Upload folder using huggingface_hub

.ipynb_checkpoints/Experimenting-checkpoint.ipynb

@@ -0,0 +1,276 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "e119e71a-f88a-4d5c-90fb-e60b84c4f42c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import (\n",
+    "    AutoModelForCausalLM,\n",
+    "    AutoTokenizer,\n",
+    "    AutoTokenizer,\n",
+    ")\n",
+    "from peft import PeftModel, PeftConfig\n",
+    "import torch\n",
+    "\n",
+    "d_map = {\"\": torch.cuda.current_device()} if torch.cuda.is_available() else None\n",
+    "local_model_path = \"outputs/checkpoint-100\"     # Path to the combined weights"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "ba591ab9-5029-46e8-b9a9-428de3896e62",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ee8258a57258444baf79b52af6444788",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "ename": "OutOfMemoryError",
+     "evalue": "CUDA out of memory. Tried to allocate 112.00 MiB. GPU 0 has a total capacty of 23.68 GiB of which 79.62 MiB is free. Process 657358 has 23.59 GiB memory in use. Of the allocated memory 23.00 GiB is allocated by PyTorch, and 357.02 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mOutOfMemoryError\u001b[0m                          Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[23], line 4\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;66;03m# Loading the base Model\u001b[39;00m\n\u001b[1;32m      2\u001b[0m config \u001b[38;5;241m=\u001b[39m PeftConfig\u001b[38;5;241m.\u001b[39mfrom_pretrained(local_model_path)\n\u001b[0;32m----> 4\u001b[0m model \u001b[38;5;241m=\u001b[39m \u001b[43mAutoModelForCausalLM\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_pretrained\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m      5\u001b[0m \u001b[43m    \u001b[49m\u001b[43mconfig\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbase_model_name_or_path\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[1;32m      6\u001b[0m \u001b[43m    \u001b[49m\u001b[43mreturn_dict\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m      7\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtorch_dtype\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfloat16\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      8\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;66;43;03m# load_in_4bit=True, \u001b[39;49;00m\n\u001b[1;32m      9\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdevice_map\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43md_map\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     10\u001b[0m \u001b[43m)\u001b[49m\n\u001b[1;32m     11\u001b[0m tokenizer \u001b[38;5;241m=\u001b[39m AutoTokenizer\u001b[38;5;241m.\u001b[39mfrom_pretrained(config\u001b[38;5;241m.\u001b[39mbase_model_name_or_path)\n",
+      "File \u001b[0;32m/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py:566\u001b[0m, in \u001b[0;36m_BaseAutoModelClass.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, *model_args, **kwargs)\u001b[0m\n\u001b[1;32m    564\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28mtype\u001b[39m(config) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m_model_mapping\u001b[38;5;241m.\u001b[39mkeys():\n\u001b[1;32m    565\u001b[0m     model_class \u001b[38;5;241m=\u001b[39m _get_model_class(config, \u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m_model_mapping)\n\u001b[0;32m--> 566\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mmodel_class\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfrom_pretrained\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    567\u001b[0m \u001b[43m        \u001b[49m\u001b[43mpretrained_model_name_or_path\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mmodel_args\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconfig\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mconfig\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mhub_kwargs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\n\u001b[1;32m    568\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    569\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m    570\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mUnrecognized configuration class \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mconfig\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m for this kind of AutoModel: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m.\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    571\u001b[0m     \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mModel type should be one of \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m, \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;241m.\u001b[39mjoin(c\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;250m \u001b[39m\u001b[38;5;28;01mfor\u001b[39;00m\u001b[38;5;250m \u001b[39mc\u001b[38;5;250m \u001b[39m\u001b[38;5;129;01min\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m_model_mapping\u001b[38;5;241m.\u001b[39mkeys())\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    572\u001b[0m )\n",
+      "File \u001b[0;32m/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py:3850\u001b[0m, in \u001b[0;36mPreTrainedModel.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, config, cache_dir, ignore_mismatched_sizes, force_download, local_files_only, token, revision, use_safetensors, *model_args, **kwargs)\u001b[0m\n\u001b[1;32m   3841\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m dtype_orig \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m   3842\u001b[0m         torch\u001b[38;5;241m.\u001b[39mset_default_dtype(dtype_orig)\n\u001b[1;32m   3843\u001b[0m     (\n\u001b[1;32m   3844\u001b[0m         model,\n\u001b[1;32m   3845\u001b[0m         missing_keys,\n\u001b[1;32m   3846\u001b[0m         unexpected_keys,\n\u001b[1;32m   3847\u001b[0m         mismatched_keys,\n\u001b[1;32m   3848\u001b[0m         offload_index,\n\u001b[1;32m   3849\u001b[0m         error_msgs,\n\u001b[0;32m-> 3850\u001b[0m     ) \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_load_pretrained_model\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   3851\u001b[0m \u001b[43m        \u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3852\u001b[0m \u001b[43m        \u001b[49m\u001b[43mstate_dict\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3853\u001b[0m \u001b[43m        \u001b[49m\u001b[43mloaded_state_dict_keys\u001b[49m\u001b[43m,\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# XXX: rename?\u001b[39;49;00m\n\u001b[1;32m   3854\u001b[0m \u001b[43m        \u001b[49m\u001b[43mresolved_archive_file\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3855\u001b[0m \u001b[43m        \u001b[49m\u001b[43mpretrained_model_name_or_path\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3856\u001b[0m \u001b[43m        \u001b[49m\u001b[43mignore_mismatched_sizes\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mignore_mismatched_sizes\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3857\u001b[0m \u001b[43m        \u001b[49m\u001b[43msharded_metadata\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msharded_metadata\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3858\u001b[0m \u001b[43m        \u001b[49m\u001b[43m_fast_init\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m_fast_init\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3859\u001b[0m \u001b[43m        \u001b[49m\u001b[43mlow_cpu_mem_usage\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mlow_cpu_mem_usage\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3860\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdevice_map\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdevice_map\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3861\u001b[0m \u001b[43m        \u001b[49m\u001b[43moffload_folder\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moffload_folder\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3862\u001b[0m \u001b[43m        \u001b[49m\u001b[43moffload_state_dict\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moffload_state_dict\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3863\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdtype\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtorch_dtype\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3864\u001b[0m \u001b[43m        \u001b[49m\u001b[43mis_quantized\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mgetattr\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mquantization_method\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m==\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mQuantizationMethod\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mBITS_AND_BYTES\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3865\u001b[0m \u001b[43m        \u001b[49m\u001b[43mkeep_in_fp32_modules\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mkeep_in_fp32_modules\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3866\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3868\u001b[0m model\u001b[38;5;241m.\u001b[39mis_loaded_in_4bit \u001b[38;5;241m=\u001b[39m load_in_4bit\n\u001b[1;32m   3869\u001b[0m model\u001b[38;5;241m.\u001b[39mis_loaded_in_8bit \u001b[38;5;241m=\u001b[39m load_in_8bit\n",
+      "File \u001b[0;32m/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py:4284\u001b[0m, in \u001b[0;36mPreTrainedModel._load_pretrained_model\u001b[0;34m(cls, model, state_dict, loaded_keys, resolved_archive_file, pretrained_model_name_or_path, ignore_mismatched_sizes, sharded_metadata, _fast_init, low_cpu_mem_usage, device_map, offload_folder, offload_state_dict, dtype, is_quantized, keep_in_fp32_modules)\u001b[0m\n\u001b[1;32m   4280\u001b[0m                     set_module_quantized_tensor_to_device(\n\u001b[1;32m   4281\u001b[0m                         model_to_load, key, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcpu\u001b[39m\u001b[38;5;124m\"\u001b[39m, torch\u001b[38;5;241m.\u001b[39mempty(\u001b[38;5;241m*\u001b[39mparam\u001b[38;5;241m.\u001b[39msize(), dtype\u001b[38;5;241m=\u001b[39mdtype)\n\u001b[1;32m   4282\u001b[0m                     )\n\u001b[1;32m   4283\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4284\u001b[0m         new_error_msgs, offload_index, state_dict_index \u001b[38;5;241m=\u001b[39m \u001b[43m_load_state_dict_into_meta_model\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   4285\u001b[0m \u001b[43m            \u001b[49m\u001b[43mmodel_to_load\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4286\u001b[0m \u001b[43m            \u001b[49m\u001b[43mstate_dict\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4287\u001b[0m \u001b[43m            \u001b[49m\u001b[43mloaded_keys\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4288\u001b[0m \u001b[43m            \u001b[49m\u001b[43mstart_prefix\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4289\u001b[0m \u001b[43m            \u001b[49m\u001b[43mexpected_keys\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4290\u001b[0m \u001b[43m            \u001b[49m\u001b[43mdevice_map\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdevice_map\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4291\u001b[0m \u001b[43m            \u001b[49m\u001b[43moffload_folder\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moffload_folder\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4292\u001b[0m \u001b[43m            \u001b[49m\u001b[43moffload_index\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moffload_index\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4293\u001b[0m \u001b[43m            \u001b[49m\u001b[43mstate_dict_folder\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstate_dict_folder\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4294\u001b[0m \u001b[43m            \u001b[49m\u001b[43mstate_dict_index\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstate_dict_index\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4295\u001b[0m \u001b[43m            \u001b[49m\u001b[43mdtype\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdtype\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4296\u001b[0m \u001b[43m            \u001b[49m\u001b[43mis_quantized\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mis_quantized\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4297\u001b[0m \u001b[43m            \u001b[49m\u001b[43mis_safetensors\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mis_safetensors\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4298\u001b[0m \u001b[43m            \u001b[49m\u001b[43mkeep_in_fp32_modules\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mkeep_in_fp32_modules\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4299\u001b[0m \u001b[43m            \u001b[49m\u001b[43munexpected_keys\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43munexpected_keys\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   4300\u001b[0m \u001b[43m        \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4301\u001b[0m         error_msgs \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m new_error_msgs\n\u001b[1;32m   4302\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n",
