+      ],
+      "metadata": {
+        "id": "aecb8a6a"
+      }
+    },
+    {
+      "id": "1c473e9d",
+      "cell_type": "code",
+      "source": [
+        "# Load MNIST — keep as numpy, only send batches to GPU\n",
+        "ds = tfds.as_numpy(tfds.load('mnist', split='train', batch_size=-1))\n",
+        "all_data = ds['image'].astype(np.float32) / 127.5 - 1.0\n",
+        "all_data = all_data.reshape(-1, 28, 28, 1)\n",
+        "all_data = np.tile(all_data, (1, 1, 1, 3))  # tile to 3ch for UNet compatibility\n",
+        "dataset_size = all_data.shape[0]\n",
+        "print(f'Loaded {dataset_size} samples, shape {all_data.shape} ({all_data.nbytes/1e6:.0f} MB on CPU)')\n",
+        "\n",
+        "# Noise process\n",
+        "schedule = schedules.CosineSchedule()\n",
+        "process = gaussian.GaussianProcess(schedule=schedule)\n",
+        "print(f'Noise schedule: {schedule}')"
+      ],
+      "metadata": {
+        "executionInfo": {
+          "elapsed": 4119,
+          "status": "ok",
+          "timestamp": 1784325390152,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "d6d9c71d-a289-4df9-c298-bc028acb3343",
+        "id": "1c473e9d"
+      },
+      "execution_count": 2,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Loaded 60000 samples, shape (60000, 28, 28, 3) (564 MB on CPU)\n",
+            "Noise schedule: <hackable_diffusion.lib.corruption.schedules.CosineSchedule object at 0x3306bd6dfcb0>\n"
+          ]
+        }
+      ]
+    },
+    {
+      "id": "f39ccb2a",
+      "cell_type": "markdown",
+      "source": [
+        "## UNet Critic"
+      ],
+      "metadata": {
+        "id": "f39ccb2a"
+      }
+    },
+    {
+      "id": "a949d79f",
+      "cell_type": "code",
+      "source": [
+        "class ResBlock(nn.Module):\n",
+        "  \"\"\"ResBlock with AdaGN time conditioning.\"\"\"\n",
+        "  channels: int\n",
+        "\n",
+        "  @nn.compact\n",
+        "  def __call__(self, x, t_emb, is_training=True):\n",
+        "    residual = x\n",
+        "    h = nn.GroupNorm(num_groups=max(1, min(x.shape[-1] // 4, 32)))(x)\n",
+        "    h = jax.nn.silu(h)\n",
+        "    h = nn.Conv(self.channels, (3, 3), padding='SAME')(h)\n",
+        "\n",
+        "    t_proj = nn.Dense(2 * self.channels)(jax.nn.silu(t_emb))\n",
+        "    scale, shift = jnp.split(t_proj, 2, axis=-1)\n",
+        "    h = nn.GroupNorm(num_groups=max(1, min(self.channels // 4, 32)))(h)\n",
+        "    h = h * (1 + scale[:, None, None, :]) + shift[:, None, None, :]\n",
+        "\n",
+        "    h = jax.nn.silu(h)\n",
+        "    h = nn.Conv(self.channels, (3, 3), padding='SAME',\n",
+        "                kernel_init=nn.initializers.truncated_normal(stddev=1e-8))(h)\n",
+        "\n",
+        "    if residual.shape[-1] != self.channels:\n",
+        "      residual = nn.Conv(self.channels, (1, 1))(residual)\n",
+        "    return (residual + h) / jnp.sqrt(2.0)\n",
+        "\n",
+        "\n",
+        "class UNetCritic(nn.Module):\n",
+        "  \"\"\"UNet critic with multi-scale feature extraction for DMMD.\"\"\"\n",
+        "  channels: int = 32\n",
+        "  channel_multipliers: tuple[int, ...] = (1, 2)\n",
+        "  num_residual_blocks: int = 1\n",
+        "  time_embedding_dim: int = 256\n",
+        "  feature_dim_per_stage: int = 32\n",
+        "  use_remat: bool = True\n",
+        "\n",
+        "  @nn.compact\n",
+        "  def __call__(self, x, t, is_training=True):\n",
+        "    B = x.shape[0]\n",
+        "    num_levels = len(self.channel_multipliers)\n",
+        "    Block = nn.remat(ResBlock) if self.use_remat else ResBlock\n",
+        "\n",
