6d1f84dc2a6eebd464cd01c33bf56eb90d04ebf2ed9dd4ccb7c2bc2e8a0a0f68

trainer-v4-unit-test/model-init.ipynb

@@ -3,13 +3,13 @@
   {
    "attachments": {},
    "cell_type": "markdown",
-   "id": "52b890f4",
+   "id": "ee8cfec1",
    "metadata": {
     "papermill": {
-     "duration": 0.002262,
-     "end_time": "2023-08-23T10:53:59.090685",
+     "duration": 0.001903,
+     "end_time": "2023-08-23T19:20:40.353029",
      "exception": false,
-     "start_time": "2023-08-23T10:53:59.088423",
+     "start_time": "2023-08-23T19:20:40.351126",
      "status": "completed"
     },
     "tags": []
@@ -27,16 +27,16 @@
   {
    "attachments": {},
    "cell_type": "markdown",
-   "id": "f89c248f",
+   "id": "7713f8b0",
    "metadata": {
     "notebookRunGroups": {
      "groupValue": ""
     },
     "papermill": {
-     "duration": 0.001598,
-     "end_time": "2023-08-23T10:53:59.094424",
+     "duration": 0.001217,
+     "end_time": "2023-08-23T19:20:40.355733",
      "exception": false,
-     "start_time": "2023-08-23T10:53:59.092826",
+     "start_time": "2023-08-23T19:20:40.354516",
      "status": "completed"
     },
     "tags": []
@@ -48,19 +48,19 @@
   {
    "cell_type": "code",
    "execution_count": 1,
-   "id": "65ffd8d5",
+   "id": "54684bc2",
    "metadata": {
     "execution": {
-     "iopub.execute_input": "2023-08-23T10:53:59.099519Z",
-     "iopub.status.busy": "2023-08-23T10:53:59.099000Z",
-     "iopub.status.idle": "2023-08-23T10:53:59.849771Z",
-     "shell.execute_reply": "2023-08-23T10:53:59.848760Z"
+     "iopub.execute_input": "2023-08-23T19:20:40.360445Z",
+     "iopub.status.busy": "2023-08-23T19:20:40.359491Z",
+     "iopub.status.idle": "2023-08-23T19:20:41.112004Z",
+     "shell.execute_reply": "2023-08-23T19:20:41.111011Z"
     },
     "papermill": {
-     "duration": 0.756019,
-     "end_time": "2023-08-23T10:53:59.852126",
+     "duration": 0.757092,
+     "end_time": "2023-08-23T19:20:41.114108",
      "exception": false,
-     "start_time": "2023-08-23T10:53:59.096107",
+     "start_time": "2023-08-23T19:20:40.357016",
      "status": "completed"
     },
     "tags": []
@@ -76,19 +76,19 @@
   {
    "cell_type": "code",
    "execution_count": 2,
-   "id": "c3b4c1b4",
+   "id": "80da2afe",
    "metadata": {
     "execution": {
-     "iopub.execute_input": "2023-08-23T10:53:59.858552Z",
-     "iopub.status.busy": "2023-08-23T10:53:59.858038Z",
-     "iopub.status.idle": "2023-08-23T10:54:08.846245Z",
-     "shell.execute_reply": "2023-08-23T10:54:08.844779Z"
+     "iopub.execute_input": "2023-08-23T19:20:41.119086Z",
+     "iopub.status.busy": "2023-08-23T19:20:41.118584Z",
+     "iopub.status.idle": "2023-08-23T19:20:50.840612Z",
+     "shell.execute_reply": "2023-08-23T19:20:50.839437Z"
     },
     "papermill": {
-     "duration": 8.994708,
-     "end_time": "2023-08-23T10:54:08.848911",
+     "duration": 9.727624,
+     "end_time": "2023-08-23T19:20:50.843488",
      "exception": false,
-     "start_time": "2023-08-23T10:53:59.854203",
+     "start_time": "2023-08-23T19:20:41.115864",
      "status": "completed"
     },
     "tags": []
@@ -98,14 +98,20 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[2023-08-23 10:54:04,147] [INFO] [real_accelerator.py:133:get_accelerator] Setting ds_accelerator to cuda (auto detect)\r\n"