+      "File \u001b[0;32m/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py:805\u001b[0m, in \u001b[0;36m_load_state_dict_into_meta_model\u001b[0;34m(model, state_dict, loaded_state_dict_keys, start_prefix, expected_keys, device_map, offload_folder, offload_index, state_dict_folder, state_dict_index, dtype, is_quantized, is_safetensors, keep_in_fp32_modules, unexpected_keys)\u001b[0m\n\u001b[1;32m    802\u001b[0m     state_dict_index \u001b[38;5;241m=\u001b[39m offload_weight(param, param_name, state_dict_folder, state_dict_index)\n\u001b[1;32m    803\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_quantized:\n\u001b[1;32m    804\u001b[0m     \u001b[38;5;66;03m# For backward compatibility with older versions of `accelerate`\u001b[39;00m\n\u001b[0;32m--> 805\u001b[0m     \u001b[43mset_module_tensor_to_device\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mparam_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mparam_device\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mset_module_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    806\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m param\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;129;01min\u001b[39;00m (torch\u001b[38;5;241m.\u001b[39mint8, torch\u001b[38;5;241m.\u001b[39muint8) \u001b[38;5;129;01mand\u001b[39;00m is_quantized:\n\u001b[1;32m    807\u001b[0m     \u001b[38;5;66;03m# handling newly quantized weights and loaded quantized weights\u001b[39;00m\n\u001b[1;32m    808\u001b[0m     \u001b[38;5;66;03m# edit the param.dtype restrictions and is_quantized condition when adding new quant methods\u001b[39;00m\n\u001b[1;32m    809\u001b[0m     quantized_stats \u001b[38;5;241m=\u001b[39m {}\n",
+      "File \u001b[0;32m/usr/local/lib/python3.10/dist-packages/accelerate/utils/modeling.py:347\u001b[0m, in \u001b[0;36mset_module_tensor_to_device\u001b[0;34m(module, tensor_name, device, value, dtype, fp16_statistics)\u001b[0m\n\u001b[1;32m    345\u001b[0m             module\u001b[38;5;241m.\u001b[39m_parameters[tensor_name] \u001b[38;5;241m=\u001b[39m param_cls(new_value, requires_grad\u001b[38;5;241m=\u001b[39mold_value\u001b[38;5;241m.\u001b[39mrequires_grad)\n\u001b[1;32m    346\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(value, torch\u001b[38;5;241m.\u001b[39mTensor):\n\u001b[0;32m--> 347\u001b[0m     new_value \u001b[38;5;241m=\u001b[39m \u001b[43mvalue\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mto\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdevice\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    348\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    349\u001b[0m     new_value \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mtensor(value, device\u001b[38;5;241m=\u001b[39mdevice)\n",
+      "\u001b[0;31mOutOfMemoryError\u001b[0m: CUDA out of memory. Tried to allocate 112.00 MiB. GPU 0 has a total capacty of 23.68 GiB of which 79.62 MiB is free. Process 657358 has 23.59 GiB memory in use. Of the allocated memory 23.00 GiB is allocated by PyTorch, and 357.02 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF"
+     ]
+    }
+   ],
+   "source": [
+    "# Loading the base Model\n",
+    "config = PeftConfig.from_pretrained(local_model_path)\n",
+    "\n",
+    "model = AutoModelForCausalLM.from_pretrained(\n",
+    "    config.base_model_name_or_path, \n",
+    "    return_dict=True,\n",
+    "    torch_dtype=torch.float16,\n",
+    "    # load_in_4bit=True, \n",
+    "    device_map=d_map,\n",
+    ")\n",
+    "tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "c1d36c14-0bfc-4215-8576-bb390a3a6114",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# load the base model with the Lora model\n",
+    "model = PeftModel.from_pretrained(model, local_model_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "72cdeaa8-4b0c-45d0-a585-2f86626e280b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# model.eval()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "09fa4575-0dec-4e62-a43f-77e57f68c4a9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def inferance(query: str, model, tokenizer, temp = 1.0, limit = 200) -> str:\n",
+    "  device = d_map\n",
+    "\n",
+    "  prompt_template = \"\"\"\n",
+    "  Below is an instruction that describes a task. Write a response that appropriately completes the request.\n",
+    "  ### Question:\n",
+    "  {query}\n",
+    "\n",
+    "  ### Answer:\n",
+    "  \"\"\"\n",
+    "  prompt = prompt_template.format(query=query)\n",
+    "\n",
+    "  encodeds = tokenizer(prompt, return_tensors=\"pt\", add_special_tokens=True)\n",
+    "\n",
+    "  model_inputs = encodeds.to(device)\n",
+    "\n",
+    "  generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)\n",
+    "  decoded = tokenizer.batch_decode(generated_ids)\n",
+    "  return (decoded[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "ba47700b-0787-4677-a5a1-c1a1b4063fe2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://127.0.0.1:7862\n",
+      "Running on public URL: https://f4d39c90e01dcf849c.gradio.live\n",
+      "\n",
+      "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"https://f4d39c90e01dcf849c.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": []
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Attempting to cast a BatchEncoding to type {'': 0}. This is not supported.\n",
+      "/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py:1413: UserWarning: You are calling .generate() with the `input_ids` being on a device type different than your model's device. `input_ids` is on cpu, whereas the model is on cuda. You may experience unexpected behaviors or slower generation. Please make sure that you have put `input_ids` to the correct device by calling for example input_ids = input_ids.to('cuda') before running `.generate()`.\n",
+      "  warnings.warn(\n",
+      "Traceback (most recent call last):\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/gradio/queueing.py\", line 495, in call_prediction\n",
+      "    output = await route_utils.call_process_api(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/gradio/route_utils.py\", line 230, in call_process_api\n",
+      "    output = await app.get_blocks().process_api(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/gradio/blocks.py\", line 1590, in process_api\n",
+      "    result = await self.call_function(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/gradio/blocks.py\", line 1176, in call_function\n",
+      "    prediction = await anyio.to_thread.run_sync(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py\", line 33, in run_sync\n",
+      "    return await get_async_backend().run_sync_in_worker_thread(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py\", line 2106, in run_sync_in_worker_thread\n",
+      "    return await future\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py\", line 833, in run\n",
+      "    result = context.run(func, *args)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/gradio/utils.py\", line 678, in wrapper\n",
+      "    response = f(*args, **kwargs)\n",
+      "  File \"/tmp/ipykernel_812/3792119410.py\", line 5, in predict\n",
+      "    out = inferance(prompt, model, tokenizer, temp = 1.0, limit = 200)\n",
+      "  File \"/tmp/ipykernel_812/3078938966.py\", line 17, in inferance\n",
+      "    generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/peft/peft_model.py\", line 1140, in generate\n",
+      "    outputs = self.base_model.generate(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py\", line 115, in decorate_context\n",
+      "    return func(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py\", line 1525, in generate\n",
+      "    return self.sample(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py\", line 2622, in sample\n",
+      "    outputs = self(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\", line 1518, in _wrapped_call_impl\n",
+      "    return self._call_impl(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\", line 1527, in _call_impl\n",
+      "    return forward_call(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/transformers/models/mistral/modeling_mistral.py\", line 1154, in forward\n",
+      "    outputs = self.model(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\", line 1518, in _wrapped_call_impl\n",
+      "    return self._call_impl(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\", line 1527, in _call_impl\n",
+      "    return forward_call(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/transformers/models/mistral/modeling_mistral.py\", line 984, in forward\n",
+      "    inputs_embeds = self.embed_tokens(input_ids)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\", line 1518, in _wrapped_call_impl\n",
+      "    return self._call_impl(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py\", line 1527, in _call_impl\n",
+      "    return forward_call(*args, **kwargs)\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/modules/sparse.py\", line 162, in forward\n",
+      "    return F.embedding(\n",
+      "  File \"/usr/local/lib/python3.10/dist-packages/torch/nn/functional.py\", line 2233, in embedding\n",
+      "    return torch.embedding(weight, input, padding_idx, scale_grad_by_freq, sparse)\n",
+      "RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument index in method wrapper_CUDA__index_select)\n"
+     ]
+    }
+   ],
+   "source": [
+    "import gradio as gr\n",
+    "\n",
+    "def predict(temp, limit, text):\n",
+    "    prompt = text\n",
+    "    out = inferance(prompt, model, tokenizer, temp = 1.0, limit = 200)\n",
+    "    return out\n",
+    "\n",
+    "pred = gr.Interface(\n",
+    "    predict,\n",
+    "    inputs=[\n",
+    "        gr.Slider(0.001, 10, value=0.1, label=\"Temperature\"),\n",
+    "        gr.Slider(1, 1024, value=128, label=\"Token Limit\"),\n",
+    "        gr.Textbox(\n",
+    "            label=\"Input\",\n",
+    "            lines=1,\n",
+    "            value=\"#### Human: What's the capital of Australia?#### Assistant: \",\n",
+    "        ),\n",
+    "    ],\n",
+    "    outputs='text',\n",
+    ")\n",
+    "\n",
+    "pred.launch(share=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "62622184-9d02-4bd0-8c4e-d6775ce20f75",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "###some factors to try\n",
+    "from_pt=True"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,6 @@
+---
+title: finance
+app_file: app.py
+sdk: gradio
+sdk_version: 4.17.0
+---