+        "    # Time embedding\n",
+        "    half = self.time_embedding_dim // 2\n",
+        "    freq = jnp.exp(jnp.arange(half) * -(jnp.log(10000.0) / (half - 1)))\n",
+        "    te = jnp.concatenate([jnp.sin(t[:, None] * freq), jnp.cos(t[:, None] * freq)], -1)\n",
+        "    te = jax.nn.silu(nn.Dense(self.time_embedding_dim, name='te0')(te))\n",
+        "    te = nn.Dense(self.time_embedding_dim, name='te1')(te)\n",
+        "\n",
+        "    # Encoder\n",
+        "    ch0 = self.channels * self.channel_multipliers[0]\n",
+        "    h = nn.Conv(ch0, (3, 3), padding='SAME', name='in_conv')(x)\n",
+        "    stack = [h]\n",
+        "    pre_down = h\n",
+        "\n",
+        "    for i, m in enumerate(self.channel_multipliers):\n",
+        "      ch = self.channels * m\n",
+        "      for j in range(self.num_residual_blocks):\n",
+        "        h = Block(ch, name=f'd{i}_{j}')(h, te, is_training)\n",
+        "        stack.append(h)\n",
+        "      if i < num_levels - 1:\n",
+        "        h = nn.avg_pool(h, (2, 2), strides=(2, 2))\n",
+        "        stack.append(h)\n",
+        "    down = h\n",
+        "\n",
+        "    # Middle\n",
+        "    ch_mid = self.channels * self.channel_multipliers[-1]\n",
+        "    h = Block(ch_mid, name='m0')(h, te, is_training)\n",
+        "    h = Block(ch_mid, name='m1')(h, te, is_training)\n",
+        "    mid = h\n",
+        "\n",
+        "    # Decoder\n",
+        "    for i in reversed(range(num_levels)):\n",
+        "      ch = self.channels * self.channel_multipliers[i]\n",
+        "      for j in range(self.num_residual_blocks + 1):\n",
+        "        h = jnp.concatenate([h, stack.pop()], axis=-1)\n",
+        "        h = Block(ch, name=f'u{i}_{j}')(h, te, is_training)\n",
+        "      if i > 0:\n",
+        "        h = jax.image.resize(h, (B, h.shape[1]*2, h.shape[2]*2, h.shape[3]), 'nearest')\n",
+        "        h = nn.Conv(self.channels * self.channel_multipliers[i-1], (3, 3),\n",
+        "                    padding='SAME', name=f'up_c{i}')(h)\n",
+        "    assert len(stack) == 0\n",
+        "    up = h\n",
+        "\n",
+        "    h = jax.nn.silu(nn.GroupNorm(max(1, min(h.shape[-1]//4, 32)), name='on')(h))\n",
+        "    h = nn.Conv(x.shape[-1], (3, 3), padding='SAME', name='out',\n",
+        "                kernel_init=nn.initializers.truncated_normal(stddev=1e-8))(h)\n",
+        "    last = h\n",
+        "\n",
+        "    # Multi-scale features: flatten + Dense\n",
+        "    feats = []\n",
+        "    # for name, tensor in [('pd', pre_down), ('dn', down), ('md', mid), ('up', up), ('la', last)]:\n",
+        "    for name, tensor in [('la', last)]:\n",
+        "      f = jax.nn.silu(nn.GroupNorm(max(1, min(tensor.shape[-1]//4, 32)), name=f'fn_{name}')(tensor))\n",
+        "      f = nn.Dense(self.feature_dim_per_stage, name=f'fp_{name}')(f.reshape(B, -1))\n",
+        "      feats.append(f)\n",
+        "    return jnp.concatenate(feats, axis=-1)\n",
+        "\n",
+        "\n",
+        "# critic = UNetCritic(channels=28, channel_multipliers=(1, 2, 2), num_residual_blocks=2,\n",
+        "#                     feature_dim_per_stage=32, use_remat=True)\n",
+        "\n",
+        "critic = UNetCritic(channels=28, channel_multipliers=(1, 1), num_residual_blocks=1,\n",
+        "                    feature_dim_per_stage=32, use_remat=True)\n",
+        "\n",
+        "rng = jax.random.PRNGKey(42)\n",
+        "rng, init_rng = jax.random.split(rng)\n",
+        "params = critic.init({'params': init_rng, 'dropout': init_rng},\n",
