diff --git a/examples/01_Model_Training.ipynb b/examples/01_Model_Training.ipynb
index 46f0c786..e05f85fe 100644
--- a/examples/01_Model_Training.ipynb
+++ b/examples/01_Model_Training.ipynb
@@ -44,49 +44,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/linux3_i1/segreto/miniconda3/envs/apax/lib/python3.11/pty.py:89: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.\n",
-      "  pid, fd = os.forkpty()\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1m                                                                                \u001b[0m\n",
-      "\u001b[1m \u001b[0m\u001b[1;33mUsage: \u001b[0m\u001b[1mapax [OPTIONS] COMMAND [ARGS]...\u001b[0m\u001b[1m                                       \u001b[0m\u001b[1m \u001b[0m\n",
-      "\u001b[1m                                                                                \u001b[0m\n",
-      "\u001b[2m╭─\u001b[0m\u001b[2m Options \u001b[0m\u001b[2m───────────────────────────────────────────────────────────────────\u001b[0m\u001b[2m─╮\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m-\u001b[0m\u001b[1;36m-version\u001b[0m             \u001b[1;32m-V\u001b[0m                                                     \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m-\u001b[0m\u001b[1;36m-install\u001b[0m\u001b[1;36m-completion\u001b[0m            Install completion for the current shell.    \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m-\u001b[0m\u001b[1;36m-show\u001b[0m\u001b[1;36m-completion\u001b[0m               Show completion for the current shell, to    \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m                                 copy it or customize the installation.       \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m-\u001b[0m\u001b[1;36m-help\u001b[0m                \u001b[1;32m-h\u001b[0m        Show this message and exit.                  \u001b[2m│\u001b[0m\n",
-      "\u001b[2m╰──────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n",
-      "\u001b[2m╭─\u001b[0m\u001b[2m Commands \u001b[0m\u001b[2m──────────────────────────────────────────────────────────────────\u001b[0m\u001b[2m─╮\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36mdocs     \u001b[0m\u001b[1;36m \u001b[0m Opens the documentation website in your browser.                  \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36meval     \u001b[0m\u001b[1;36m \u001b[0m Starts performing the evaluation of the test dataset with         \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m          \u001b[0m parameters provided by a configuration file.                      \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36mmd       \u001b[0m\u001b[1;36m \u001b[0m Starts performing a molecular dynamics simulation (currently only \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m          \u001b[0m NHC thermostat) with parameters provided by a configuration file. \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36mtemplate \u001b[0m\u001b[1;36m \u001b[0m Create configuration file templates.                              \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36mtrain    \u001b[0m\u001b[1;36m \u001b[0m Starts the training of a model with parameters provided by a      \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m          \u001b[0m configuration file.                                               \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36mvalidate \u001b[0m\u001b[1;36m \u001b[0m Validate training or MD config files.                             \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36mvisualize\u001b[0m\u001b[1;36m \u001b[0m Visualize a model based on a configuration file. A CO molecule is \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m          \u001b[0m taken as sample input (influences number of atoms, number of      \u001b[2m│\u001b[0m\n",
-      "\u001b[2m│\u001b[0m \u001b[1;36m          \u001b[0m species is set to 10).                                            \u001b[2m│\u001b[0m\n",
-      "\u001b[2m╰──────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n",
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "!apax -h"
    ]
@@ -100,7 +60,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -132,7 +92,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -168,18 +128,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[32mSuccess!\u001b[0m\n",
-      "config.yaml is a valid training config.\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "!apax validate train config.yaml"
    ]
@@ -194,7 +145,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -209,21 +160,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "1 validation error for Config\n",
-      "n_epochs\n",
-      "  Input should be greater than 0 [type=greater_than, input_value=-1000, input_type=int]\n",
-      "    For further information visit https://errors.pydantic.dev/2.6/v/greater_than\n",
-      "\u001b[31mConfiguration Invalid!\u001b[0m\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "!apax validate train error_config.yaml"
    ]
@@ -239,32 +178,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "INFO | 21:56:40 | Initializing Callbacks\n",
-      "INFO | 21:56:40 | Initializing Loss Function\n",
-      "INFO | 21:56:40 | Initializing Metrics\n",
-      "INFO | 21:56:40 | Running Input Pipeline\n",
-      "INFO | 21:56:40 | Read data file project/benzene_mod.xyz\n",
-      "INFO | 21:56:40 | Loading data from project/benzene_mod.xyz\n",
-      "INFO | 21:56:50 | Precomputing neighborlists\n",
-      "Precomputing NL: 100%|███████████████████████████████████████| 1000/1000 [00:00<00:00, 13135.40it/s]\n",
-      "INFO | 21:56:50 | Computing per element energy regression.\n",
-      "INFO | 21:56:57 | Precomputing neighborlists\n",
-      "Precomputing NL: 100%|█████████████████████████████████████████| 100/100 [00:00<00:00, 12868.33it/s]\n",
-      "INFO | 21:56:57 | Initializing Model\n",
-      "INFO | 21:56:58 | initializing 1 models\n",
-      "INFO | 21:57:04 | Initializing Optimizer\n",
-      "INFO | 21:57:04 | Beginning Training\n",
-      "Epochs: 100%|████████████████████████████████████████| 10/10 [00:28<00:00,  2.82s/it, val_loss=0.63]\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "!apax train config.yaml"
    ]
@@ -287,24 +203,14 @@
     "\n",
     "If training is interrupted for any reason, re-running the above `train` command will resume training from the latest checkpoint.\n",
     "\n",
-    "Furthermore, an Apax trianing can easily be started within a scriped."