.ipynb_checkpoints/app-checkpoint.py

@@ -0,0 +1,100 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+from peft import PeftModel, PeftConfig
+import torch
+import gradio as gr
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+local_model_path = "outputs/checkpoint-100"     # Path to the combined weights
+
+# Loading the base Model
+config = PeftConfig.from_pretrained(local_model_path)
+
+model = AutoModelForCausalLM.from_pretrained(
+    config.base_model_name_or_path, 
+    return_dict=True,
+    torch_dtype=torch.float16,
+    device_map=d_map,
+)
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path, trust_remote_code=True)
+
+# load the base model with the Lora model
+mergedModel = PeftModel.from_pretrained(model, local_model_path)
+# model = model.merge_and_unload()
+mergedModel.eval()
+
+
+
+def inferance(query: str, model, tokenizer, temp = 1.0, limit = 200) -> str:
+  device = "cuda:0"
+
+  prompt_template = """
+  Below is an instruction that describes a task. Write a response that appropriately completes the request.
+  ### Question:
+  {query}
+
+  ### Answer:
+  """
+  prompt = prompt_template.format(query=query)
+
+  encodeds = tokenizer(prompt, return_tensors="pt", add_special_tokens=True)
+
+  model_inputs = encodeds.to(device)
+
+  generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)
+  decoded = tokenizer.batch_decode(generated_ids)
+  return (decoded[0])
+
+
+
+def predict(temp, limit, text):
+    prompt = text
+    out = inferance(prompt, mergedModel, tokenizer, temp = 1.0, limit = 200)
+    return out
+
+pred = gr.Interface(
+    predict,
+    inputs=[
+        gr.Slider(0.001, 10, value=0.1, label="Temperature"),
+        gr.Slider(1, 1024, value=128, label="Token Limit"),
+        gr.Textbox(
+            label="Input",
+            lines=1,
+            value="#### Human: What's the capital of Australia?#### Assistant: ",
+        ),
+    ],
+    outputs='text',
+)
+
+pred.launch(share=True)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

.ipynb_checkpoints/baseInference-checkpoint.py

@@ -0,0 +1,66 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+from peft import PeftModel, PeftConfig
+import torch
+import gradio as gr
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+local_model_path = "outputs/checkpoint-100"     # Path to the combined weights
+
+# Loading the base Model
+config = PeftConfig.from_pretrained(local_model_path)
+
+model = AutoModelForCausalLM.from_pretrained(
+    config.base_model_name_or_path, 
+    return_dict=True,
+    torch_dtype=torch.float16,
+    device_map=d_map,
+)
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path, trust_remote_code=True)
+
+
+def inferance(query: str, model, tokenizer, temp = 1.0, limit = 200) -> str:
+  device = "cuda:0"
+
+  prompt_template = """
+  Below is an instruction that describes a task. Write a response that appropriately completes the request.
+  ### Question:
+  {query}
+
+  ### Answer:
+  """
+  prompt = prompt_template.format(query=query)
+
+  encodeds = tokenizer(prompt, return_tensors="pt", add_special_tokens=True)
+
+  model_inputs = encodeds.to(device)
+
+  generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)
+  decoded = tokenizer.batch_decode(generated_ids)
+  return (decoded[0])
+
+
+
+def predict(temp, limit, text):
+    prompt = text
+    out = inferance(prompt, model, tokenizer, temp = 1.0, limit = 200)
+    return out
+
+pred = gr.Interface(
+    predict,
+    inputs=[
+        gr.Slider(0.001, 10, value=0.1, label="Temperature"),
+        gr.Slider(1, 1024, value=128, label="Token Limit"),
+        gr.Textbox(
+            label="Input",
+            lines=1,
+            value="#### Human: What's the capital of Australia?#### Assistant: ",
+        ),
+    ],
+    outputs='text',
+)
+
+pred.launch(share=True)