+        "                     jnp.ones((1, 28, 28, 3)), jnp.ones((1,)))['params']\n",
+        "num_params = sum(p.size for p in jax.tree.leaves(params))\n",
+        "print(f\"Critic: {num_params:,} params, remat=True\")"
+      ],
+      "metadata": {
+        "executionInfo": {
+          "elapsed": 1541,
+          "status": "ok",
+          "timestamp": 1784325597570,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "7e6cafdb-5df8-4bd3-f668-27350de723d5",
+        "id": "a949d79f"
+      },
+      "execution_count": 8,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Critic: 479,777 params, remat=True\n"
+          ]
+        }
+      ]
+    },
+    {
+      "id": "4828b2dc",
+      "cell_type": "markdown",
+      "source": [
+        "## DMMD Loss\n",
+        "\n",
+        "Single function: MMD² + L2 + gradient penalty, tiled over noise levels. Standard batched VJP for GP."
+      ],
+      "metadata": {
+        "id": "4828b2dc"
+      }
+    },
+    {
+      "id": "760435e6",
+      "cell_type": "code",
+      "source": [
+        "# ---- Hyperparameters (matching original MMDOptimizationConfig defaults) ----\n",
+        "num_noise_levels = 4\n",
+        "l2_coeff = 0.1\n",
+        "gp_coeff = 1.0\n",
+        "time_eps = 1e-3\n",
+        "\n",
+        "\n",
+        "def forward_fn(x, t, params):\n",
+        "  \"\"\"Forward pass through critic.\"\"\"\n",
+        "  return critic.apply(\n",
+        "      {'params': params}, x, t,\n",
+        "      is_training=True, rngs={'dropout': jax.random.PRNGKey(0)},\n",
+        "  )\n",
+        "\n",
+        "\n",
+        "def dmmd_loss_fn(params, x0, rng):\n",
+        "  \"\"\"DMMD loss: -MMD² + L2 + GP.\n",
+        "\n",
+        "  Faithful reproduction of _compute_total_loss_optimized from the original\n",
+        "  MMDDiffusionGradientFlowModel.\n",
+        "  \"\"\"\n",
+        "  B = x0.shape[0]\n",
+        "  ndim = len(x0.shape[1:])  # 3\n",
+        "\n",
+        "  rng, trng, nrng, arng = jax.random.split(rng, 4)\n",
+        "\n",
+        "  # Sample noise levels\n",
+        "  times = jax.random.uniform(\n",
+        "      trng, (num_noise_levels,), minval=time_eps, maxval=1.0 - time_eps\n",
+        "  )\n",
+        "\n",
+        "  # Tile data and time — [NL*B, ...]\n",
+        "  t_rep = jnp.repeat(times, B)                                   # [NL*B]\n",
+        "  x0_rep = jnp.tile(x0, (num_noise_levels,) + (1,) * ndim)       # [NL*B, H, W, C]\n",
+        "\n",
+        "  # Corrupt\n",
+        "  xt, _ = process.corrupt(key=nrng, x0=x0_rep, time=t_rep)\n",
+        "\n",
+        "  # Forward — single batched call each\n",
+        "  phi_clean = forward_fn(x0_rep, t_rep, params)   # [NL*B, D]\n",
+        "  phi_noisy = forward_fn(xt, t_rep, params)        # [NL*B, D]\n",
+        "\n",
+        "  # Reshape to [NL, B, D]\n",
+        "  D = phi_clean.shape[-1]\n",
+        "  phi_c = phi_clean.reshape(num_noise_levels, B, D)\n",
+        "  phi_n = phi_noisy.reshape(num_noise_levels, B, D)\n",
+        "\n",
+        "  # ---------- MMD² (unbiased, linear kernel) ----------\n",
+        "  sum_phi_c = jnp.sum(phi_c, axis=1, keepdims=True)   # [NL, 1, D]\n",
+        "  sum_phi_n = jnp.sum(phi_n, axis=1, keepdims=True)\n",
+        "\n",
+        "  n_sq = B * B\n",
+        "  n_sq_m1 = B * (B - 1)\n",
+        "\n",
+        "  def _dot(x, y):\n",
+        "    return jnp.sum(x * y, axis=-1)\n",
+        "\n",
+        "  k_xx = jnp.mean(\n",
+        "      _dot(sum_phi_n, sum_phi_n)\n",
+        "      - jnp.sum(_dot(phi_n, phi_n), axis=1, keepdims=True)\n",
+        "  ) / n_sq_m1\n",
+        "\n",
+        "  k_yy = jnp.mean(\n",