+    "Furthermore, an Apax trianing can easily be started within a script."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Precomputing NL: 100%|███████████████████████████████████████| 1000/1000 [00:00<00:00, 10410.16it/s]\n",
-      "Precomputing NL: 100%|█████████████████████████████████████████| 100/100 [00:00<00:00, 11413.38it/s]\n",
-      "Epochs: 100%|██████████████████████████████████████| 100/100 [03:43<00:00,  2.24s/it, val_loss=0.31]\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from apax.train.run import run\n",
     "\n",
@@ -328,50 +234,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1800x400 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1800x400 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1800x400 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1800x400 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "import csv\n",
     "import matplotlib.pyplot as plt\n",
@@ -399,21 +264,22 @@
     "            key = headers[idx]\n",
     "            data_dict[key].append(float(value))\n",
     "\n",
-    "for key in keys:\n",
-    "    fig, axes = plt.subplots(1, 2, sharey=True, sharex=True, figsize=(18, 4))\n",
+    "fig, axes = plt.subplots(2, 2, constrained_layout=True)\n",
+    "axes = axes.ravel()\n",
+    "fig.suptitle(f'Metrics', fontsize=16)\n",
     "\n",
+    "for id, key in enumerate(keys):\n",
     "    val = np.array(data_dict[f\"val_{key}\"])\n",
     "    train = np.array(data_dict[f\"train_{key}\"])\n",
     "    epoch = np.array(data_dict[\"epoch\"])\n",
     "\n",
-    "    axes[0].plot(epoch, val)\n",
-    "    axes[1].plot(epoch, train)\n",
+    "    axes[id].plot(epoch, val, label=\"val data\")\n",
+    "    axes[id].plot(epoch, train, label=\"train data\")\n",
     "\n",
-    "    axes[0].set_ylabel(f\"val_{key}\")\n",
-    "    axes[0].set_xlabel(r\"epoch\")\n",
-    "    axes[1].set_ylabel(f\"train_{key}\")\n",
-    "    axes[1].set_xlabel(r\"epoch\")\n",
-    "    fig.show()"
+    "    axes[id].set_ylabel(f\"{key}\")\n",
+    "    axes[id].set_xlabel(r\"epoch\")\n",
+    "\n",
+    "plt.show()"
    ]
   },
   {
@@ -428,31 +294,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/linux3_i1/segreto/miniconda3/envs/apax/lib/python3.11/pty.py:89: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.\n",
-      "  pid, fd = os.forkpty()\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[33mUsage: \u001b[0mapax [OPTIONS] COMMAND [ARGS]...\n",
-      "\u001b[2mTry \u001b[0m\u001b[2;34m'apax \u001b[0m\u001b[1;2;34m-h\u001b[0m\u001b[2;34m'\u001b[0m\u001b[2m for help.\u001b[0m\n",
-      "\u001b[31m╭─\u001b[0m\u001b[31m Error \u001b[0m\u001b[31m─────────────────────────────────────────────────────────────────────\u001b[0m\u001b[31m─╮\u001b[0m\n",
-      "\u001b[31m│\u001b[0m No such command 'evaluate'.                                                  \u001b[31m│\u001b[0m\n",
-      "\u001b[31m╰──────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "!apax evaluate config_minimal.yaml"
+    "from apax.train.eval import eval_model\n",
+    "\n",
+    "eval_model(config_dict, n_test=100)\n",
+    "# !apax eval config.yaml"
    ]
   },
   {
@@ -471,14 +320,96 @@
    "source": [
     "## A Closer Look At Training Parameters\n",
     "\n",
-    "TODO"
+    "```yaml\n",
+    "n_epochs: <NUMBER OF EPOCHS> # Number of training epochs.\n",
+    "seed: 1 # Seed for initialising random numbers\n",
+    "patience: None # Number of epochs without improvement before trainings gets terminated.\n",
+    "n_models: 1 # Number of models to be trained at once.\n",
+    "n_jitted_steps: 1 # Number of train batches to be processed in a compiled loop. Can yield singificant speedups for small structures or small batch sizes.\n",
+    "\n",
+    "data:\n",
+    "  directory: models/ # Path to the directory where the training results and checkpoints will be written.\n",
+    "  experiment: apax # Name of  the model. Distinguishes it from the other models trained in the same `directory`.\n",
+    "  data_path: <PATH> # Path to a single dataset file. Set either this or `val_data_path` and `train_data_path`.\n",
+    "  train_data_path: <PATH> # Path to a training dataset. Set this and `val_data_path` if your data comes pre-split.\n",