.ipynb_checkpoints/finetune-checkpoint.py

@@ -0,0 +1,147 @@
+# -*- coding: utf-8 -*-
+"""PreFinetuningForRunPod.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/1LtsUCcWfL2VpWLJXVkE5076XX5k3PTyg
+"""
+
+# IMPORTS
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
+import datasets
+from datasets import load_dataset
+from trl import SFTTrainer
+from peft import PeftConfig, PeftModel
+from multiprocessing import cpu_count
+from peft import prepare_model_for_kbit_training
+from peft import LoraConfig, get_peft_model
+import bitsandbytes as bnb
+import transformers
+
+# LOGIN TO HUB FOR MODEL DEPLOYMENT
+# from huggingface_hub import notebook_login
+# notebook_login()
+
+# LOADING THE TOKENIZER
+model_id = "mistralai/Mistral-7B-v0.1"
+print("-----------------------------loading tokenizer-----------------------------------------------------------")
+tokenizer = AutoTokenizer.from_pretrained(model_id, add_eos_token=True)
+
+
+print("-----------------------------loading dataset-----------------------------------------------------------")
+# LOAD DATA FROM HUGGINFACE
+data = load_dataset("gbharti/finance-alpaca", split='train')
+
+# PREPARE DATA FOR FINE-TUNING
+def generate_prompt(data_point):
+    """Gen. input text based on a prompt, task instruction, (context info.), and answer
+
+    :param data_point: dict: Data point
+    :return: dict: tokenzed prompt
+    """
+    # Samples with additional context into.
+    if data_point['input']:
+        text = 'Below is an instruction that describes a task, paired with an input that provides' \
+               ' further context. Write a response that appropriately completes the request.\n\n'
+        text += f'### Instruction:\n{data_point["instruction"]}\n\n'
+        text += f'### Input:\n{data_point["input"]}\n\n'
+        text += f'### Response:\n{data_point["output"]}'
+
+    # Without context
+    else:
+        text = 'Below is an instruction that describes a task. Write a response that ' \
+               'appropriately completes the request.\n\n'
+        text += f'### Instruction:\n{data_point["instruction"]}\n\n'
+        text += f'### Response:\n{data_point["output"]}'
+    return text
+
+print("-----------------------------Preparing dataset for fine-tuning-----------------------------------------------------------")
+prompt = [generate_prompt(data_point) for data_point in data]
+data = data.add_column("prompt", prompt);
+data = data.map(lambda sample: tokenizer(sample["prompt"]),num_proc=cpu_count(), batched=True)
+# data = data.remove_columns(['Context', 'Response'])
+data = data.shuffle(seed=1234)
+data = data.train_test_split(test_size=0.1)
+train_data = data["train"]
+test_data = data["test"]
+
+
+# LOADING MODEL IN N(4, 8.....) BIT
+bnb_config = BitsAndBytesConfig(
+    load_in_4bit = True,
+    bnb_4bit_use_double_quant=True,
+    bnb_4bit_quant_type="nf4",
+    bnb_4bit_compute_dtype=torch.bfloat16
+)
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+
+print("-----------------------------loading model-----------------------------------------------------------")
+model = AutoModelForCausalLM.from_pretrained(
+model_id,
+ torch_dtype="auto",
+  use_cache=False, # set to False as we're going to use gradient checkpointing
+  quantization_config=bnb_config,
+  device_map=d_map
+)
+
+
+def find_all_linear_names(model):
+  cls = bnb.nn.Linear4bit #if args.bits == 4 else (bnb.nn.Linear8bitLt if args.bits == 8 else torch.nn.Linear)
+  lora_module_names = set()
+  for name, module in model.named_modules():
+    if isinstance(module, cls):
+      names = name.split('.')
+      lora_module_names.add(names[0] if len(names) == 1 else names[-1])
+    if 'lm_head' in lora_module_names: # needed for 16-bit
+      lora_module_names.remove('lm_head')
+  return list(lora_module_names)
+
+modules = find_all_linear_names(model)
+
+
+lora_config = LoraConfig(
+    r=8,                                   # Number of quantization levels
+    lora_alpha=32,                         # Hyperparameter for LoRA
+    target_modules = modules, # Modules to apply LoRA to
+    lora_dropout=0.05,                     # Dropout probability
+    bias="none",                           # Type of bias
+    task_type="CAUSAL_LM"                  # Task type (in this case, Causal Language Modeling)
+)
+model.gradient_checkpointing_enable()
+model = prepare_model_for_kbit_training(model)
+model = get_peft_model(model, lora_config)
+
+
+# trainable, total = model.get_nb_trainable_parameters()
+# print(f"Trainable: {trainable} | total: {total} | Percentage: {trainable/total*100:.4f}%")
+
+tokenizer.pad_token = tokenizer.eos_token
+torch.cuda.empty_cache()
+
+trainer = SFTTrainer(
+    model=model,
+    train_dataset=train_data,
+    eval_dataset=test_data,
+    dataset_text_field="prompt",
+    peft_config=lora_config,
+    args=transformers.TrainingArguments(
+        per_device_train_batch_size=1,
+        gradient_accumulation_steps=4,
+        gradient_checkpointing=True,
+        fp16=True,
+        warmup_steps=0.03,
+        max_steps=10,
+        learning_rate=2e-4,
+        logging_steps=1,
+        output_dir="outputs",
+        optim="paged_adamw_8bit",
+        save_strategy="epoch",
+    ),
+    data_collator=transformers.DataCollatorForLanguageModeling(tokenizer, mlm=False),
+)
+print("-----------------------------finetuning starts------------------------------------------------------------")
+model.config.use_cache = False
+trainer.train()

.ipynb_checkpoints/mergePeftWithBaseModel-checkpoint.py

@@ -0,0 +1,33 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+from peft import PeftModel, PeftConfig
+import torch
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+local_model_path = "outputs/checkpoint-100"     # Path to the combined weights
+
+# Loading the base Model
+config = PeftConfig.from_pretrained(local_model_path)
+
+model = AutoModelForCausalLM.from_pretrained(
+    config.base_model_name_or_path, 
+    return_dict=True, 
+    # load_in_4bit=True, 
+    device_map=d_map,
+    ignore_mismatched_sizes=True,
+    # from_tf=True,  
+)
+
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
+
+# load the base model with the Lora model
+model = PeftModel.from_pretrained(model, local_model_path)
+
+merged = model.merge_and_unload()
+
+merged.save_pretrained("outputs/merged")
+
+tokenizer.save_pretrained("outputs/merged")

.ipynb_checkpoints/requiretments-checkpoint.txt

@@ -0,0 +1,8 @@
+bitsandbytes
+transformers
+peft
+accelerate
+datasets
+trl
+flash-attn
+gradio

.ipynb_checkpoints/uploadToHub-checkpoint.py

@@ -0,0 +1,42 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+
+import torch
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+merged_model_path = "outputs/merged"     # Path to the combined weights
+repo_name = "Financial_Analyst"            # HuggingFace repo name
+
+
+model = AutoModelForCausalLM.from_pretrained(
+    merged_model_path,
+    ignore_mismatched_sizes=True,
+    from_tf=True,
+    trust_remote_code=True, 
+    device_map=d_map, 
+    torch_dtype=torch.float16,
+).eval()
+
+tokenizer = AutoTokenizer.from_pretrained(merged_model_path)
+
+
+model.push_to_hub(repo_name, token=hf_token)
+tokenizer.push_to_hub(repo_name, token=hf_token)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