+        "      _dot(sum_phi_c, sum_phi_c)\n",
+        "      - jnp.sum(_dot(phi_c, phi_c), axis=1, keepdims=True)\n",
+        "  ) / n_sq_m1\n",
+        "\n",
+        "  k_xy = jnp.mean(_dot(sum_phi_n, sum_phi_c)) / n_sq\n",
+        "\n",
+        "  mmd_sq = k_xx + k_yy - 2 * k_xy\n",
+        "  total_loss = -mmd_sq\n",
+        "\n",
+        "  # ---------- L2 penalty ----------\n",
+        "  l2 = jnp.mean(jnp.sum(phi_n**2 + phi_c**2, axis=-1))\n",
+        "  total_loss += l2_coeff * l2\n",
+        "\n",
+        "  # ---------- Gradient penalty (WGAN-GP, original formulation) ----------\n",
+        "  # Interpolate between noisy and clean\n",
+        "  alpha = jax.random.uniform(\n",
+        "      arng, (num_noise_levels * B,) + (1,) * ndim\n",
+        "  )\n",
+        "  mixed = xt * alpha + (1.0 - alpha) * x0_rep   # [NL*B, H, W, C]\n",
+        "\n",
+        "  # Per-noise-level witness direction, tiled to [NL*B, D]\n",
+        "  diff = jax.lax.stop_gradient(\n",
+        "      jnp.mean(phi_c - phi_n, axis=1, keepdims=True)   # [NL, 1, D]\n",
+        "  )\n",
+        "  diff = jnp.tile(diff, (1, B, 1)).reshape(num_noise_levels * B, D)\n",
+        "\n",
+        "  # witness_train: dot(diff, phi(z)) — per-sample scalar\n",
+        "  def witness_fn(x):\n",
+        "    phi = forward_fn(x, t_rep, params)        # [NL*B, D]\n",
+        "    return jnp.sum(diff * phi, axis=-1)        # [NL*B]\n",
+        "\n",
+        "  # Batched per-sample gradient via VJP (original _vgrad pattern)\n",
+        "  def _vgrad(f, x):\n",
+        "    y, vjp_fn = jax.vjp(f, x)\n",
+        "    return vjp_fn(jnp.ones(y.shape))[0]\n",
+        "\n",
+        "  witness_grads = _vgrad(witness_fn, mixed)     # [NL*B, H, W, C]\n",
+        "  witness_grads = witness_grads.reshape(num_noise_levels * B, -1)\n",
+        "  grad_norms = jnp.sqrt(\n",
+        "      1e-8 + jnp.sum(jnp.square(witness_grads), axis=-1)\n",
+        "  )\n",
+        "  gp = jnp.mean(jnp.square(grad_norms - 1.0))\n",
+        "  total_loss += gp_coeff * gp\n",
+        "\n",
+        "  return total_loss, {\n",
+        "      'mmd_sq': mmd_sq, 'l2': l2, 'gp': gp, 'loss': total_loss,\n",
+        "  }\n",
+        "\n",
+        "\n",
+        "print(f\"Loss config: NL={num_noise_levels}, l2={l2_coeff}, gp={gp_coeff}\")"
+      ],
+      "metadata": {
+        "executionInfo": {
+          "elapsed": 54,
+          "status": "ok",
+          "timestamp": 1784325599648,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "a49989cb-132b-484f-e16f-d6cb89c76128",
+        "id": "760435e6"
+      },
+      "execution_count": 9,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Loss config: NL=4, l2=0.1, gp=1.0\n"
+          ]
+        }
+      ]
+    },
+    {
+      "id": "1faca00a",
+      "cell_type": "markdown",
+      "source": [
+        "## Training"
+      ],
+      "metadata": {
+        "id": "1faca00a"
+      }
+    },
+    {
+      "id": "7452dd2f",
+      "cell_type": "code",
+      "source": [
+        "optimizer = optax.chain(\n",
+        "    optax.clip_by_global_norm(max_norm=1.0),\n",
+        "    optax.scale_by_adam(b1=0.9, b2=0.999, eps=1e-8),\n",
+        "    optax.scale_by_schedule(optax.constant_schedule(value=2e-4)),\n",
+        "    optax.scale(-1.0),\n",
+        ")\n",
+        "opt_state = optimizer.init(params)\n",
+        "\n",
+        "\n",
+        "@jax.jit\n",
+        "def train_step(params, opt_state, x0, rng):\n",
+        "  grads, metrics = jax.grad(dmmd_loss_fn, has_aux=True)(params, x0, rng)\n",
+        "  updates, opt_state = optimizer.update(grads, opt_state)\n",
+        "  params = optax.apply_updates(params, updates)\n",
+        "  return params, opt_state, metrics\n",
+        "\n",
+        "\n",
+        "batch_size = 32\n",
+        "nepochs = 20\n",
+        "steps_per_epoch = dataset_size // batch_size\n",
+        "\n",
+        "losses, mmd_sqs = [], []\n",
+        "for epoch in tqdm.tqdm(range(1, nepochs + 1)):\n",
+        "  rng, shrng = jax.random.split(rng)\n",
+        "  perm = np.array(jax.random.permutation(shrng, dataset_size))\n",
+        "\n",
+        "  ep_loss, ep_mmd, ep_l2, ep_gp = 0.0, 0.0, 0.0, 0.0\n",
+        "  for i in range(steps_per_epoch):\n",
+        "    x0 = jnp.array(all_data[perm[i*batch_size:(i+1)*batch_size]])\n",
+        "    rng, srng = jax.random.split(rng)\n",
+        "    params, opt_state, m = train_step(params, opt_state, x0, srng)\n",
+        "    ep_loss += float(m['loss'])\n",
+        "    ep_mmd += float(m['mmd_sq'])\n",
+        "    ep_l2 += float(m['l2'])\n",
+        "    ep_gp += float(m['gp'])\n",
+        "\n",
+        "  n = steps_per_epoch\n",
+        "  losses.append(ep_loss / n)\n",
+        "  mmd_sqs.append(ep_mmd / n)\n",
+        "  print(\n",
+        "      f'Epoch {epoch:2d}  loss={ep_loss/n:.4f}  '\n",
+        "      f'MMD²={ep_mmd/n:.4f}  L2={ep_l2/n:.4f}  GP={ep_gp/n:.4f}'\n",
+        "  )"
+      ],
+      "metadata": {
+        "executionInfo": {
+          "elapsed": 1374248,
+          "status": "ok",
+          "timestamp": 1784327242649,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "fe59373e-5232-43c5-c5af-ecac7f0f9241",
+        "id": "7452dd2f"
+      },
+      "execution_count": 13,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "  5%|▌         | 1/20 [01:34<29:56, 94.57s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  1  loss=-30.9098  MMD²=1550.1995  L2=778.0146  GP=1441.4882\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 10%|█         | 2/20 [02:42<23:35, 78.62s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  2  loss=-31.0163  MMD²=1551.7946  L2=778.8761  GP=1442.8907\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 15%|█▌        | 3/20 [03:48<20:38, 72.85s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  3  loss=-30.8160  MMD²=1541.7975  L2=773.4673  GP=1433.6348\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 20%|██        | 4/20 [04:53<18:41, 70.12s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  4  loss=-30.4913  MMD²=1535.1530  L2=770.0579  GP=1427.6559\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 25%|██▌       | 5/20 [05:59<17:09, 68.60s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  5  loss=-31.4627  MMD²=1541.9113  L2=773.4177  GP=1433.1068\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 30%|███       | 6/20 [07:07<15:54, 68.20s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  6  loss=-31.2200  MMD²=1555.7363  L2=780.2735  GP=1446.4890\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 35%|███▌      | 7/20 [08:15<14:48, 68.36s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  7  loss=-30.1000  MMD²=1513.9629  L2=759.1867  GP=1407.9442\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 40%|████      | 8/20 [09:21<13:31, 67.61s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  8  loss=-30.8421  MMD²=1524.5666  L2=764.5307  GP=1417.2714\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 45%|████▌     | 9/20 [10:27<12:17, 67.06s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch  9  loss=-30.4659  MMD²=1548.1768  L2=776.1558  GP=1440.0953\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 50%|█████     | 10/20 [11:33<11:07, 66.76s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 10  