+      "[2023-08-23 19:20:45,786] [INFO] [real_accelerator.py:133:get_accelerator] Setting ds_accelerator to cuda (auto detect)\r\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[RWKV.model] Running RWKV model using 'torch-jit' with torch '2.0.1+cu118'\r\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[RWKV.model] Running RWKV model using 'torch-jit' with torch '2.0.1+cu118'\r\n",
       "---- Initializing model ----\r\n",
       "No of layers: 6\r\n",
       "Embedding size: 512\r\n",
@@ -129,22 +135,22 @@
      "output_type": "stream",
      "text": [
       "Building extension module wkv_1_bf16...\r\n",
-      "Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)\r\n",
-      "ninja: no work to do.\r\n",
-      "Loading extension module wkv_1_bf16...\r\n"
+      "Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)\r\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[RWKV.model]: Finished initial model load\r\n"
+      "ninja: no work to do.\r\n",
+      "Loading extension module wkv_1_bf16...\r\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      "[RWKV.model]: Finished initial model load\r\n",
       "50277 512   -0.0001 emb.weight\r\n"
      ]
     },
@@ -207,15 +213,15 @@
      "text": [
       "512   512   0    blocks.3.att.receptance.weight\r\n",
       "512   512   0    blocks.3.att.output.weight\r\n",
-      "2048  512   1.0  blocks.3.ffn.key.weight\r\n"
+      "2048  512   1.0  blocks.3.ffn.key.weight\r\n",
+      "512   512   0    blocks.3.ffn.receptance.weight\r\n",
+      "512   2048  0    blocks.3.ffn.value.weight\r\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "512   512   0    blocks.3.ffn.receptance.weight\r\n",
-      "512   2048  0    blocks.3.ffn.value.weight\r\n",
       "512   512   0    blocks.4.att.key.weight\r\n",
       "512   512   1.0  blocks.4.att.value.weight\r\n",
       "512   512   0    blocks.4.att.receptance.weight\r\n",
@@ -230,23 +236,17 @@
       "512   512   0    blocks.4.ffn.receptance.weight\r\n",
       "512   2048  0    blocks.4.ffn.value.weight\r\n",
       "512   512   0    blocks.5.att.key.weight\r\n",
-      "512   512   1.0  blocks.5.att.value.weight\r\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
+      "512   512   1.0  blocks.5.att.value.weight\r\n",
       "512   512   0    blocks.5.att.receptance.weight\r\n",
       "512   512   0    blocks.5.att.output.weight\r\n",
-      "2048  512   1.0  blocks.5.ffn.key.weight\r\n",
-      "512   512   0    blocks.5.ffn.receptance.weight\r\n"
+      "2048  512   1.0  blocks.5.ffn.key.weight\r\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      "512   512   0    blocks.5.ffn.receptance.weight\r\n",
       "512   2048  0    blocks.5.ffn.value.weight\r\n",
       "50277 512   0.5  head.weight\r\n"
      ]
@@ -278,14 +278,14 @@
   },
   "papermill": {
    "default_parameters": {},
-   "duration": 11.521627,
-   "end_time": "2023-08-23T10:54:09.274570",
+   "duration": 12.198584,
+   "end_time": "2023-08-23T19:20:51.170107",
    "environment_variables": {},
    "exception": null,
    "input_path": "/actions-runner/_work/RWKV-infctx-trainer/RWKV-infctx-trainer/notebook/trainer-v4-unit-test/model-init.ipynb",
    "output_path": "/actions-runner/_work/RWKV-infctx-trainer/RWKV-infctx-trainer/output/trainer-v4-unit-test/model-init.ipynb",
    "parameters": {},
-   "start_time": "2023-08-23T10:53:57.752943",
+   "start_time": "2023-08-23T19:20:38.971523",
    "version": "2.4.0"
   }
  },