Experimenting.ipynb

@@ -0,0 +1,659 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "e119e71a-f88a-4d5c-90fb-e60b84c4f42c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import (\n",
+    "    AutoModelForCausalLM,\n",
+    "    AutoTokenizer,\n",
+    "    AutoTokenizer,\n",
+    ")\n",
+    "from peft import PeftModel, PeftConfig\n",
+    "import torch\n",
+    "import gradio as gr\n",
+    "\n",
+    "d_map = {\"\": torch.cuda.current_device()} if torch.cuda.is_available() else None\n",
+    "local_model_path = \"outputs/checkpoint-100\"     # Path to the combined weights"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "ba591ab9-5029-46e8-b9a9-428de3896e62",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "913fc55a00aa4a2799729c6461060d28",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "config.json:   0%|          | 0.00/571 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4bf4ee1d7d0446d7a9c5775bdd823d3b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model.safetensors.index.json:   0%|          | 0.00/25.1k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4d512679e32f4ca4ae7aad1f821fda4d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "665dd2a779b94e6fba7c5fdc6313f504",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00001-of-00002.safetensors:   0%|          | 0.00/9.94G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9c1efb636ab24518b58b78bb2fa90cf7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00002-of-00002.safetensors:   0%|          | 0.00/4.54G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e502f8a721354757a12a34879ceba413",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "80a0725ee12c4659b5271308553d2b23",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "generation_config.json:   0%|          | 0.00/116 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3d1e7af016a64550949738a7dfb0daed",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer_config.json:   0%|          | 0.00/967 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d85c88ad430748b48026e227e2b58b8d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer.model:   0%|          | 0.00/493k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "196bda4d38fc4d198a26154b06089754",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer.json:   0%|          | 0.00/1.80M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "692a1a2768724c40aa90d6cec836d16a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "special_tokens_map.json:   0%|          | 0.00/72.0 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Loading the base Model\n",
+    "config = PeftConfig.from_pretrained(local_model_path)\n",
+    "\n",
+    "model = AutoModelForCausalLM.from_pretrained(\n",
+    "    config.base_model_name_or_path, \n",
+    "    return_dict=True,\n",
+    "    torch_dtype=torch.float16,\n",
+    "    device_map=d_map,\n",
+    "    trust_remote_code=True\n",
+    ")\n",
+    "tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path, trust_remote_code=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "c1d36c14-0bfc-4215-8576-bb390a3a6114",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "PeftModelForCausalLM(\n",
+       "  (base_model): LoraModel(\n",
+       "    (model): MistralForCausalLM(\n",
+       "      (model): MistralModel(\n",
+       "        (embed_tokens): Embedding(32000, 4096)\n",
+       "        (layers): ModuleList(\n",
+       "          (0-31): 32 x MistralDecoderLayer(\n",
+       "            (self_attn): MistralAttention(\n",
+       "              (q_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=4096, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=4096, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (k_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=4096, out_features=1024, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=4096, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=1024, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (v_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=4096, out_features=1024, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=4096, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=1024, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (o_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=4096, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=4096, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (rotary_emb): MistralRotaryEmbedding()\n",
+       "            )\n",
+       "            (mlp): MistralMLP(\n",
+       "              (gate_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=4096, out_features=14336, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=4096, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=14336, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (up_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=4096, out_features=14336, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=4096, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=14336, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (down_proj): lora.Linear(\n",
+       "                (base_layer): Linear(in_features=14336, out_features=4096, bias=False)\n",
+       "                (lora_dropout): ModuleDict(\n",
+       "                  (default): Dropout(p=0.05, inplace=False)\n",
+       "                )\n",
+       "                (lora_A): ModuleDict(\n",
+       "                  (default): Linear(in_features=14336, out_features=8, bias=False)\n",
+       "                )\n",
+       "                (lora_B): ModuleDict(\n",
+       "                  (default): Linear(in_features=8, out_features=4096, bias=False)\n",
+       "                )\n",
+       "                (lora_embedding_A): ParameterDict()\n",
+       "                (lora_embedding_B): ParameterDict()\n",
+       "              )\n",
+       "              (act_fn): SiLU()\n",
+       "            )\n",
+       "            (input_layernorm): MistralRMSNorm()\n",
+       "            (post_attention_layernorm): MistralRMSNorm()\n",
+       "          )\n",
+       "        )\n",
+       "        (norm): MistralRMSNorm()\n",
+       "      )\n",
+       "      (lm_head): Linear(in_features=4096, out_features=32000, bias=False)\n",
+       "    )\n",
+       "  )\n",
+       ")"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# load the base model with the Lora model\n",
+    "mergedModel = PeftModel.from_pretrained(model, local_model_path)\n",
+    "# model = model.merge_and_unload()\n",
+    "mergedModel.eval()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "09fa4575-0dec-4e62-a43f-77e57f68c4a9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def inferance(query: str, model, tokenizer, temp = 1.0, limit = 200) -> str:\n",
+    "  device = \"cuda:0\"\n",
+    "\n",
+    "  prompt_template = \"\"\"\n",
+    "  Below is an instruction that describes a task. Write a response that appropriately completes the request.\n",
+    "  ### Question:\n",
+    "  {query}\n",
+    "\n",
+    "  ### Answer:\n",
+    "  \"\"\"\n",
+    "  prompt = prompt_template.format(query=query)\n",
+    "\n",
+    "  encodeds = tokenizer(prompt, return_tensors=\"pt\", add_special_tokens=True)\n",
+    "\n",
+    "  model_inputs = encodeds.to(device)\n",
+    "\n",
+    "  generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)\n",
+    "  decoded = tokenizer.batch_decode(generated_ids)\n",
+    "  return (decoded[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "ba47700b-0787-4677-a5a1-c1a1b4063fe2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://127.0.0.1:7864\n",
+      "Running on public URL: https://b9a01680128206a745.gradio.live\n",
+      "\n",
+      "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"https://b9a01680128206a745.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": []
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "\n",
+    "def predict(temp, limit, text):\n",
+    "    prompt = text\n",
+    "    out = inferance(prompt, mergedModel, tokenizer, temp = 1.0, limit = 200)\n",
+    "    return out\n",
+    "\n",
+    "pred = gr.Interface(\n",
+    "    predict,\n",
+    "    inputs=[\n",
+    "        gr.Slider(0.001, 10, value=0.1, label=\"Temperature\"),\n",
+    "        gr.Slider(1, 1024, value=128, label=\"Token Limit\"),\n",
+    "        gr.Textbox(\n",
+    "            label=\"Input\",\n",
+    "            lines=1,\n",
+    "            value=\"#### Human: What's the capital of Australia?#### Assistant: \",\n",
+    "        ),\n",
+    "    ],\n",
+    "    outputs='text',\n",
+    ")\n",
+    "\n",
+    "pred.launch(share=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "8e0529f7-5e0d-4822-b826-ca49ca10c172",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
+      "To disable this warning, you can either:\n",
+      "\t- Avoid using `tokenizers` before the fork if possible\n",
+      "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Need \u001b[32m'write'\u001b[0m access token to create a Spaces repo.\n",
+      "\n",
+      "    _|    _|  _|    _|    _|_|_|    _|_|_|  _|_|_|  _|      _|    _|_|_|      _|_|_|_|    _|_|      _|_|_|  _|_|_|_|\n",
+      "    _|    _|  _|    _|  _|        _|          _|    _|_|    _|  _|            _|        _|    _|  _|        _|\n",
+      "    _|_|_|_|  _|    _|  _|  _|_|  _|  _|_|    _|    _|  _|  _|  _|  _|_|      _|_|_|    _|_|_|_|  _|        _|_|_|\n",
+      "    _|    _|  _|    _|  _|    _|  _|    _|    _|    _|    _|_|  _|    _|      _|        _|    _|  _|        _|\n",
+      "    _|    _|    _|_|      _|_|_|    _|_|_|  _|_|_|  _|      _|    _|_|_|      _|        _|    _|    _|_|_|  _|_|_|_|\n",
+      "\n",
+      "    To login, `huggingface_hub` requires a token generated from https://huggingface.co/settings/tokens .\n",
+      "Token: \n",
+      "\u001b[31mAborted.\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "!gradio deploy"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "1eecb055-93d7-485c-bb85-c025c06cd94c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import datasets"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "9e25a85a-5276-47ff-996d-ddb426eafac4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from datasets import load_dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "b7bece25-22f4-4489-a5e8-10a1c28aeae8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "441a29ce1fd14a259d8538497f16798a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Downloading readme:   0%|          | 0.00/709 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "57c38be620174f12b1c13f1794c29b83",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Downloading data:   0%|          | 0.00/42.9M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d955c0a3620242978f507e27ddebf41b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Generating train split: 0 examples [00:00, ? examples/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "data = load_dataset(\"gbharti/finance-alpaca\", split='train')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "7a2b2ac8-f6ce-40ff-b785-d79e8e673aa6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'output': \"This is definitely a scam. I had a friend sign up for a very similar offer and what they did was send a fake check and then asked to transfer the same amount to them. So now you just send them a couple grand and you're holding a fake check.\",\n",
+       " 'text': '',\n",
+       " 'instruction': 'What risks are there acting as a broker between PayPal and electronic bank transfers?',\n",
+       " 'input': ''}"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data[1021]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "c20eb320-5c64-4118-82a9-0cd65319bd57",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://127.0.0.1:7863\n",
+      "Running on public URL: https://f9ab2457c317a5c384.gradio.live\n",
+      "\n",
+      "This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"https://f9ab2457c317a5c384.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": []
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import gradio as gr\n",
+    "\n",
+    "def predict(temp, limit, text):\n",
+    "    prompt = text\n",
+    "    out = inferance(prompt, model, tokenizer, temp = 1.0, limit = 200)\n",
+    "    return out\n",
+    "\n",
+    "pred = gr.Interface(\n",
+    "    predict,\n",
+    "    inputs=[\n",
+    "        gr.Slider(0.001, 10, value=0.1, label=\"Temperature\"),\n",
+    "        gr.Slider(1, 1024, value=128, label=\"Token Limit\"),\n",
+    "        gr.Textbox(\n",
+    "            label=\"Input\",\n",
+    "            lines=1,\n",
+    "            value=\"#### Human: What's the capital of Australia?#### Assistant: \",\n",
+    "        ),\n",
+    "    ],\n",
+    "    outputs='text',\n",
+    ")\n",
+    "\n",
+    "pred.launch(share=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "6cac6e2f-d6eb-4311-95da-082bd92145c5",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

ImportCommands.txt

@@ -0,0 +1 @@
+pip install -r requirements.txt

README.md

@@ -1,12 +1,6 @@
 ---
-title: Finance
-emoji: 🌖
-colorFrom: gray
-colorTo: green
+title: finance
+app_file: app.py
 sdk: gradio
 sdk_version: 4.17.0
-app_file: app.py
-pinned: false
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,100 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+from peft import PeftModel, PeftConfig
+import torch
+import gradio as gr
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+local_model_path = "outputs/checkpoint-100"     # Path to the combined weights
+
+# Loading the base Model
+config = PeftConfig.from_pretrained(local_model_path)
+
+model = AutoModelForCausalLM.from_pretrained(
+    config.base_model_name_or_path, 
+    return_dict=True,
+    torch_dtype=torch.float16,
+    device_map=d_map,
+)
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path, trust_remote_code=True)
+
+# load the base model with the Lora model
+mergedModel = PeftModel.from_pretrained(model, local_model_path)
+# model = model.merge_and_unload()
+mergedModel.eval()
+
+
+
+def inferance(query: str, model, tokenizer, temp = 1.0, limit = 200) -> str:
+  device = "cuda:0"
+
+  prompt_template = """
+  Below is an instruction that describes a task. Write a response that appropriately completes the request.
+  ### Question:
+  {query}
+
+  ### Answer:
+  """
+  prompt = prompt_template.format(query=query)
+
+  encodeds = tokenizer(prompt, return_tensors="pt", add_special_tokens=True)
+
+  model_inputs = encodeds.to(device)
+
+  generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)
+  decoded = tokenizer.batch_decode(generated_ids)
+  return (decoded[0])
+
+
+
+def predict(temp, limit, text):
+    prompt = text
+    out = inferance(prompt, mergedModel, tokenizer, temp = 1.0, limit = 200)
+    return out
+
+pred = gr.Interface(
+    predict,
+    inputs=[
+        gr.Slider(0.001, 10, value=0.1, label="Temperature"),
+        gr.Slider(1, 1024, value=128, label="Token Limit"),
+        gr.Textbox(
+            label="Input",
+            lines=1,
+            value="#### Human: What's the capital of Australia?#### Assistant: ",
+        ),
+    ],
+    outputs='text',
+)
+
+pred.launch(share=True)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

baseInference.py

@@ -0,0 +1,66 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+from peft import PeftModel, PeftConfig
+import torch
+import gradio as gr
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+local_model_path = "outputs/checkpoint-100"     # Path to the combined weights
+
+# Loading the base Model
+config = PeftConfig.from_pretrained(local_model_path)
+
+model = AutoModelForCausalLM.from_pretrained(
+    config.base_model_name_or_path, 
+    return_dict=True,
+    torch_dtype=torch.float16,
+    device_map=d_map,
+)
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path, trust_remote_code=True)
+
+
+def inferance(query: str, model, tokenizer, temp = 1.0, limit = 200) -> str:
+  device = "cuda:0"
+
+  prompt_template = """
+  Below is an instruction that describes a task. Write a response that appropriately completes the request.
+  ### Question:
+  {query}
+
+  ### Answer:
+  """
+  prompt = prompt_template.format(query=query)
+
+  encodeds = tokenizer(prompt, return_tensors="pt", add_special_tokens=True)
+
+  model_inputs = encodeds.to(device)
+
+  generated_ids = model.generate(**model_inputs, max_new_tokens=int(limit), temperature=temp, do_sample=True, pad_token_id=tokenizer.eos_token_id)
+  decoded = tokenizer.batch_decode(generated_ids)
+  return (decoded[0])
+
+
+
+def predict(temp, limit, text):
+    prompt = text
+    out = inferance(prompt, model, tokenizer, temp = 1.0, limit = 200)
+    return out
+
+pred = gr.Interface(
+    predict,
+    inputs=[
+        gr.Slider(0.001, 10, value=0.1, label="Temperature"),
+        gr.Slider(1, 1024, value=128, label="Token Limit"),
+        gr.Textbox(
+            label="Input",
+            lines=1,
+            value="#### Human: What's the capital of Australia?#### Assistant: ",
+        ),
+    ],
+    outputs='text',
+)
+
+pred.launch(share=True)

finetune.py

@@ -0,0 +1,147 @@
+# -*- coding: utf-8 -*-
+"""PreFinetuningForRunPod.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/1LtsUCcWfL2VpWLJXVkE5076XX5k3PTyg
+"""
+
+# IMPORTS
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
+import datasets
+from datasets import load_dataset
+from trl import SFTTrainer
+from peft import PeftConfig, PeftModel
+from multiprocessing import cpu_count
+from peft import prepare_model_for_kbit_training
+from peft import LoraConfig, get_peft_model
+import bitsandbytes as bnb
+import transformers
+
+# LOGIN TO HUB FOR MODEL DEPLOYMENT
+# from huggingface_hub import notebook_login
+# notebook_login()
+
+# LOADING THE TOKENIZER
+model_id = "mistralai/Mistral-7B-v0.1"
+print("-----------------------------loading tokenizer-----------------------------------------------------------")
+tokenizer = AutoTokenizer.from_pretrained(model_id, add_eos_token=True)
+
+
+print("-----------------------------loading dataset-----------------------------------------------------------")
+# LOAD DATA FROM HUGGINFACE
+data = load_dataset("gbharti/finance-alpaca", split='train')
+
+# PREPARE DATA FOR FINE-TUNING
+def generate_prompt(data_point):
+    """Gen. input text based on a prompt, task instruction, (context info.), and answer
+
+    :param data_point: dict: Data point
+    :return: dict: tokenzed prompt
+    """
+    # Samples with additional context into.
+    if data_point['input']:
+        text = 'Below is an instruction that describes a task, paired with an input that provides' \
+               ' further context. Write a response that appropriately completes the request.\n\n'
+        text += f'### Instruction:\n{data_point["instruction"]}\n\n'
+        text += f'### Input:\n{data_point["input"]}\n\n'
+        text += f'### Response:\n{data_point["output"]}'
+
+    # Without context
+    else:
+        text = 'Below is an instruction that describes a task. Write a response that ' \
+               'appropriately completes the request.\n\n'
+        text += f'### Instruction:\n{data_point["instruction"]}\n\n'
+        text += f'### Response:\n{data_point["output"]}'
+    return text
+
+print("-----------------------------Preparing dataset for fine-tuning-----------------------------------------------------------")
+prompt = [generate_prompt(data_point) for data_point in data]
+data = data.add_column("prompt", prompt);
+data = data.map(lambda sample: tokenizer(sample["prompt"]),num_proc=cpu_count(), batched=True)
+# data = data.remove_columns(['Context', 'Response'])
+data = data.shuffle(seed=1234)
+data = data.train_test_split(test_size=0.1)
+train_data = data["train"]
+test_data = data["test"]
+
+
+# LOADING MODEL IN N(4, 8.....) BIT
+bnb_config = BitsAndBytesConfig(
+    load_in_4bit = True,
+    bnb_4bit_use_double_quant=True,
+    bnb_4bit_quant_type="nf4",
+    bnb_4bit_compute_dtype=torch.bfloat16
+)
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+
+print("-----------------------------loading model-----------------------------------------------------------")
+model = AutoModelForCausalLM.from_pretrained(
+model_id,
+ torch_dtype="auto",
+  use_cache=False, # set to False as we're going to use gradient checkpointing
+  quantization_config=bnb_config,
+  device_map=d_map
+)
+
+
+def find_all_linear_names(model):
+  cls = bnb.nn.Linear4bit #if args.bits == 4 else (bnb.nn.Linear8bitLt if args.bits == 8 else torch.nn.Linear)
+  lora_module_names = set()
+  for name, module in model.named_modules():
+    if isinstance(module, cls):
+      names = name.split('.')
+      lora_module_names.add(names[0] if len(names) == 1 else names[-1])
+    if 'lm_head' in lora_module_names: # needed for 16-bit
+      lora_module_names.remove('lm_head')
+  return list(lora_module_names)
+
+modules = find_all_linear_names(model)
+
+
+lora_config = LoraConfig(
+    r=8,                                   # Number of quantization levels
+    lora_alpha=32,                         # Hyperparameter for LoRA
+    target_modules = modules, # Modules to apply LoRA to
+    lora_dropout=0.05,                     # Dropout probability
+    bias="none",                           # Type of bias
+    task_type="CAUSAL_LM"                  # Task type (in this case, Causal Language Modeling)
+)
+model.gradient_checkpointing_enable()
+model = prepare_model_for_kbit_training(model)
+model = get_peft_model(model, lora_config)
+
+
+# trainable, total = model.get_nb_trainable_parameters()
+# print(f"Trainable: {trainable} | total: {total} | Percentage: {trainable/total*100:.4f}%")
+
+tokenizer.pad_token = tokenizer.eos_token
+torch.cuda.empty_cache()
+
+trainer = SFTTrainer(
+    model=model,
+    train_dataset=train_data,
+    eval_dataset=test_data,
+    dataset_text_field="prompt",
+    peft_config=lora_config,
+    args=transformers.TrainingArguments(
+        per_device_train_batch_size=1,
+        gradient_accumulation_steps=4,
+        gradient_checkpointing=True,
+        fp16=True,
+        warmup_steps=0.03,
+        max_steps=10,
+        learning_rate=2e-4,
+        logging_steps=1,
+        output_dir="outputs",
+        optim="paged_adamw_8bit",
+        save_strategy="epoch",
+    ),
+    data_collator=transformers.DataCollatorForLanguageModeling(tokenizer, mlm=False),
+)
+print("-----------------------------finetuning starts------------------------------------------------------------")
+model.config.use_cache = False
+trainer.train()

flagged/log.csv

@@ -0,0 +1,2 @@
+Temperature,Token Limit,Input,output,flag,username,timestamp
+0.001,1,hello,,,,2024-02-08 01:33:00.412258

mergePeftWithBaseModel.py

@@ -0,0 +1,33 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+from peft import PeftModel, PeftConfig
+import torch
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+local_model_path = "outputs/checkpoint-100"     # Path to the combined weights
+
+# Loading the base Model
+config = PeftConfig.from_pretrained(local_model_path)
+
+model = AutoModelForCausalLM.from_pretrained(
+    config.base_model_name_or_path, 
+    return_dict=True, 
+    # load_in_4bit=True, 
+    device_map=d_map,
+    ignore_mismatched_sizes=True,
+    # from_tf=True,  
+)
+
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
+
+# load the base model with the Lora model
+model = PeftModel.from_pretrained(model, local_model_path)
+
+merged = model.merge_and_unload()
+
+merged.save_pretrained("outputs/merged")
+
+tokenizer.save_pretrained("outputs/merged")

outputs/checkpoint-10/README.md

@@ -0,0 +1,204 @@
+---
+library_name: peft
+base_model: mistralai/Mistral-7B-v0.1
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure 
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+
+
+### Framework versions
+
+- PEFT 0.8.2

outputs/checkpoint-10/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "gate_proj",
+    "q_proj",
+    "o_proj",
+    "down_proj",
+    "v_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_rslora": false
+}

outputs/checkpoint-10/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5cf8302b03366081bd1d38bb1cb1879be97e37561f3d27948bad2acd0d923e4
+size 83945296

outputs/checkpoint-10/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6d55b6511f1729b6376635f4163abd016e8d7c205f15ae4b1daab9e8d8e77d4
+size 42545748

outputs/checkpoint-10/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:466f8edfa57a71fc1c5033963e7ade8afb45dd92f2ae2bb543284593e1998663
+size 14244

outputs/checkpoint-10/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92895d149359090ddf3ae4a97578a9224915364b8880d14de783a868db38bfc8
+size 1064

outputs/checkpoint-10/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

outputs/checkpoint-10/tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

outputs/checkpoint-10/trainer_state.json

@@ -0,0 +1,81 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.0006449532408900354,
+  "eval_steps": 500,
+  "global_step": 10,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00018054162487462388,
+      "loss": 2.5461,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.000160481444332999,
+      "loss": 2.1187,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00014042126379137414,
+      "loss": 2.1673,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00012036108324974924,
+      "loss": 1.8731,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00010030090270812438,
+      "loss": 1.4151,
+      "step": 5
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.02407221664995e-05,
+      "loss": 1.5411,
+      "step": 6
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.018054162487462e-05,
+      "loss": 1.4401,
+      "step": 7
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.012036108324975e-05,
+      "loss": 0.85,
+      "step": 8
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.0060180541624874e-05,
+      "loss": 1.0114,
+      "step": 9
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0,
+      "loss": 1.8774,
+      "step": 10
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 10,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 336712855117824.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

outputs/checkpoint-10/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c51fe40f7d5d576653d66e0860715cd1f38681cfe0dc755470225987a86fec53
+size 4664

outputs/checkpoint-100/README.md

@@ -0,0 +1,204 @@
+---
+library_name: peft
+base_model: mistralai/Mistral-7B-v0.1
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure 
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+
+
+### Framework versions
+
+- PEFT 0.8.2

outputs/checkpoint-100/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "q_proj",
+    "k_proj",
+    "o_proj",
+    "gate_proj",
+    "down_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_rslora": false
+}

outputs/checkpoint-100/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d4c76655486ba0755cda4b98e8cf44c51ac55c316a16716adfe72a893f33f59
+size 83945296

outputs/checkpoint-100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3fe5cd332d7c70e0ea8516a38e95aef229624258fb58b07d39e5793024acdc2
+size 42545748

outputs/checkpoint-100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a2da9732d59f29842d306635d89f785d09764a48e4c3ddf83b669fcba436f2e
+size 14244

outputs/checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06de337c477ec1811259baac2b8aa88593bc99d405bbc7f63f504e99114af4cf
+size 1064

outputs/checkpoint-100/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

outputs/checkpoint-100/tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

outputs/checkpoint-100/trainer_state.json

@@ -0,0 +1,621 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.0064495324089003546,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00019805941782534764,
+      "loss": 2.1283,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00019605881764529358,
+      "loss": 1.9699,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00019405821746523957,
+      "loss": 1.4391,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00019205761728518557,
+      "loss": 1.175,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00019005701710513156,
+      "loss": 0.9018,
+      "step": 5
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00018805641692507753,
+      "loss": 1.074,
+      "step": 6
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00018605581674502352,
+      "loss": 1.5826,
+      "step": 7
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00018405521656496952,
+      "loss": 0.9327,
+      "step": 8
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00018205461638491548,
+      "loss": 1.258,
+      "step": 9
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00018005401620486148,
+      "loss": 1.0259,
+      "step": 10
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00017805341602480744,
+      "loss": 0.788,
+      "step": 11
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00017605281584475344,
+      "loss": 1.3443,
+      "step": 12
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00017405221566469943,
+      "loss": 1.207,
+      "step": 13
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00017205161548464542,
+      "loss": 0.7044,
+      "step": 14
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00017005101530459136,
+      "loss": 1.0694,
+      "step": 15
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00016805041512453736,
+      "loss": 1.1858,
+      "step": 16
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00016604981494448335,
+      "loss": 1.2622,
+      "step": 17
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00016404921476442935,
+      "loss": 0.7919,
+      "step": 18
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0001620486145843753,
+      "loss": 1.0567,
+      "step": 19
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0001600480144043213,
+      "loss": 1.2917,
+      "step": 20
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0001580474142242673,
+      "loss": 0.9172,
+      "step": 21
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00015604681404421327,
+      "loss": 1.3542,
+      "step": 22
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00015404621386415926,
+      "loss": 1.4117,
+      "step": 23
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00015204561368410523,
+      "loss": 0.6285,
+      "step": 24
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00015004501350405122,
+      "loss": 1.4488,
+      "step": 25
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00014804441332399721,
+      "loss": 0.618,
+      "step": 26
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0001460438131439432,
+      "loss": 1.3617,
+      "step": 27
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00014404321296388918,
+      "loss": 1.1482,
+      "step": 28
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00014204261278383514,
+      "loss": 1.0907,
+      "step": 29
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00014004201260378114,
+      "loss": 1.0434,
+      "step": 30
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00013804141242372713,
+      "loss": 1.3318,
+      "step": 31
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00013604081224367312,
+      "loss": 1.109,
+      "step": 32
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0001340402120636191,
+      "loss": 1.0989,
+      "step": 33
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00013203961188356508,
+      "loss": 1.2508,
+      "step": 34
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00013003901170351108,
+      "loss": 1.3718,
+      "step": 35
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00012803841152345704,
+      "loss": 0.8635,
+      "step": 36
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.000126037811343403,
+      "loss": 1.0131,
+      "step": 37
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.000124037211163349,
+      "loss": 1.0554,
+      "step": 38
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.000122036610983295,
+      "loss": 0.8635,
+      "step": 39
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00012003601080324098,
+      "loss": 0.7278,
+      "step": 40
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00011803541062318697,
+      "loss": 1.5206,
+      "step": 41
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00011603481044313295,
+      "loss": 1.0913,
+      "step": 42
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00011403421026307892,
+      "loss": 0.7845,
+      "step": 43
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00011203361008302491,
+      "loss": 0.8502,
+      "step": 44
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.0001100330099029709,
+      "loss": 1.8437,
+      "step": 45
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00010803240972291689,
+      "loss": 1.6084,
+      "step": 46
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00010603180954286287,
+      "loss": 1.1675,
+      "step": 47
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00010403120936280886,
+      "loss": 1.1029,
+      "step": 48
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00010203060918275482,
+      "loss": 1.151,
+      "step": 49
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 0.00010003000900270081,
+      "loss": 1.061,
+      "step": 50
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.802940882264679e-05,
+      "loss": 1.536,
+      "step": 51
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.602880864259278e-05,
+      "loss": 0.616,
+      "step": 52
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.402820846253876e-05,
+      "loss": 0.8911,
+      "step": 53
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.202760828248476e-05,
+      "loss": 1.3183,
+      "step": 54
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.002700810243074e-05,
+      "loss": 1.186,
+      "step": 55
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.802640792237672e-05,
+      "loss": 1.301,
+      "step": 56
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.802640792237672e-05,
+      "loss": 0.9267,
+      "step": 57
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.602580774232271e-05,
+      "loss": 1.6136,
+      "step": 58
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.402520756226868e-05,
+      "loss": 0.6503,
+      "step": 59
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.202460738221467e-05,
+      "loss": 1.1043,
+      "step": 60
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.002400720216065e-05,
+      "loss": 1.098,
+      "step": 61
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.802340702210663e-05,
+      "loss": 0.8298,
+      "step": 62
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.602280684205261e-05,
+      "loss": 1.3,
+      "step": 63
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.402220666199861e-05,
+      "loss": 1.2021,
+      "step": 64
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.202160648194459e-05,
+      "loss": 1.2454,
+      "step": 65
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.002100630189057e-05,
+      "loss": 1.2646,
+      "step": 66
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.802040612183656e-05,
+      "loss": 1.2226,
+      "step": 67
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.601980594178254e-05,
+      "loss": 1.2412,
+      "step": 68
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.401920576172852e-05,
+      "loss": 1.4806,
+      "step": 69
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.20186055816745e-05,
+      "loss": 0.908,
+      "step": 70
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.001800540162049e-05,
+      "loss": 1.1458,
+      "step": 71
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.801740522156648e-05,
+      "loss": 1.2067,
+      "step": 72
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.601680504151246e-05,
+      "loss": 1.1893,
+      "step": 73
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.4016204861458444e-05,
+      "loss": 1.1854,
+      "step": 74
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.201560468140443e-05,
+      "loss": 0.9617,
+      "step": 75
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.0015004501350405e-05,
+      "loss": 1.0145,
+      "step": 76
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.801440432129639e-05,
+      "loss": 1.3273,
+      "step": 77
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.601380414124238e-05,
+      "loss": 0.841,
+      "step": 78
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.401320396118836e-05,
+      "loss": 1.0926,
+      "step": 79
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.201260378113434e-05,
+      "loss": 0.9073,
+      "step": 80
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0012003601080326e-05,
+      "loss": 0.6963,
+      "step": 81
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.801140342102631e-05,
+      "loss": 1.023,
+      "step": 82
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6010803240972294e-05,
+      "loss": 1.079,
+      "step": 83
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.401020306091828e-05,
+      "loss": 1.0577,
+      "step": 84
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.200960288086426e-05,
+      "loss": 0.8747,
+      "step": 85
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0009002700810245e-05,
+      "loss": 1.0794,
+      "step": 86
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.800840252075623e-05,
+      "loss": 0.8603,
+      "step": 87
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6007802340702216e-05,
+      "loss": 1.4064,
+      "step": 88
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4007202160648196e-05,
+      "loss": 1.2233,
+      "step": 89
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.200660198059418e-05,
+      "loss": 1.6015,
+      "step": 90
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0006001800540163e-05,
+      "loss": 1.1858,
+      "step": 91
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8005401620486147e-05,
+      "loss": 0.676,
+      "step": 92
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.600480144043213e-05,
+      "loss": 1.4585,
+      "step": 93
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4004201260378114e-05,
+      "loss": 1.3953,
+      "step": 94
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2003601080324098e-05,
+      "loss": 0.9336,
+      "step": 95
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.0003000900270082e-05,
+      "loss": 0.8244,
+      "step": 96
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 8.002400720216065e-06,
+      "loss": 0.8921,
+      "step": 97
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.001800540162049e-06,
+      "loss": 1.1946,
+      "step": 98
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.001200360108033e-06,
+      "loss": 1.3317,
+      "step": 99
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0006001800540163e-06,
+      "loss": 0.6798,
+      "step": 100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 100,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 2762380453404672.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

outputs/checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73450ff8bcd630c54efa079eb5886af4c4249e60576ac187998637f94ce85529
+size 4664

outputs/merged/.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,38 @@
+{
+  "_name_or_path": "mistralai/Mistral-7B-v0.1",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "quantization_config": {
+    "bnb_4bit_compute_dtype": "float32",
+    "bnb_4bit_quant_type": "fp4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.37.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

outputs/merged/config.json

@@ -0,0 +1,38 @@
+{
+  "_name_or_path": "mistralai/Mistral-7B-v0.1",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "quantization_config": {
+    "bnb_4bit_compute_dtype": "float32",
+    "bnb_4bit_quant_type": "fp4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.37.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

outputs/merged/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.37.2"
+}

outputs/merged/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e5252c0ef4cf3bf6aa7defa8f1f2afbf7c743a859ba1aec2e0a7d846ba5578e
+size 4450837496

outputs/merged/special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

outputs/merged/tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

requiretments.txt

@@ -0,0 +1,8 @@
+bitsandbytes
+transformers
+peft
+accelerate
+datasets
+trl
+flash-attn
+gradio

uploadToHub.py

@@ -0,0 +1,42 @@
+from transformers import (
+    AutoModelForCausalLM,
+    AutoTokenizer,
+    AutoTokenizer,
+)
+
+import torch
+
+d_map = {"": torch.cuda.current_device()} if torch.cuda.is_available() else None
+merged_model_path = "outputs/merged"     # Path to the combined weights
+repo_name = "Financial_Analyst"            # HuggingFace repo name
+
+
+model = AutoModelForCausalLM.from_pretrained(
+    merged_model_path,
+    ignore_mismatched_sizes=True,
+    from_tf=True,
+    trust_remote_code=True, 
+    device_map=d_map, 
+    torch_dtype=torch.float16,
+).eval()
+
+tokenizer = AutoTokenizer.from_pretrained(merged_model_path)
+
+
+model.push_to_hub(repo_name, token=hf_token)
+tokenizer.push_to_hub(repo_name, token=hf_token)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+