loss=-30.4850  MMD²=1555.0813  L2=779.6286  GP=1446.6334\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 55%|█████▌    | 11/20 [12:42<10:05, 67.23s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 11  loss=-30.8509  MMD²=1544.4678  L2=774.3719  GP=1436.1797\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 60%|██████    | 12/20 [13:51<09:02, 67.80s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 12  loss=-30.4405  MMD²=1549.0609  L2=776.5380  GP=1440.9666\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 65%|██████▌   | 13/20 [14:59<07:54, 67.78s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 13  loss=-31.3037  MMD²=1552.4509  L2=778.0128  GP=1443.3459\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 70%|███████   | 14/20 [16:05<06:43, 67.27s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 14  loss=-30.7358  MMD²=1549.6897  L2=776.6884  GP=1441.2850\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 75%|███████▌  | 15/20 [17:11<05:34, 66.93s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 15  loss=-31.6441  MMD²=1565.8537  L2=784.8329  GP=1455.7263\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 80%|████████  | 16/20 [18:17<04:26, 66.71s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 16  loss=-30.6257  MMD²=1565.2247  L2=784.5474  GP=1456.1443\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 85%|████████▌ | 17/20 [19:27<03:23, 67.67s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 17  loss=-30.9462  MMD²=1546.1134  L2=774.8564  GP=1437.6815\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 90%|█████████ | 18/20 [20:36<02:16, 68.11s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 18  loss=-31.3862  MMD²=1533.1580  L2=768.1681  GP=1424.9551\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\r 95%|█████████▌| 19/20 [21:42<01:07, 67.53s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 19  loss=-30.7256  MMD²=1545.5131  L2=774.6784  GP=1437.3197\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "100%|██████████| 20/20 [22:54<00:00, 68.71s/it]"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Epoch 20  loss=-30.9601  MMD²=1559.5342  L2=781.4278  GP=1450.4313\n"
+          ]
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "\n"
+          ]
+        }
+      ]
+    },
+    {
+      "id": "26641591",
+      "cell_type": "code",
+      "source": [
+        "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))\n",
+        "ax1.plot(losses); ax1.set(xlabel='Epoch', ylabel='Loss', title='Total Loss (should decrease)')\n",
+        "ax2.plot(mmd_sqs); ax2.set(xlabel='Epoch', ylabel='MMD²', title='MMD² (should increase)')\n",
+        "plt.tight_layout(); plt.show()"
+      ],
+      "metadata": {
+        "colab": {
+          "height": 296
+        },
+        "executionInfo": {
+          "elapsed": 507,
+          "status": "ok",
+          "timestamp": 1784327243450,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "a3903fe8-4bb8-475c-98ee-c13bc2c2762e",
+        "id": "26641591"
+      },
+      "execution_count": 14,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 864x288 with 2 Axes>"
+            ]
+          },
+          "metadata": {
+            "image/png": {
+              "height": 279,
+              "width": 856
+            },
+            "needs_background": "light"
+          }
+        }
+      ]
+    },
+    {
+      "id": "61eed3b7",
+      "cell_type": "markdown",
+      "source": [
+        "## Sampling via gradient flow\n",
+        "\n",
+        "Starting from noise $z \\sim \\mathcal{N}(0, I)$, solve the gradient flow $dz/dt = -\\nabla_z \\text{witness}(z, t)$ backward in time from $t=1$ to $t=0$."
+      ],
+      "metadata": {
+        "id": "61eed3b7"
+      }
+    },
+    {
+      "id": "d5d64ea3",
+      "cell_type": "code",
+      "source": [
+        "# ---- Sampling: gradient flow on the witness function ----\n",
+        "# Faithful reproduction of witness_f_sampling + sample() from the original.\n",
+        "#\n",
+        "# witness_f(z_i) = <E_loo[φ(Z)] − E[φ(X₀)], φ(z_i)>\n",
+        "#\n",
+        "# where E_loo[φ(Z)] = (Σ_j φ(z_j) − stop_grad(φ(z_i))) / (m−1)\n",
+        "# is the leave-one-out mean of ALL particles (not training data).\n",
+        "\n",
+        "\n",
+        "@jax.jit\n",
+        "def _gradient_step(xt, phi_clean_mean, t, params, lr):\n",
+        "  \"\"\"One gradient flow step for all particles.\"\"\"\n",
+        "  N = xt.shape[0]\n",
+        "  t_arr = jnp.full((N,), t)\n",
+        "\n",
+        "  # Pre-compute features for ALL particles (1 batched forward)\n",
+        "  phi_all = forward_fn(xt, t_arr, params)   # [N, D]\n",
+        "  phi_all_sum = jnp.sum(phi_all, axis=0)    # [D]\n",
+        "\n",
+        "  def witness_single(z):\n",
+        "    \"\"\"Witness value for one particle — matches witness_f_sampling.\"\"\"\n",
+        "    phi_z = forward_fn(z[None], jnp.full((1,), t), params)[0]   # [D]\n",
+        "    # Leave-one-out mean of noisy particles (original formula)\n",
+        "    expected_noisy = (\n",
+        "        phi_all_sum - jax.lax.stop_gradient(phi_z)\n",
+        "    ) / (N - 1)\n",
+        "    expected_clean = phi_clean_mean   # [D]\n",
+        "    return jnp.sum((expected_noisy - expected_clean) * phi_z)\n",
+        "\n",
+        "  grads = jax.vmap(jax.grad(witness_single))(xt)   # [N, H, W, C]\n",
+        "  return xt - lr * grads\n",
+        "\n",
+        "\n",
+        "def sample_dmmd(params, num_samples=64, num_timesteps=10,\n",
+        "                num_steps_per_noise=5, grad_flow_lr=1.0, rng=None):\n",
+        "  \"\"\"Sample by gradient flow — matches original sample() method.\"\"\"\n",
+        "  if rng is None:\n",
+        "    rng = jax.random.PRNGKey(0)\n",
+        "  xt = jax.random.normal(rng, (num_samples, 28, 28, 3))\n",
+        "\n",
+        "  timesteps = jnp.linspace(1.0 - time_eps, time_eps, num_timesteps)\n",
+        "\n",
+        "  for t in tqdm.tqdm(timesteps, desc='Sampling'):\n",
+        "    t = float(t)\n",
+        "    # Compute E[φ(X₀)] from clean training data\n",
+        "    rng, drng = jax.random.split(rng)\n",
+        "    ref_idx = np.array(jax.random.choice(drng, dataset_size, (500,), replace=False))\n",
+        "    ref_x0 = jnp.array(all_data[ref_idx])\n",
+        "    phi_clean = forward_fn(ref_x0, jnp.full((500,), t), params)\n",
+        "    phi_clean_mean = jnp.mean(phi_clean, axis=0)   # [D]\n",
+        "\n",
+        "    for _ in range(num_steps_per_noise):\n",
+        "      xt = _gradient_step(xt, phi_clean_mean, t, params, grad_flow_lr)\n",
+        "\n",
+        "  return jnp.clip(xt, -1, 1)\n",
+        "\n",
+        "\n",
+        "print(\"Sampling function ready (original leave-one-out witness)\")"
+      ],
+      "metadata": {
+        "executionInfo": {
+          "elapsed": 55,
+          "status": "ok",
+          "timestamp": 1784327243798,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "cdb17333-f134-4126-d979-bd6225838c8a",
+        "id": "d5d64ea3"
+      },
+      "execution_count": 15,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Sampling function ready (original leave-one-out witness)\n"
+          ]
+        }
+      ]
+    },
+    {
+      "id": "500b9209",
+      "cell_type": "code",
+      "source": [
+        "samples = np.array(sample_dmmd(\n",
+        "    params, num_samples=64, num_timesteps=10,\n",
+        "    num_steps_per_noise=10, grad_flow_lr=0.1,\n",
+        "    rng=jax.random.PRNGKey(123),\n",
+        "))\n",
+        "\n",
+        "fig, axes = plt.subplots(8, 8, figsize=(8, 8))\n",
+        "for img, ax in zip(samples, axes.flatten()):\n",
+        "  ax.imshow((img[:, :, 0] + 1) / 2, cmap='gray', vmin=0, vmax=1)\n",
+        "  ax.axis('off')\n",
+        "fig.suptitle('DMMD samples', fontsize=14)\n",
+        "plt.tight_layout(); plt.show()\n",
+        "\n",
+        "# Real data for comparison\n",
+        "fig, axes = plt.subplots(8, 8, figsize=(8, 8))\n",
+        "for i, ax in enumerate(axes.flatten()):\n",
+        "  ax.imshow((all_data[i, :, :, 0] + 1) / 2, cmap='gray', vmin=0, vmax=1)\n",
+        "  ax.axis('off')\n",
+        "fig.suptitle('Real data', fontsize=14)\n",
+        "plt.tight_layout(); plt.show()"
+      ],
+      "metadata": {
+        "colab": {
+          "height": 1000
+        },
+        "executionInfo": {
+          "elapsed": 39977,
+          "status": "ok",
+          "timestamp": 1784327284058,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          },
+          "user_tz": -60
+        },
+        "outputId": "4332f496-668c-4331-c8b1-77b93cd43424",
+        "id": "500b9209"
+      },
+      "execution_count": 16,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "Sampling: 100%|██████████| 10/10 [00:36<00:00,  3.62s/it]\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 576x576 with 64 Axes>"
+            ]
+          },
+          "metadata": {
+            "image/png": {
+              "height": 568,
+              "width": 564
+            },
+            "needs_background": "light"
+          }
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 576x576 with 64 Axes>"
+            ]
+          },
+          "metadata": {
+            "image/png": {
+              "height": 568,
+              "width": 564
+            },
+            "needs_background": "light"
+          }
+        }
+      ]
+    },
+    {
+      "id": "-X-WMvAVyAGL",
+      "cell_type": "code",
+      "source": [],
+      "metadata": {
+        "id": "-X-WMvAVyAGL",
+        "executionInfo": {
+          "status": "ok",
+          "timestamp": 1784327284430,
+          "user_tz": -60,
+          "elapsed": 1,
+          "user": {
+            "displayName": "",
+            "userId": ""
+          }
+        }
+      },
+      "execution_count": 16,
+      "outputs": []
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python"
+    },
+    "colab": {
+      "provenance": [],
+      "last_runtime": {
+        "build_target": "//gdm/science/msgym/exploratory:msgym_brain_ext_notebook_gpu",
+        "kind": "private"
+      }
+    }
+  },
+  "nbformat_minor": 5,
+  "nbformat": 4
+}