+    "  val_data_path: <PATH> # Path to a validation dataset. Set this and `train_data_path` if your data comes pre-split.\n",
+    "  test_data_path: <PATH> # Path to a test dataset. Set this, `train_data_path` and `val_data_path` if your data comes pre-split.\n",
+    "\n",
+    "  n_train: 1000 # Number of training datapoints from `data_path`.\n",
+    "  n_valid: 100 # Number of validation datapoints from `data_path`.\n",
+    "\n",
+    "  batch_size: 32 # Number of training examples to be evaluated at once.\n",
+    "  valid_batch_size: 100 # Number of validation examples to be evaluated at once.\n",
+    "\n",
+    "  shift_method: \"per_element_regression_shift\"\n",
+    "  shift_options:\n",
+    "    energy_regularisation: 1.0 # Magnitude of the regularization in the per-element energy regression.\n",
+    "  shuffle_buffer_size: 1000 # Size of the `tf.data` shuffle buffer.\n",
+    "\n",
+    "  pos_unit: Ang\n",
+    "  energy_unit: eV\n",
+    "\n",
+    "  additional_properties_info: # Dict of property name, shape (ragged or fixed) pairs\n",
+    "\n",
+    "model:\n",
+    "  n_basis: 7 # Number of uncontracted gaussian basis functions.\n",
+    "  n_radial: 5 # Number of contracted basis functions.\n",
+    "  nn: [512, 512] # Number of hidden layers and units in those layers.\n",
+    "\n",
+    "  r_max: 6.0 # Position of the first uncontracted basis function's mean.\n",
+    "  r_min: 0.5 # Cutoff radius of the descriptor.\n",
+    "\n",
+    "  use_zbl: false # \n",
+    "\n",
+    "  b_init: normal # Initialization scheme for the neural network biases. Either `normal` or `zeros`.\n",
+    "  descriptor_dtype: fp64\n",
+    "  readout_dtype: fp32\n",
+    "  scale_shift_dtype: fp32\n",
+    "\n",
+    "loss:\n",
+    "- loss_type: structures # Weighting scheme for atomic contributions. See the MLIP package for reference 10.1088/2632-2153/abc9fe for details\n",
+    "  name: energy # Keyword of the quantity e.g `energy`.\n",
+    "  weight: 1.0 # Weighting factor in the overall loss function.\n",
+    "- loss_type: structures\n",
+    "  name: forces\n",
+    "  weight: 4.0\n",
+    "\n",
+    "metrics:\n",
+    "- name: energy # Keyword of the quantity e.g `energy`.\n",
+    "  reductions: # List of reductions performed on the difference between target and predictions. Can be mae, mse, rmse for energies and forces. For forces it is also possible to use `angle`.\n",
+    "  - mae\n",
+    "- name: forces\n",
+    "  reductions:\n",
+    "  - mae\n",
+    "  - mse\n",
+    "\n",
+    "optimizer:\n",
+    "  opt_name: adam # Name of the optimizer. Can be any `optax` optimizer.\n",
+    "  opt_kwargs: {} # Optimizer keyword arguments. Passed to the `optax` optimizer.\n",
+    "  emb_lr: 0.03 # Learning rate of the elemental embedding contraction coefficients.\n",
+    "  nn_lr: 0.03 # Learning rate of the neural network parameters.\n",
+    "  scale_lr: 0.001 # Learning rate of the elemental output scaling factors.\n",
+    "  shift_lr: 0.05 # Learning rate of the elemental output shifts.\n",
+    "  zbl_lr: 0.001 # \n",
+    "  transition_begin: 0 # Number of training steps (not epochs) before the start of the linear learning rate schedule.\n",
+    "\n",
+    "callbacks:\n",
+    "- name: csv # Keyword of the callback used. Currently we implement \"csv\" and \"tensorboard\".\n",
+    "\n",
+    "progress_bar:\n",
+    "  disable_epoch_pbar: false # Set to True to disable the epoch progress bar.\n",
+    "  disable_nl_pbar: false # Set to True to disable the NL precomputation progress bar.\n",
+    "\n",
+    "\n",
+    "checkpoints:\n",
+    "  ckpt_interval: 1 # Number of epochs between checkpoints.\n",
+    "  # The options below are used for transfer learning\n",
+    "  base_model_checkpoint: null # Path to the folder containing a pre-trained model ckpt.\n",
+    "  reset_layers: [] # List of layer names for which the parameters will be reinitialized.\n",
+    "\n",
+    "```"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": []
-  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -488,12 +419,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
-    "!rm -r project config.yaml error_config.yaml"
+    "# !rm -r project config.yaml error_config.yaml"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {