diff --git a/examples/conditional_generation/multitask_prompt_tuning.ipynb b/examples/conditional_generation/multitask_prompt_tuning.ipynb
index 96fd2180dd..1eaec5b016 100644
--- a/examples/conditional_generation/multitask_prompt_tuning.ipynb
+++ b/examples/conditional_generation/multitask_prompt_tuning.ipynb
@@ -9,12 +9,13 @@
    },
    "outputs": [],
    "source": [
+    "import torch\n",
     "from datasets import load_dataset\n",
     "from transformers import set_seed, AutoModelForSeq2SeqLM, AutoTokenizer\n",
     "from peft import get_peft_model, MultitaskPromptTuningConfig, TaskType, MultitaskPromptTuningInit\n",
     "\n",
     "set_seed(42)\n",
-    "\n",
+    "device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
     "model_name = \"google/flan-t5-base\"\n",
     "\n",
     "peft_config = MultitaskPromptTuningConfig(\n",
@@ -31,18 +32,18 @@
     "model = AutoModelForSeq2SeqLM.from_pretrained(model_name)\n",
     "model = get_peft_model(model, peft_config)\n",
     "\n",
-    "model = model.cuda()\n",
+    "model = model.to(device)\n",
     "\n",
     "\n",
     "def send_to_device(batch):\n",
     "    for i in batch:\n",
-    "        batch[i] = batch[i].cuda()\n",
+    "        batch[i] = batch[i].to(device)\n",
     "    return batch"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 9,
    "id": "eb112bc1-ffaf-49fa-a216-0d601ec304ee",
    "metadata": {
     "tags": []
@@ -86,7 +87,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 10,
    "id": "e5a16ec4-8fef-4ba9-95b6-a661eb51e50c",
    "metadata": {
     "tags": []
@@ -159,7 +160,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 11,
    "id": "cceecc94-f43a-4f62-8d45-926f2f02f36d",
    "metadata": {
     "tags": []
@@ -293,7 +294,7 @@
     "    num_tasks=1,\n",
     "    task_type=TaskType.SEQ_2_SEQ_LM,\n",
     "    prompt_tuning_init=MultitaskPromptTuningInit.EXACT_SOURCE_TASK,\n",
-    "    prompt_tuning_init_state_dict_path=\"checkpoints_source/50000/adapter_model.bin\",\n",
+    "    prompt_tuning_init_state_dict_path=\"checkpoints_source/50000/adapter_model.safetensors\",\n",
     "    num_virtual_tokens=50,\n",
     "    num_transformer_submodules=1,\n",
     ")\n",
@@ -302,7 +303,7 @@
     "model = AutoModelForSeq2SeqLM.from_pretrained(model_name)\n",
     "model = get_peft_model(model, peft_config)\n",
     "\n",
-    "model = model.cuda()"
+    "model = model.to(device)"
    ]
   },
   {
@@ -360,8 +361,9 @@
    "source": [
     "# load last checkpoint for now\n",
     "from peft import set_peft_model_state_dict\n",
+    "from safetensors.torch import load_file\n",
     "\n",
-    "sd_6000 = torch.load(\"checkpoints_target/6000/adapter_model.bin\")\n",
+    "sd_6000 = load_file(\"checkpoints_target/6000/adapter_model.safetensors\")\n",
     "set_peft_model_state_dict(model, sd_6000)\n",
     "\n",
     "# evaluate val\n",
@@ -382,6 +384,22 @@
     "f1 = {f1}\"\"\"\n",
     ")"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1d18325c-9607-4cb5-a5b0-5b44dfee2a75",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "43988e92-af42-45cb-8bca-f19c193ad04f",
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
@@ -400,7 +418,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.13"
+   "version": "3.11.13"
   }
  },
  "nbformat": 4,
diff --git a/examples/conditional_generation/peft_adalora_seq2seq.py b/examples/conditional_generation/peft_adalora_seq2seq.py
index eac19f3d55..32f532226b 100644
--- a/examples/conditional_generation/peft_adalora_seq2seq.py
+++ b/examples/conditional_generation/peft_adalora_seq2seq.py
@@ -11,7 +11,7 @@
 
 os.environ["TOKENIZERS_PARALLELISM"] = "false"
 
-device = "cuda"
+device = torch.accelerator.current_accelerator().type if hasattr(torch, "accelerator") else "cuda"
 model_name_or_path = "facebook/bart-base"
 tokenizer_name_or_path = "facebook/bart-base"
 
@@ -24,6 +24,20 @@
 batch_size = 8
 
 
+# loading dataset
+dataset = load_dataset("financial_phrasebank", "sentences_allagree")
+dataset = dataset["train"].train_test_split(test_size=0.1)
+dataset["validation"] = dataset["test"]
+del dataset["test"]
+
+classes = dataset["train"].features["label"].names
+dataset = dataset.map(
+    lambda x: {"text_label": [classes[label] for label in x["label"]]},
+    batched=True,
+    num_proc=1,
+)
+
+
 # creating model
 peft_config = AdaLoraConfig(
     init_r=12,
@@ -37,6 +51,7 @@
     lora_dropout=0.1,
     task_type=TaskType.SEQ_2_SEQ_LM,
     inference_mode=False,
+    total_step=len(dataset["train"]) * num_epochs,
 )
 
 model = AutoModelForSeq2SeqLM.from_pretrained(model_name_or_path)
@@ -44,20 +59,6 @@
 model.print_trainable_parameters()
 
 
-# loading dataset
-dataset = load_dataset("financial_phrasebank", "sentences_allagree")
-dataset = dataset["train"].train_test_split(test_size=0.1)
-dataset["validation"] = dataset["test"]
-del dataset["test"]
-
-classes = dataset["train"].features["label"].names
-dataset = dataset.map(
-    lambda x: {"text_label": [classes[label] for label in x["label"]]},
-    batched=True,
-    num_proc=1,
-)
-
-
 # data preprocessing
 tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
 
@@ -159,7 +160,7 @@ def preprocess_function(examples):
 model.save_pretrained(peft_model_id)
 
 
-ckpt = f"{peft_model_id}/adapter_model.bin"
+ckpt = f"{peft_model_id}/adapter_model.safetensors"
 # get_ipython().system('du -h $ckpt')
 
 
diff --git a/examples/conditional_generation/peft_ia3_seq2seq.ipynb b/examples/conditional_generation/peft_ia3_seq2seq.ipynb
index 0bd57f89ba..155fab5530 100644
--- a/examples/conditional_generation/peft_ia3_seq2seq.ipynb
+++ b/examples/conditional_generation/peft_ia3_seq2seq.ipynb
@@ -2,7 +2,8 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": null,
+   "id": "0c152fc8",
    "metadata": {
     "id": "5f93b7d1"
    },
@@ -22,7 +23,7 @@
     "from tqdm import tqdm\n",
     "from datasets import load_dataset\n",
     "\n",
-    "device = \"cuda\"\n",
+    "device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
     "model_name_or_path = \"bigscience/mt0-large\"\n",
     "tokenizer_name_or_path = \"bigscience/mt0-large\"\n",
     "\n",
@@ -37,7 +38,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 2,
+   "id": "4e23624f",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -49,10 +51,10 @@
     {
      "data": {
       "text/plain": [
-       "<module 'peft' from '/usr/local/lib/python3.10/dist-packages/peft/__init__.py'>"
+       "<module 'peft' from '/usr/local/lib/python3.11/dist-packages/peft/__init__.py'>"
       ]
      },
-     "execution_count": 13,
+     "execution_count": 2,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -65,7 +67,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": null,
+   "id": "da74b569",
    "metadata": {
     "id": "8d0850ac"
    },
@@ -79,7 +82,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 4,
+   "id": "df33fce2",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -233,7 +237,7 @@
        ")"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -244,7 +248,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 5,
+   "id": "63d7bc2d",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -257,7 +262,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "trainable params: 282,624 || all params: 1,229,863,936 || trainable%: 0.022980103060766553\n"
+      "trainable params: 282,624 || all params: 1,229,863,936 || trainable%: 0.0230\n"
      ]
     },
     {
@@ -276,11 +281,11 @@
        "                (SelfAttention): MT5Attention(\n",
        "                  (q): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                  (k): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (v): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (o): Linear(in_features=1024, out_features=1024, bias=False)\n",
@@ -293,7 +298,7 @@
        "                (DenseReluDense): MT5DenseGatedActDense(\n",
        "                  (wi_0): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                  (wi_1): Linear(\n",
-       "                    in_features=1024, out_features=2816, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 2816x1])\n",
        "                  )\n",
        "                  (wo): Linear(in_features=2816, out_features=1024, bias=False)\n",
@@ -311,11 +316,11 @@
        "                (SelfAttention): MT5Attention(\n",
        "                  (q): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                  (k): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (v): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (o): Linear(in_features=1024, out_features=1024, bias=False)\n",
@@ -327,7 +332,7 @@
        "                (DenseReluDense): MT5DenseGatedActDense(\n",
        "                  (wi_0): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                  (wi_1): Linear(\n",
-       "                    in_features=1024, out_features=2816, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 2816x1])\n",
        "                  )\n",
        "                  (wo): Linear(in_features=2816, out_features=1024, bias=False)\n",
@@ -352,11 +357,11 @@
        "                (SelfAttention): MT5Attention(\n",
        "                  (q): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                  (k): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (v): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (o): Linear(in_features=1024, out_features=1024, bias=False)\n",
@@ -369,11 +374,11 @@
        "                (EncDecAttention): MT5Attention(\n",
        "                  (q): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                  (k): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (v): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (o): Linear(in_features=1024, out_features=1024, bias=False)\n",
@@ -385,7 +390,7 @@
        "                (DenseReluDense): MT5DenseGatedActDense(\n",
        "                  (wi_0): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                  (wi_1): Linear(\n",
-       "                    in_features=1024, out_features=2816, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 2816x1])\n",
        "                  )\n",
        "                  (wo): Linear(in_features=2816, out_features=1024, bias=False)\n",
@@ -403,11 +408,11 @@
        "                (SelfAttention): MT5Attention(\n",
        "                  (q): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                  (k): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (v): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (o): Linear(in_features=1024, out_features=1024, bias=False)\n",
@@ -419,11 +424,11 @@
        "                (EncDecAttention): MT5Attention(\n",
        "                  (q): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                  (k): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (v): Linear(\n",
-       "                    in_features=1024, out_features=1024, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=1024, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 1024x1])\n",
        "                  )\n",
        "                  (o): Linear(in_features=1024, out_features=1024, bias=False)\n",
@@ -435,7 +440,7 @@
        "                (DenseReluDense): MT5DenseGatedActDense(\n",
        "                  (wi_0): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                  (wi_1): Linear(\n",
-       "                    in_features=1024, out_features=2816, bias=False\n",
+       "                    (base_layer): Linear(in_features=1024, out_features=2816, bias=False)\n",
        "                    (ia3_l): ParameterDict(  (default): Parameter containing: [torch.FloatTensor of size 2816x1])\n",
        "                  )\n",
        "                  (wo): Linear(in_features=2816, out_features=1024, bias=False)\n",
@@ -457,7 +462,7 @@
        ")"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -470,7 +475,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 6,
+   "id": "155b8728",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
@@ -519,27 +525,14 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "WARNING:datasets.builder:Found cached dataset financial_phrasebank (/root/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141)\n"
+      "Using the latest cached version of the dataset since financial_phrasebank couldn't be found on the Hugging Face Hub\n",
+      "Found the latest cached dataset configuration 'sentences_allagree' at /root/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141 (last modified on Thu Jul 31 03:15:41 2025).\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "bbfb7533b5ca459194e171df56b79566",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9e12d97af6124a5a8c6627708b300c1e",
+       "model_id": "43b03e9b6de94bf0921228482d7be1e5",
        "version_major": 2,
        "version_minor": 0
       },
@@ -553,7 +546,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "0c561dab67914ea9b6e1aab803600551",
+       "model_id": "d08de1efca67472781017b806f33870c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -567,12 +560,12 @@
     {
      "data": {
       "text/plain": [
-       "{'sentence': 'It will be operated by Nokia , and supported by its Nokia NetAct network and service management system .',\n",
+       "{'sentence': 'SCOPI Chief Business Excellence Officer , Eng .',\n",
        " 'label': 1,\n",
        " 'text_label': 'neutral'}"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -596,7 +589,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 7,
+   "id": "723fb67d",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/",
@@ -633,7 +627,63 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e1e80a68a9e7429397cafc96c3c11f80",
+       "model_id": "7e08a312e5454c188f52fc2ca902c463",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer_config.json:   0%|          | 0.00/430 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "25d5de12709748c9959cd011c5c641de",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "spiece.model:   0%|          | 0.00/4.31M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5b39c130813843c18e7f9187ffec37df",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer.json:   0%|          | 0.00/16.3M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "de27076e123243fd89dbad1c9e1f0596",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "special_tokens_map.json:   0%|          | 0.00/74.0 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1b55669bf13a4e2886f34c12d5f50354",
        "version_major": 2,
        "version_minor": 0
       },
@@ -647,7 +697,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "21f582e1208a4a38ae3c0cdce87e5c14",
+       "model_id": "f914229f180b4188925d9e804b92475c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -695,7 +745,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 8,
+   "id": "36d56ea7",
    "metadata": {
     "id": "f733a3c6"
    },
@@ -712,7 +763,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 9,
+   "id": "6b0a0536",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -725,45 +777,45 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████| 255/255 [02:33<00:00,  1.67it/s]\n",
-      "100%|██████████| 29/29 [00:08<00:00,  3.48it/s]\n"
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:52<00:00,  4.86it/s]\n",
+      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 12.67it/s]\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "epoch=0: train_ppl=tensor(1.4939, device='cuda:0') train_epoch_loss=tensor(0.4014, device='cuda:0') eval_ppl=tensor(1.0514, device='cuda:0') eval_epoch_loss=tensor(0.0501, device='cuda:0')\n"
+      "epoch=0: train_ppl=tensor(1.4686, device='xpu:0') train_epoch_loss=tensor(0.3843, device='xpu:0') eval_ppl=tensor(1.0421, device='xpu:0') eval_epoch_loss=tensor(0.0412, device='xpu:0')\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████| 255/255 [02:32<00:00,  1.67it/s]\n",
-      "100%|██████████| 29/29 [00:08<00:00,  3.43it/s]\n"
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:49<00:00,  5.20it/s]\n",
+      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 13.62it/s]\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "epoch=1: train_ppl=tensor(1.0523, device='cuda:0') train_epoch_loss=tensor(0.0510, device='cuda:0') eval_ppl=tensor(1.0383, device='cuda:0') eval_epoch_loss=tensor(0.0376, device='cuda:0')\n"
+      "epoch=1: train_ppl=tensor(1.0683, device='xpu:0') train_epoch_loss=tensor(0.0661, device='xpu:0') eval_ppl=tensor(1.0264, device='xpu:0') eval_epoch_loss=tensor(0.0261, device='xpu:0')\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████| 255/255 [02:32<00:00,  1.68it/s]\n",
-      "100%|██████████| 29/29 [00:08<00:00,  3.44it/s]"
+      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:49<00:00,  5.20it/s]\n",
+      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 13.63it/s]"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "epoch=2: train_ppl=tensor(1.0397, device='cuda:0') train_epoch_loss=tensor(0.0389, device='cuda:0') eval_ppl=tensor(1.0392, device='cuda:0') eval_epoch_loss=tensor(0.0385, device='cuda:0')\n"
+      "epoch=2: train_ppl=tensor(1.0451, device='xpu:0') train_epoch_loss=tensor(0.0441, device='xpu:0') eval_ppl=tensor(1.0191, device='xpu:0') eval_epoch_loss=tensor(0.0190, device='xpu:0')\n"
      ]
     },
     {
@@ -814,6 +866,7 @@
   {
    "cell_type": "code",
    "execution_count": 21,
+   "id": "761b90e4",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -849,6 +902,7 @@
   {
    "cell_type": "code",
    "execution_count": 22,
+   "id": "8e0658ac",
    "metadata": {
     "id": "a8de6005"
    },
@@ -861,7 +915,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": null,
+   "id": "ef7fbf9c",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -874,18 +929,19 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "1.2M\tbigscience/mt0-large_IA3_SEQ_2_SEQ_LM/adapter_model.bin\n"
+      "1.2M\tbigscience/mt0-large_IA3_SEQ_2_SEQ_LM/adapter_model.safetensors\n"
      ]
     }
    ],
    "source": [
-    "ckpt = f\"{peft_model_id}/adapter_model.bin\"\n",
+    "ckpt = f\"{peft_model_id}/adapter_model.safetensors\"\n",
     "!du -h $ckpt"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 24,
+   "id": "4774d931",
    "metadata": {
     "id": "76c2fc29"
    },
@@ -903,6 +959,7 @@
   {
    "cell_type": "code",
    "execution_count": 25,
+   "id": "996ddf0a",
    "metadata": {
     "colab": {
      "base_uri": "https://localhost:8080/"
@@ -946,6 +1003,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "id": "701eda1b",
    "metadata": {
     "id": "66c65ea4"
    },
@@ -955,6 +1013,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "id": "7d7718c5",
    "metadata": {
     "id": "65e71f78"
    },
@@ -970,7 +1029,7 @@
    "provenance": []
   },
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -984,7 +1043,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.3"
+   "version": "3.11.13"
   },
   "vscode": {
    "interpreter": {
diff --git a/examples/conditional_generation/peft_lora_seq2seq.ipynb b/examples/conditional_generation/peft_lora_seq2seq.ipynb
index 6cbd4f1cb9..55a3f8aae2 100644
--- a/examples/conditional_generation/peft_lora_seq2seq.ipynb
+++ b/examples/conditional_generation/peft_lora_seq2seq.ipynb
@@ -2,26 +2,10 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "5f93b7d1",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "===================================BUG REPORT===================================\n",
-      "Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues\n",
-      "For effortless bug reporting copy-paste your error into this form: https://docs.google.com/forms/d/e/1FAIpQLScPB8emS3Thkp66nvqwmjTEgxp8Y9ufuWTzFyr9kJ5AoI47dQ/viewform?usp=sf_link\n",
-      "================================================================================\n",
-      "CUDA SETUP: CUDA runtime path found: /home/sourab/miniconda3/envs/ml/lib/libcudart.so\n",
-      "CUDA SETUP: Highest compute capability among GPUs detected: 7.5\n",
-      "CUDA SETUP: Detected CUDA version 117\n",
-      "CUDA SETUP: Loading binary /home/sourab/miniconda3/envs/ml/lib/python3.10/site-packages/bitsandbytes/libbitsandbytes_cuda117.so...\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from transformers import AutoModelForSeq2SeqLM\n",
     "from peft import get_peft_config, get_peft_model, get_peft_model_state_dict, LoraConfig, TaskType\n",
@@ -36,7 +20,7 @@
     "from tqdm import tqdm\n",
     "from datasets import load_dataset\n",
     "\n",
-    "device = \"cuda\"\n",
+    "device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
     "model_name_or_path = \"bigscience/mt0-large\"\n",
     "tokenizer_name_or_path = \"bigscience/mt0-large\"\n",
     "\n",
@@ -51,7 +35,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "id": "8d0850ac",
    "metadata": {},
    "outputs": [],
@@ -75,18 +59,19 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Found cached dataset financial_phrasebank (/home/sourab/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141)\n"
+      "Using the latest cached version of the dataset since financial_phrasebank couldn't be found on the Hugging Face Hub\n",
+      "Found the latest cached dataset configuration 'sentences_allagree' at /root/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141 (last modified on Thu Jul 31 05:47:32 2025).\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3403bf3d718042018b0531848cc30209",
+       "model_id": "867f7bbb679d4b6eae344812fb797c19",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?it/s]"
+       "Map:   0%|          | 0/2037 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -95,12 +80,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "d3d5c45e3776469f9560b6eaa9346f8f",
+       "model_id": "a6964a9de5e64d4e80c1906e2bed9f21",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/3 [00:00<?, ?ba/s]"
+       "Map:   0%|          | 0/227 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -108,22 +93,8 @@
     },
     {
      "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e9736f26e9aa450b8d65f95c0b9c81cc",
-       "version_major": 2,
-       "version_minor": 0
-      },
       "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?ba/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "{'sentence': \"The 10,000-odd square metre plot that Stockmann has bought for the Nevsky Center shopping center is located on Nevsky Prospect , St Petersburg 's high street , next to the Vosstaniya Square underground station , in the immediate vicinity of Moscow Station .\",\n",
+       "{'sentence': 'The bank VTB24 provides mortgage loans to buy apartments in the complex at 11-13 % per annum in rubles .',\n",
        " 'label': 1,\n",
        " 'text_label': 'neutral'}"
       ]
@@ -159,12 +130,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "c460989d4ab24e3f97d81ef040b1d1b4",
+       "model_id": "a867fe83918c435ab8a52bee2737f4f3",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Running tokenizer on dataset:   0%|          | 0/3 [00:00<?, ?ba/s]"
+       "Running tokenizer on dataset:   0%|          | 0/2037 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -173,12 +144,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "1acc389b08b94f8a87900b9fbdbccce4",
+       "model_id": "97ceaf1285f348bd8272e2bec54050c6",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Running tokenizer on dataset:   0%|          | 0/1 [00:00<?, ?ba/s]"
+       "Running tokenizer on dataset:   0%|          | 0/227 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -237,63 +208,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "id": "6b3a4090",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [02:21<00:00,  1.81it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:07<00:00,  4.13it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=0: train_ppl=tensor(14.6341, device='cuda:0') train_epoch_loss=tensor(2.6834, device='cuda:0') eval_ppl=tensor(1.0057, device='cuda:0') eval_epoch_loss=tensor(0.0057, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [02:00<00:00,  2.11it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:05<00:00,  5.66it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=1: train_ppl=tensor(1.7576, device='cuda:0') train_epoch_loss=tensor(0.5640, device='cuda:0') eval_ppl=tensor(1.0052, device='cuda:0') eval_epoch_loss=tensor(0.0052, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [01:33<00:00,  2.74it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:04<00:00,  6.23it/s]"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=2: train_ppl=tensor(1.3830, device='cuda:0') train_epoch_loss=tensor(0.3243, device='cuda:0') eval_ppl=tensor(1.0035, device='cuda:0') eval_epoch_loss=tensor(0.0035, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# training and evaluation\n",
     "model = model.to(device)\n",
@@ -375,7 +293,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "id": "bd20cd4c",
    "metadata": {},
    "outputs": [
@@ -383,12 +301,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "9,2M\tbigscience/mt0-large_LORA_SEQ_2_SEQ_LM/adapter_model.bin\r\n"
+      "9,2M\tbigscience/mt0-large_LORA_SEQ_2_SEQ_LM/adapter_model.safetensors\r\n"
      ]
     }
    ],
    "source": [
-    "ckpt = f\"{peft_model_id}/adapter_model.bin\"\n",
+    "ckpt = f\"{peft_model_id}/adapter_model.safetensors\"\n",
     "!du -h $ckpt"
    ]
   },
@@ -473,7 +391,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.5"
+   "version": "3.11.13"
   },
   "vscode": {
    "interpreter": {
diff --git a/examples/conditional_generation/peft_lora_seq2seq_accelerate_big_model_inference.ipynb b/examples/conditional_generation/peft_lora_seq2seq_accelerate_big_model_inference.ipynb
deleted file mode 100644
index 132cb01556..0000000000
--- a/examples/conditional_generation/peft_lora_seq2seq_accelerate_big_model_inference.ipynb
+++ /dev/null
@@ -1,253 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "71fbfca2",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from transformers import AutoModelForSeq2SeqLM\n",
-    "from peft import PeftModel, PeftConfig\n",
-    "import torch\n",
-    "from datasets import load_dataset\n",
-    "import os\n",
-    "from transformers import AutoTokenizer\n",
-    "from torch.utils.data import DataLoader\n",
-    "from transformers import default_data_collator, get_linear_schedule_with_warmup\n",
-    "from tqdm import tqdm\n",
-    "from datasets import load_dataset\n",
-    "\n",
-    "dataset_name = \"twitter_complaints\"\n",
-    "text_column = \"Tweet text\"\n",
-    "label_column = \"text_label\"\n",
-    "batch_size = 8\n",
-    "\n",
-    "peft_model_id = \"smangrul/twitter_complaints_bigscience_T0_3B_LORA_SEQ_2_SEQ_LM\"\n",
-    "config = PeftConfig.from_pretrained(peft_model_id)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "cc55820a",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "peft_model_id = \"smangrul/twitter_complaints_bigscience_T0_3B_LORA_SEQ_2_SEQ_LM\"\n",
-    "max_memory = {0: \"6GIB\", 1: \"0GIB\", 2: \"0GIB\", 3: \"0GIB\", 4: \"0GIB\", \"cpu\": \"30GB\"}\n",
-    "config = PeftConfig.from_pretrained(peft_model_id)\n",
-    "model = AutoModelForSeq2SeqLM.from_pretrained(config.base_model_name_or_path, device_map=\"auto\", max_memory=max_memory)\n",
-    "model = PeftModel.from_pretrained(model, peft_model_id, device_map=\"auto\", max_memory=max_memory)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "e1a3648b",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from datasets import load_dataset\n",
-    "\n",
-    "dataset = load_dataset(\"ought/raft\", dataset_name)\n",
-    "\n",
-    "classes = [k.replace(\"_\", \" \") for k in dataset[\"train\"].features[\"Label\"].names]\n",
-    "print(classes)\n",
-    "dataset = dataset.map(\n",
-    "    lambda x: {\"text_label\": [classes[label] for label in x[\"Label\"]]},\n",
-    "    batched=True,\n",
-    "    num_proc=1,\n",
-    ")\n",
-    "print(dataset)\n",
-    "dataset[\"train\"][0]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "fe12d4d3",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)\n",
-    "target_max_length = max([len(tokenizer(class_label)[\"input_ids\"]) for class_label in classes])\n",
-    "\n",
-    "\n",
-    "def preprocess_function(examples):\n",
-    "    inputs = examples[text_column]\n",
-    "    targets = examples[label_column]\n",
-    "    model_inputs = tokenizer(inputs, truncation=True)\n",
-    "    labels = tokenizer(\n",
-    "        targets, max_length=target_max_length, padding=\"max_length\", truncation=True, return_tensors=\"pt\"\n",
-    "    )\n",
-    "    labels = labels[\"input_ids\"]\n",
-    "    labels[labels == tokenizer.pad_token_id] = -100\n",
-    "    model_inputs[\"labels\"] = labels\n",
-    "    return model_inputs\n",
-    "\n",
-    "\n",
-    "processed_datasets = dataset.map(\n",
-    "    preprocess_function,\n",
-    "    batched=True,\n",
-    "    num_proc=1,\n",
-    "    remove_columns=dataset[\"train\"].column_names,\n",
-    "    load_from_cache_file=True,\n",
-    "    desc=\"Running tokenizer on dataset\",\n",
-    ")\n",
-    "\n",
-    "train_dataset = processed_datasets[\"train\"]\n",
-    "eval_dataset = processed_datasets[\"train\"]\n",
-    "test_dataset = processed_datasets[\"test\"]\n",
-    "\n",
-    "\n",
-    "def collate_fn(examples):\n",
-    "    return tokenizer.pad(examples, padding=\"longest\", return_tensors=\"pt\")\n",
-    "\n",
-    "\n",
-    "train_dataloader = DataLoader(\n",
-    "    train_dataset, shuffle=True, collate_fn=collate_fn, batch_size=batch_size, pin_memory=True\n",
-    ")\n",
-    "eval_dataloader = DataLoader(eval_dataset, collate_fn=collate_fn, batch_size=batch_size, pin_memory=True)\n",
-    "test_dataloader = DataLoader(test_dataset, collate_fn=collate_fn, batch_size=batch_size, pin_memory=True)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "id": "b33be5e6",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "@NYTsupport i have complained a dozen times &amp; yet my papers are still thrown FAR from my door. Why is this so hard to resolve?\n",
-      "{'input_ids': tensor([[25335,  1499,     3,    10,  3320, 12056,   382, 20390,     3,    23,\n",
-      "            43, 25932,     3,     9,  9611,   648,     3,   184,  4624,   117,\n",
-      "           780,    82,  5778,    33,   341,     3, 12618,   377,  4280,    45,\n",
-      "            82,  1365,     5,  1615,    19,    48,    78,   614,    12,  7785,\n",
-      "            58, 16229,     3,    10,     3,     1]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
-      "         1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]])}\n",
-      "tensor([[    0, 10394,     1]], device='cuda:0')\n",
-      "['complaint']\n"
-     ]
-    }
-   ],
-   "source": [
-    "model.eval()\n",
-    "i = 15\n",
-    "inputs = tokenizer(f'{text_column} : {dataset[\"test\"][i][\"Tweet text\"]} Label : ', return_tensors=\"pt\")\n",
-    "print(dataset[\"test\"][i][\"Tweet text\"])\n",
-    "print(inputs)\n",
-    "\n",
-    "with torch.no_grad():\n",
-    "    outputs = model.generate(input_ids=inputs[\"input_ids\"].to(\"cuda\"), max_new_tokens=10)\n",
-    "    print(outputs)\n",
-    "    print(tokenizer.batch_decode(outputs.detach().cpu().numpy(), skip_special_tokens=True))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "id": "b6d6cd5b",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "  0%|                                                                                                    | 0/7 [00:00<?, ?it/s]You're using a T5TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
-      "100%|████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00,  1.48s/it]\n"
-     ]
-    }
-   ],
-   "source": [
-    "model.eval()\n",
-    "eval_preds = []\n",
-    "for _, batch in enumerate(tqdm(eval_dataloader)):\n",
-    "    batch = {k: v.to(\"cuda\") for k, v in batch.items() if k != \"labels\"}\n",
-    "    with torch.no_grad():\n",
-    "        outputs = model.generate(**batch, max_new_tokens=10)\n",
-    "    preds = outputs.detach().cpu().numpy()\n",
-    "    eval_preds.extend(tokenizer.batch_decode(preds, skip_special_tokens=True))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "id": "61264abe",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "accuracy=100.0\n",
-      "eval_preds[:10]=['no complaint', 'no complaint', 'complaint', 'complaint', 'no complaint', 'no complaint', 'no complaint', 'complaint', 'complaint', 'no complaint']\n",
-      "dataset['train'][label_column][:10]=['no complaint', 'no complaint', 'complaint', 'complaint', 'no complaint', 'no complaint', 'no complaint', 'complaint', 'complaint', 'no complaint']\n"
-     ]
-    }
-   ],
-   "source": [
-    "correct = 0\n",
-    "total = 0\n",
-    "for pred, true in zip(eval_preds, dataset[\"train\"][label_column]):\n",
-    "    if pred.strip() == true.strip():\n",
-    "        correct += 1\n",
-    "    total += 1\n",
-    "accuracy = correct / total * 100\n",
-    "print(f\"{accuracy=}\")\n",
-    "print(f\"{eval_preds[:10]=}\")\n",
-    "print(f\"{dataset['train'][label_column][:10]=}\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "a70802a3",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "model.eval()\n",
-    "test_preds = []\n",
-    "\n",
-    "for _, batch in enumerate(tqdm(test_dataloader)):\n",
-    "    batch = {k: v for k, v in batch.items() if k != \"labels\"}\n",
-    "    with torch.no_grad():\n",
-    "        outputs = model.generate(**batch, max_new_tokens=10)\n",
-    "    preds = outputs.detach().cpu().numpy()\n",
-    "    test_preds.extend(tokenizer.batch_decode(preds, skip_special_tokens=True))\n",
-    "    if len(test_preds) > 100:\n",
-    "        break\n",
-    "test_preds"
-   ]
-  }
- ],
- "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": "3.10.5 (v3.10.5:f377153967, Jun  6 2022, 12:36:10) [Clang 13.0.0 (clang-1300.0.29.30)]"
-  },
-  "vscode": {
-   "interpreter": {
-    "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49"
-   }
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/examples/conditional_generation/peft_lora_seq2seq_accelerate_ds_zero3_offload.py b/examples/conditional_generation/peft_lora_seq2seq_accelerate_ds_zero3_offload.py
index 4be3e1666b..50bd15a89e 100644
--- a/examples/conditional_generation/peft_lora_seq2seq_accelerate_ds_zero3_offload.py
+++ b/examples/conditional_generation/peft_lora_seq2seq_accelerate_ds_zero3_offload.py
@@ -54,10 +54,12 @@ def b2mb(x):
 # This context manager is used to track the peak memory usage of the process
 class TorchTracemalloc:
     def __enter__(self):
+        self.device_type = torch.accelerator.current_accelerator().type if hasattr(torch, "accelerator") else "cuda"
+        self.device_module = getattr(torch, self.device_type, torch.cuda)
         gc.collect()
-        torch.cuda.empty_cache()
-        torch.cuda.reset_max_memory_allocated()  # reset the peak gauge to zero
-        self.begin = torch.cuda.memory_allocated()
+        self.device_module.empty_cache()
+        self.device_module.reset_peak_memory_stats()  # reset the peak gauge to zero
+        self.begin = self.device_module.memory_allocated()
         self.process = psutil.Process()
 
         self.cpu_begin = self.cpu_mem_used()
@@ -87,9 +89,9 @@ def __exit__(self, *exc):
         self.peak_monitoring = False
 
         gc.collect()
-        torch.cuda.empty_cache()
-        self.end = torch.cuda.memory_allocated()
-        self.peak = torch.cuda.max_memory_allocated()
+        self.device_module.empty_cache()
+        self.end = self.device_module.memory_allocated()
+        self.peak = self.device_module.max_memory_allocated()
         self.used = b2mb(self.end - self.begin)
         self.peaked = b2mb(self.peak - self.begin)
 
@@ -199,12 +201,18 @@ def collate_fn(examples):
                 optimizer.step()
                 lr_scheduler.step()
                 optimizer.zero_grad()
-        # Printing the GPU memory usage details such as allocated memory, peak memory, and total memory usage
-        accelerator.print(f"GPU Memory before entering the train : {b2mb(tracemalloc.begin)}")
-        accelerator.print(f"GPU Memory consumed at the end of the train (end-begin): {tracemalloc.used}")
-        accelerator.print(f"GPU Peak Memory consumed during the train (max-begin): {tracemalloc.peaked}")
+        # Printing the device memory usage details such as allocated memory, peak memory, and total memory usage
         accelerator.print(
-            f"GPU Total Peak Memory consumed during the train (max): {tracemalloc.peaked + b2mb(tracemalloc.begin)}"
+            f"{accelerator.device.type.upper()} Memory before entering the train : {b2mb(tracemalloc.begin)}"
+        )
+        accelerator.print(
+            f"{accelerator.device.type.upper()} Memory consumed at the end of the train (end-begin): {tracemalloc.used}"
+        )
+        accelerator.print(
+            f"{accelerator.device.type.upper()} Peak Memory consumed during the train (max-begin): {tracemalloc.peaked}"
+        )
+        accelerator.print(
+            f"{accelerator.device.type.upper()} Total Peak Memory consumed during the train (max): {tracemalloc.peaked + b2mb(tracemalloc.begin)}"
         )
 
         accelerator.print(f"CPU Memory before entering the train : {b2mb(tracemalloc.cpu_begin)}")
@@ -230,12 +238,18 @@ def collate_fn(examples):
                 preds = accelerator.gather_for_metrics(outputs).detach().cpu().numpy()
                 eval_preds.extend(tokenizer.batch_decode(preds, skip_special_tokens=True))
 
-        # Printing the GPU memory usage details such as allocated memory, peak memory, and total memory usage
-        accelerator.print(f"GPU Memory before entering the eval : {b2mb(tracemalloc.begin)}")
-        accelerator.print(f"GPU Memory consumed at the end of the eval (end-begin): {tracemalloc.used}")
-        accelerator.print(f"GPU Peak Memory consumed during the eval (max-begin): {tracemalloc.peaked}")
+        # Printing the device memory usage details such as allocated memory, peak memory, and total memory usage
+        accelerator.print(
+            f"{accelerator.device.type.upper()} Memory before entering the eval : {b2mb(tracemalloc.begin)}"
+        )
+        accelerator.print(
+            f"{accelerator.device.type.upper()} Memory consumed at the end of the eval (end-begin): {tracemalloc.used}"
+        )
+        accelerator.print(
+            f"{accelerator.device.type.upper()} Peak Memory consumed during the eval (max-begin): {tracemalloc.peaked}"
+        )
         accelerator.print(
-            f"GPU Total Peak Memory consumed during the eval (max): {tracemalloc.peaked + b2mb(tracemalloc.begin)}"
+            f"{accelerator.device.type.upper()} Total Peak Memory consumed during the eval (max): {tracemalloc.peaked + b2mb(tracemalloc.begin)}"
         )
 
         accelerator.print(f"CPU Memory before entering the eval : {b2mb(tracemalloc.cpu_begin)}")
diff --git a/examples/conditional_generation/peft_prefix_tuning_seq2seq.ipynb b/examples/conditional_generation/peft_prefix_tuning_seq2seq.ipynb
index aa85f9a743..259fc665dc 100644
--- a/examples/conditional_generation/peft_prefix_tuning_seq2seq.ipynb
+++ b/examples/conditional_generation/peft_prefix_tuning_seq2seq.ipynb
@@ -2,26 +2,10 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "5f93b7d1",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "===================================BUG REPORT===================================\n",
-      "Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues\n",
-      "For effortless bug reporting copy-paste your error into this form: https://docs.google.com/forms/d/e/1FAIpQLScPB8emS3Thkp66nvqwmjTEgxp8Y9ufuWTzFyr9kJ5AoI47dQ/viewform?usp=sf_link\n",
-      "================================================================================\n",
-      "CUDA SETUP: CUDA runtime path found: /home/sourab/miniconda3/envs/ml/lib/libcudart.so\n",
-      "CUDA SETUP: Highest compute capability among GPUs detected: 7.5\n",
-      "CUDA SETUP: Detected CUDA version 117\n",
-      "CUDA SETUP: Loading binary /home/sourab/miniconda3/envs/ml/lib/python3.10/site-packages/bitsandbytes/libbitsandbytes_cuda117.so...\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from transformers import AutoModelForSeq2SeqLM\n",
     "from peft import get_peft_config, get_peft_model, get_peft_model_state_dict, PrefixTuningConfig, TaskType\n",
@@ -30,14 +14,13 @@
     "import os\n",
     "\n",
     "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
-    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"3\"\n",
     "from transformers import AutoTokenizer\n",
     "from torch.utils.data import DataLoader\n",
     "from transformers import default_data_collator, get_linear_schedule_with_warmup\n",
     "from tqdm import tqdm\n",
     "from datasets import load_dataset\n",
     "\n",
-    "device = \"cuda\"\n",
+    "device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
     "model_name_or_path = \"t5-large\"\n",
     "tokenizer_name_or_path = \"t5-large\"\n",
     "\n",
@@ -52,7 +35,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "id": "8d0850ac",
    "metadata": {},
    "outputs": [],
@@ -76,18 +59,19 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Found cached dataset financial_phrasebank (/home/sourab/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141)\n"
+      "Using the latest cached version of the dataset since financial_phrasebank couldn't be found on the Hugging Face Hub\n",
+      "Found the latest cached dataset configuration 'sentences_allagree' at /root/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141 (last modified on Thu Jul 31 06:23:15 2025).\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "ec4be98991b84181bfa75f8846422b8b",
+       "model_id": "3b321971d6f942418bd5ef6105a1aa65",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?it/s]"
+       "Map:   0%|          | 0/2037 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -96,12 +80,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "82a6bd694c4f4751a23c370ab51f01a4",
+       "model_id": "5997543529a849bf97719e59a5ec95b2",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/3 [00:00<?, ?ba/s]"
+       "Map:   0%|          | 0/227 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -109,22 +93,8 @@
     },
     {
      "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3844878631534468a1495e435563e4b0",
-       "version_major": 2,
-       "version_minor": 0
-      },
       "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?ba/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "{'sentence': 'Finnish elevators and escalators maker KONE Corporation said on Tuesday ( 18 March ) that it has received a major order from Sir Robert McAlpine to supply all elevators and escalators for the Watermark Place project in the City of London .',\n",
+       "{'sentence': \"Progress Group , QPR 's representative in Saudi Arabia and North Africa , has signed a framework agreement for a long term strategic relationship with ISE .\",\n",
        " 'label': 2,\n",
        " 'text_label': 'positive'}"
       ]
@@ -158,26 +128,42 @@
    "metadata": {},
    "outputs": [
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/sourab/transformers/src/transformers/models/t5/tokenization_t5_fast.py:155: FutureWarning: This tokenizer was incorrectly instantiated with a model max length of 512 which will be corrected in Transformers v5.\n",
-      "For now, this behavior is kept to avoid breaking backwards compatibility when padding/encoding with `truncation is True`.\n",
-      "- Be aware that you SHOULD NOT rely on t5-large automatically truncating your input to 512 when padding/encoding.\n",
-      "- If you want to encode/pad to sequences longer than 512 you can either instantiate this tokenizer with `model_max_length` or pass `max_length` when encoding/padding.\n",
-      "- To avoid this warning, please instantiate this tokenizer with `model_max_length` set to your preferred value.\n",
-      "  warnings.warn(\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ee9bf13a2e3f4812a51a87346a4614f3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "spiece.model:   0%|          | 0.00/792k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "332fcaa33dc343e7a20b24cec7ec97e9",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "tokenizer.json:   0%|          | 0.00/1.39M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "4af8c12efb5643659573347509079f3a",
+       "model_id": "a2135262c7a44377b35fe32b8d86d6c6",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Running tokenizer on dataset:   0%|          | 0/3 [00:00<?, ?ba/s]"
+       "Running tokenizer on dataset:   0%|          | 0/2037 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -186,12 +172,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "86033b6257384584afd034075af808cb",
+       "model_id": "dc8598b160484939b27c65be001c694c",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "Running tokenizer on dataset:   0%|          | 0/1 [00:00<?, ?ba/s]"
+       "Running tokenizer on dataset:   0%|          | 0/227 [00:00<?, ? examples/s]"
       ]
      },
      "metadata": {},
@@ -250,86 +236,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "id": "6b3a4090",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:49<00:00,  5.15it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:03<00:00,  7.56it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=0: train_ppl=tensor(2760654.5000, device='cuda:0') train_epoch_loss=tensor(14.8310, device='cuda:0') eval_ppl=tensor(1.0124, device='cuda:0') eval_epoch_loss=tensor(0.0124, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:40<00:00,  6.22it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:05<00:00,  5.05it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=1: train_ppl=tensor(2.7329, device='cuda:0') train_epoch_loss=tensor(1.0054, device='cuda:0') eval_ppl=tensor(1.0081, device='cuda:0') eval_epoch_loss=tensor(0.0080, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:58<00:00,  4.36it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:05<00:00,  5.05it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=2: train_ppl=tensor(2.1698, device='cuda:0') train_epoch_loss=tensor(0.7747, device='cuda:0') eval_ppl=tensor(1.0057, device='cuda:0') eval_epoch_loss=tensor(0.0057, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:58<00:00,  4.35it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:05<00:00,  5.06it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=3: train_ppl=tensor(2.0724, device='cuda:0') train_epoch_loss=tensor(0.7287, device='cuda:0') eval_ppl=tensor(1.0051, device='cuda:0') eval_epoch_loss=tensor(0.0051, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|████████████████████████████████████████████████████████████████████████████████████████| 255/255 [01:02<00:00,  4.10it/s]\n",
-      "100%|██████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:06<00:00,  4.74it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=4: train_ppl=tensor(1.7598, device='cuda:0') train_epoch_loss=tensor(0.5652, device='cuda:0') eval_ppl=tensor(1.0047, device='cuda:0') eval_epoch_loss=tensor(0.0047, device='cuda:0')\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# training and evaluation\n",
     "model = model.to(device)\n",
@@ -411,7 +321,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "id": "bd20cd4c",
    "metadata": {},
    "outputs": [
@@ -419,12 +329,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "3,8M\tt5-large_PREFIX_TUNING_SEQ_2_SEQ_LM/adapter_model.bin\r\n"
+      "3,8M\tt5-large_PREFIX_TUNING_SEQ_2_SEQ_LM/adapter_model.safetensors\r\n"
      ]
     }
    ],
    "source": [
-    "ckpt = f\"{peft_model_id}/adapter_model.bin\"\n",
+    "ckpt = f\"{peft_model_id}/adapter_model.safetensors\"\n",
     "!du -h $ckpt"
    ]
   },
@@ -503,7 +413,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.5"
+   "version": "3.11.13"
   },
   "vscode": {
    "interpreter": {
diff --git a/examples/conditional_generation/peft_prompt_tuning_seq2seq.ipynb b/examples/conditional_generation/peft_prompt_tuning_seq2seq.ipynb
index fb2462b6fd..75806436ee 100644
--- a/examples/conditional_generation/peft_prompt_tuning_seq2seq.ipynb
+++ b/examples/conditional_generation/peft_prompt_tuning_seq2seq.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "5f93b7d1",
    "metadata": {
     "ExecuteTime": {
@@ -10,50 +10,7 @@
      "start_time": "2023-05-30T08:37:56.881307Z"
     }
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "===================================BUG REPORT===================================\n",
-      "Welcome to bitsandbytes. For bug reports, please run\n",
-      "\n",
-      "python -m bitsandbytes\n",
-      "\n",
-      " and submit this information together with your error trace to: https://github.com/TimDettmers/bitsandbytes/issues\n",
-      "================================================================================\n",
-      "bin /udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/libbitsandbytes_cuda117.so\n",
-      "CUDA_SETUP: WARNING! libcudart.so not found in any environmental path. Searching in backup paths...\n",
-      "CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so.11.0\n",
-      "CUDA SETUP: Highest compute capability among GPUs detected: 8.0\n",
-      "CUDA SETUP: Detected CUDA version 117\n",
-      "CUDA SETUP: Loading binary /udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/libbitsandbytes_cuda117.so...\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: /udir/tschilla/anaconda3 did not contain ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] as expected! Searching further paths...\n",
-      "  warn(msg)\n",
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('Europe/Paris')}\n",
-      "  warn(msg)\n",
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/udir/tschilla/.cache/dotnet_bundle_extract')}\n",
-      "  warn(msg)\n",
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('5002'), PosixPath('http'), PosixPath('//127.0.0.1')}\n",
-      "  warn(msg)\n",
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('() {  ( alias;\\n eval ${which_declare} ) | /usr/bin/which --tty-only --read-alias --read-functions --show-tilde --show-dot $@\\n}')}\n",
-      "  warn(msg)\n",
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('module'), PosixPath('//matplotlib_inline.backend_inline')}\n",
-      "  warn(msg)\n",
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/bitsandbytes/cuda_setup/main.py:149: UserWarning: Found duplicate ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] files: {PosixPath('/usr/local/cuda/lib64/libcudart.so.11.0'), PosixPath('/usr/local/cuda/lib64/libcudart.so')}.. We'll flip a coin and try one of these, in order to fail forward.\n",
-      "Either way, this might cause trouble in the future:\n",
-      "If you get `CUDA error: invalid device function` errors, the above might be the cause and the solution is to make sure only one ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] in the paths that we search based on your env.\n",
-      "  warn(msg)\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import os\n",
     "\n",
@@ -66,7 +23,7 @@
     "\n",
     "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
     "\n",
-    "device = \"cuda\"\n",
+    "device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
     "model_name_or_path = \"t5-large\"\n",
     "tokenizer_name_or_path = \"t5-large\"\n",
     "\n",
@@ -94,19 +51,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "trainable params: 40960 || all params: 737709056 || trainable%: 0.005552324411210698\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/models/t5/tokenization_t5_fast.py:155: FutureWarning: This tokenizer was incorrectly instantiated with a model max length of 512 which will be corrected in Transformers v5.\n",
-      "For now, this behavior is kept to avoid breaking backwards compatibility when padding/encoding with `truncation is True`.\n",
-      "- Be aware that you SHOULD NOT rely on t5-large automatically truncating your input to 512 when padding/encoding.\n",
-      "- If you want to encode/pad to sequences longer than 512 you can either instantiate this tokenizer with `model_max_length` or pass `max_length` when encoding/padding.\n",
-      "- To avoid this warning, please instantiate this tokenizer with `model_max_length` set to your preferred value.\n",
-      "  warnings.warn(\n"
+      "trainable params: 40,960 || all params: 737,709,056 || trainable%: 0.0056\n"
      ]
     },
     {
@@ -295,27 +240,14 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Found cached dataset financial_phrasebank (/data/proxem/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141)\n"
+      "Using the latest cached version of the dataset since financial_phrasebank couldn't be found on the Hugging Face Hub\n",
+      "Found the latest cached dataset configuration 'sentences_allagree' at /root/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141 (last modified on Thu Jul 31 06:37:38 2025).\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "fb63f50cb7cb4f5aae10648ba74d6c4e",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "2b64258700bd40548ddcd626f3920c9a",
        "version_major": 2,
        "version_minor": 0
       },
@@ -329,7 +261,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "d95de9b1dc3d417da35118c14d02b986",
        "version_major": 2,
        "version_minor": 0
       },
@@ -343,9 +275,9 @@
     {
      "data": {
       "text/plain": [
-       "{'sentence': '`` Lining stone sales were also good in the early autumn , and order books are strong to the end of the year .',\n",
-       " 'label': 2,\n",
-       " 'text_label': 'positive'}"
+       "{'sentence': 'The 2500-passenger ferry will have dimensions of 185 m length overall , 170 m length between perpendiculars , 27.70 m breadth and 6.55 m design draught .',\n",
+       " 'label': 1,\n",
+       " 'text_label': 'neutral'}"
       ]
      },
      "execution_count": 3,
@@ -384,7 +316,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "a78b4b1a041546248b8e6703eaec0969",
        "version_major": 2,
        "version_minor": 0
       },
@@ -398,7 +330,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "bbe2962a25144b5488c297e186df824f",
        "version_major": 2,
        "version_minor": 0
       },
@@ -470,7 +402,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "id": "6b3a4090",
    "metadata": {
     "ExecuteTime": {
@@ -478,90 +410,7 @@
      "start_time": "2023-05-30T08:38:50.102263Z"
     }
    },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:42<00:00,  6.05it/s]\n",
-      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.40it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=0: train_ppl=tensor(8.0846, device='cuda:0') train_epoch_loss=tensor(2.0900, device='cuda:0') eval_ppl=tensor(1.3542, device='cuda:0') eval_epoch_loss=tensor(0.3032, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:41<00:00,  6.15it/s]\n",
-      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.42it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=1: train_ppl=tensor(1.5088, device='cuda:0') train_epoch_loss=tensor(0.4113, device='cuda:0') eval_ppl=tensor(1.2692, device='cuda:0') eval_epoch_loss=tensor(0.2384, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:41<00:00,  6.18it/s]\n",
-      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.45it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=2: train_ppl=tensor(1.5322, device='cuda:0') train_epoch_loss=tensor(0.4267, device='cuda:0') eval_ppl=tensor(1.2065, device='cuda:0') eval_epoch_loss=tensor(0.1877, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:41<00:00,  6.17it/s]\n",
-      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.38it/s]\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=3: train_ppl=tensor(1.4475, device='cuda:0') train_epoch_loss=tensor(0.3699, device='cuda:0') eval_ppl=tensor(1.2346, device='cuda:0') eval_epoch_loss=tensor(0.2107, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 255/255 [00:42<00:00,  5.94it/s]\n",
-      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 29/29 [00:02<00:00, 14.42it/s]"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "epoch=4: train_ppl=tensor(1.3428, device='cuda:0') train_epoch_loss=tensor(0.2948, device='cuda:0') eval_ppl=tensor(1.2041, device='cuda:0') eval_epoch_loss=tensor(0.1857, device='cuda:0')\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# training and evaluation\n",
     "model = model.to(device)\n",
@@ -653,7 +502,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "id": "bd20cd4c",
    "metadata": {
     "ExecuteTime": {
@@ -666,12 +515,12 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "164K\tt5-large_PROMPT_TUNING_SEQ_2_SEQ_LM/adapter_model.bin\r\n"
+      "164K\tt5-large_PROMPT_TUNING_SEQ_2_SEQ_LM/adapter_model.safetensors\r\n"
      ]
     }
    ],
    "source": [
-    "ckpt = f\"{peft_model_id}/adapter_model.bin\"\n",
+    "ckpt = f\"{peft_model_id}/adapter_model.safetensors\"\n",
     "!du -h $ckpt"
    ]
   },
@@ -735,9 +584,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "peft",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
-   "name": "peft"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -749,7 +598,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.16"
+   "version": "3.11.13"
   },
   "toc": {
    "base_numbering": 1,
diff --git a/examples/conditional_generation/peft_prompt_tuning_seq2seq_with_generate.ipynb b/examples/conditional_generation/peft_prompt_tuning_seq2seq_with_generate.ipynb
index 84bbed52e6..95be106162 100644
--- a/examples/conditional_generation/peft_prompt_tuning_seq2seq_with_generate.ipynb
+++ b/examples/conditional_generation/peft_prompt_tuning_seq2seq_with_generate.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "5f93b7d1",
    "metadata": {
     "ExecuteTime": {
@@ -10,50 +10,7 @@
      "start_time": "2023-05-30T09:49:54.494916Z"
     }
    },
-   "outputs": [
-    {
-     "ename": "KeyboardInterrupt",
-     "evalue": "",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[1], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtransformers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m AutoModelForSeq2SeqLM, Seq2SeqTrainingArguments, Seq2SeqTrainer, GenerationConfig\n\u001b[1;32m      2\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mpeft\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m get_peft_model, PromptTuningInit, PromptTuningConfig, TaskType\n\u001b[1;32m      3\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap>:1055\u001b[0m, in \u001b[0;36m_handle_fromlist\u001b[0;34m(module, fromlist, import_, recursive)\u001b[0m\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/utils/import_utils.py:1076\u001b[0m, in \u001b[0;36m_LazyModule.__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m   1074\u001b[0m     value \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_get_module(name)\n\u001b[1;32m   1075\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m name \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_class_to_module\u001b[38;5;241m.\u001b[39mkeys():\n\u001b[0;32m-> 1076\u001b[0m     module \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_module\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_class_to_module\u001b[49m\u001b[43m[\u001b[49m\u001b[43mname\u001b[49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1077\u001b[0m     value \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mgetattr\u001b[39m(module, name)\n\u001b[1;32m   1078\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/utils/import_utils.py:1086\u001b[0m, in \u001b[0;36m_LazyModule._get_module\u001b[0;34m(self, module_name)\u001b[0m\n\u001b[1;32m   1084\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_get_module\u001b[39m(\u001b[38;5;28mself\u001b[39m, module_name: \u001b[38;5;28mstr\u001b[39m):\n\u001b[1;32m   1085\u001b[0m     \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 1086\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mimportlib\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mimport_module\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m.\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mmodule_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[38;5;18;43m__name__\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1087\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m   1088\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mRuntimeError\u001b[39;00m(\n\u001b[1;32m   1089\u001b[0m             \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFailed to import \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\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;132;01m{\u001b[39;00mmodule_name\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m because of the following error (look up to see its\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m   1090\u001b[0m             \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m traceback):\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;132;01m{\u001b[39;00me\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m   1091\u001b[0m         ) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01me\u001b[39;00m\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/importlib/__init__.py:127\u001b[0m, in \u001b[0;36mimport_module\u001b[0;34m(name, package)\u001b[0m\n\u001b[1;32m    125\u001b[0m             \u001b[38;5;28;01mbreak\u001b[39;00m\n\u001b[1;32m    126\u001b[0m         level \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[0;32m--> 127\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_bootstrap\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_gcd_import\u001b[49m\u001b[43m(\u001b[49m\u001b[43mname\u001b[49m\u001b[43m[\u001b[49m\u001b[43mlevel\u001b[49m\u001b[43m:\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mpackage\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlevel\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/training_args_seq2seq.py:21\u001b[0m\n\u001b[1;32m     18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtyping\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Optional, Union\n\u001b[1;32m     20\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mgeneration\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mconfiguration_utils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m GenerationConfig\n\u001b[0;32m---> 21\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtraining_args\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m TrainingArguments\n\u001b[1;32m     22\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m add_start_docstrings\n\u001b[1;32m     25\u001b[0m logger \u001b[38;5;241m=\u001b[39m logging\u001b[38;5;241m.\u001b[39mgetLogger(\u001b[38;5;18m__name__\u001b[39m)\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/training_args.py:29\u001b[0m\n\u001b[1;32m     25\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtyping\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Any, Dict, List, Optional, Union\n\u001b[1;32m     27\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mpackaging\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m version\n\u001b[0;32m---> 29\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdebug_utils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m DebugOption\n\u001b[1;32m     30\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtrainer_utils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[1;32m     31\u001b[0m     EvaluationStrategy,\n\u001b[1;32m     32\u001b[0m     FSDPOption,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     36\u001b[0m     ShardedDDPOption,\n\u001b[1;32m     37\u001b[0m )\n\u001b[1;32m     38\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[1;32m     39\u001b[0m     ExplicitEnum,\n\u001b[1;32m     40\u001b[0m     cached_property,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     53\u001b[0m     requires_backends,\n\u001b[1;32m     54\u001b[0m )\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/debug_utils.py:21\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m ExplicitEnum, is_torch_available, logging\n\u001b[1;32m     20\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m is_torch_available():\n\u001b[0;32m---> 21\u001b[0m     \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\n\u001b[1;32m     24\u001b[0m logger \u001b[38;5;241m=\u001b[39m logging\u001b[38;5;241m.\u001b[39mget_logger(\u001b[38;5;18m__name__\u001b[39m)\n\u001b[1;32m     27\u001b[0m \u001b[38;5;28;01mclass\u001b[39;00m \u001b[38;5;21;01mDebugUnderflowOverflow\u001b[39;00m:\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/__init__.py:1465\u001b[0m\n\u001b[1;32m   1463\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m library\n\u001b[1;32m   1464\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m TYPE_CHECKING:\n\u001b[0;32m-> 1465\u001b[0m     \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m _meta_registrations\n\u001b[1;32m   1467\u001b[0m \u001b[38;5;66;03m# Enable CUDA Sanitizer\u001b[39;00m\n\u001b[1;32m   1468\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mTORCH_CUDA_SANITIZER\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;129;01min\u001b[39;00m os\u001b[38;5;241m.\u001b[39menviron:\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_meta_registrations.py:7\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims_common\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mutils\u001b[39;00m\n\u001b[1;32m      6\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Tensor\n\u001b[0;32m----> 7\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_decomp\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m _add_op_to_registry, global_decomposition_table, meta_table\n\u001b[1;32m      8\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_ops\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m OpOverload\n\u001b[1;32m      9\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m _elementwise_meta, ELEMENTWISE_PRIM_TYPE_PROMOTION_KIND\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_decomp/__init__.py:169\u001b[0m\n\u001b[1;32m    165\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m decompositions\n\u001b[1;32m    168\u001b[0m \u001b[38;5;66;03m# populate the table\u001b[39;00m\n\u001b[0;32m--> 169\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_decomp\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdecompositions\u001b[39;00m\n\u001b[1;32m    170\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_refs\u001b[39;00m\n\u001b[1;32m    172\u001b[0m \u001b[38;5;66;03m# This list was copied from torch/_inductor/decomposition.py\u001b[39;00m\n\u001b[1;32m    173\u001b[0m \u001b[38;5;66;03m# excluding decompositions that results in prim ops\u001b[39;00m\n\u001b[1;32m    174\u001b[0m \u001b[38;5;66;03m# Resulting opset of decomposition is core aten ops\u001b[39;00m\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_decomp/decompositions.py:10\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtyping\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Callable, cast, Iterable, List, Optional, Tuple, Union\n\u001b[1;32m      9\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\n\u001b[0;32m---> 10\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mprims\u001b[39;00m\n\u001b[1;32m     11\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims_common\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mutils\u001b[39;00m\n\u001b[1;32m     12\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mnn\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctional\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mF\u001b[39;00m\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_prims/__init__.py:33\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims_common\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[1;32m     18\u001b[0m     check,\n\u001b[1;32m     19\u001b[0m     Dim,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     30\u001b[0m     type_to_dtype,\n\u001b[1;32m     31\u001b[0m )\n\u001b[1;32m     32\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims_common\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mwrappers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m backwards_not_supported\n\u001b[0;32m---> 33\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_subclasses\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfake_tensor\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m FakeTensor, FakeTensorMode\n\u001b[1;32m     34\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01moverrides\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m handle_torch_function, has_torch_function\n\u001b[1;32m     35\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_pytree\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m tree_flatten, tree_map, tree_unflatten\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_subclasses/__init__.py:3\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\n\u001b[0;32m----> 3\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_subclasses\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfake_tensor\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[1;32m      4\u001b[0m     DynamicOutputShapeException,\n\u001b[1;32m      5\u001b[0m     FakeTensor,\n\u001b[1;32m      6\u001b[0m     FakeTensorMode,\n\u001b[1;32m      7\u001b[0m     UnsupportedFakeTensorException,\n\u001b[1;32m      8\u001b[0m )\n\u001b[1;32m     10\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_subclasses\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfake_utils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m CrossRefFakeMode\n\u001b[1;32m     12\u001b[0m __all__ \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m     13\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFakeTensor\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m     14\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFakeTensorMode\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     17\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCrossRefFakeMode\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m     18\u001b[0m ]\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_subclasses/fake_tensor.py:13\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mweakref\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m ReferenceType\n\u001b[1;32m     12\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\n\u001b[0;32m---> 13\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_guards\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Source\n\u001b[1;32m     14\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_ops\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m OpOverload\n\u001b[1;32m     15\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m_prims_common\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[1;32m     16\u001b[0m     elementwise_dtypes,\n\u001b[1;32m     17\u001b[0m     ELEMENTWISE_TYPE_PROMOTION_KIND,\n\u001b[1;32m     18\u001b[0m     is_float_dtype,\n\u001b[1;32m     19\u001b[0m     is_integer_dtype,\n\u001b[1;32m     20\u001b[0m )\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/torch/_guards.py:14\u001b[0m\n\u001b[1;32m     11\u001b[0m \u001b[38;5;66;03m# TODO(voz): Stolen pattern, not sure why this is the case,\u001b[39;00m\n\u001b[1;32m     12\u001b[0m \u001b[38;5;66;03m# but mypy complains.\u001b[39;00m\n\u001b[1;32m     13\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m---> 14\u001b[0m     \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m  \u001b[38;5;66;03m# type: ignore[import]\u001b[39;00m\n\u001b[1;32m     15\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mImportError\u001b[39;00m:\n\u001b[1;32m     16\u001b[0m     log\u001b[38;5;241m.\u001b[39mwarning(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mNo sympy found\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/__init__.py:74\u001b[0m\n\u001b[1;32m     67\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlogic\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (to_cnf, to_dnf, to_nnf, And, Or, Not, Xor, Nand, Nor,\n\u001b[1;32m     68\u001b[0m         Implies, Equivalent, ITE, POSform, SOPform, simplify_logic, bool_map,\n\u001b[1;32m     69\u001b[0m         true, false, satisfiable)\n\u001b[1;32m     71\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01massumptions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (AppliedPredicate, Predicate, AssumptionsContext,\n\u001b[1;32m     72\u001b[0m         assuming, Q, ask, register_handler, remove_handler, refine)\n\u001b[0;32m---> 74\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (Poly, PurePoly, poly_from_expr, parallel_poly_from_expr,\n\u001b[1;32m     75\u001b[0m         degree, total_degree, degree_list, LC, LM, LT, pdiv, prem, pquo,\n\u001b[1;32m     76\u001b[0m         pexquo, div, rem, quo, exquo, half_gcdex, gcdex, invert,\n\u001b[1;32m     77\u001b[0m         subresultants, resultant, discriminant, cofactors, gcd_list, gcd,\n\u001b[1;32m     78\u001b[0m         lcm_list, lcm, terms_gcd, trunc, monic, content, primitive, compose,\n\u001b[1;32m     79\u001b[0m         decompose, sturm, gff_list, gff, sqf_norm, sqf_part, sqf_list, sqf,\n\u001b[1;32m     80\u001b[0m         factor_list, factor, intervals, refine_root, count_roots, real_roots,\n\u001b[1;32m     81\u001b[0m         nroots, ground_roots, nth_power_roots_poly, cancel, reduced, groebner,\n\u001b[1;32m     82\u001b[0m         is_zero_dimensional, GroebnerBasis, poly, symmetrize, horner,\n\u001b[1;32m     83\u001b[0m         interpolate, rational_interpolate, viete, together,\n\u001b[1;32m     84\u001b[0m         BasePolynomialError, ExactQuotientFailed, PolynomialDivisionFailed,\n\u001b[1;32m     85\u001b[0m         OperationNotSupported, HeuristicGCDFailed, HomomorphismFailed,\n\u001b[1;32m     86\u001b[0m         IsomorphismFailed, ExtraneousFactors, EvaluationFailed,\n\u001b[1;32m     87\u001b[0m         RefinementFailed, CoercionFailed, NotInvertible, NotReversible,\n\u001b[1;32m     88\u001b[0m         NotAlgebraic, DomainError, PolynomialError, UnificationFailed,\n\u001b[1;32m     89\u001b[0m         GeneratorsError, GeneratorsNeeded, ComputationFailed,\n\u001b[1;32m     90\u001b[0m         UnivariatePolynomialError, MultivariatePolynomialError,\n\u001b[1;32m     91\u001b[0m         PolificationFailed, OptionError, FlagError, minpoly,\n\u001b[1;32m     92\u001b[0m         minimal_polynomial, primitive_element, field_isomorphism,\n\u001b[1;32m     93\u001b[0m         to_number_field, isolate, round_two, prime_decomp, prime_valuation,\n\u001b[1;32m     94\u001b[0m         galois_group, itermonomials, Monomial, lex, grlex,\n\u001b[1;32m     95\u001b[0m         grevlex, ilex, igrlex, igrevlex, CRootOf, rootof, RootOf,\n\u001b[1;32m     96\u001b[0m         ComplexRootOf, RootSum, roots, Domain, FiniteField, IntegerRing,\n\u001b[1;32m     97\u001b[0m         RationalField, RealField, ComplexField, PythonFiniteField,\n\u001b[1;32m     98\u001b[0m         GMPYFiniteField, PythonIntegerRing, GMPYIntegerRing, PythonRational,\n\u001b[1;32m     99\u001b[0m         GMPYRationalField, AlgebraicField, PolynomialRing, FractionField,\n\u001b[1;32m    100\u001b[0m         ExpressionDomain, FF_python, FF_gmpy, ZZ_python, ZZ_gmpy, QQ_python,\n\u001b[1;32m    101\u001b[0m         QQ_gmpy, GF, FF, ZZ, QQ, ZZ_I, QQ_I, RR, CC, EX, EXRAW,\n\u001b[1;32m    102\u001b[0m         construct_domain, swinnerton_dyer_poly, cyclotomic_poly,\n\u001b[1;32m    103\u001b[0m         symmetric_poly, random_poly, interpolating_poly, jacobi_poly,\n\u001b[1;32m    104\u001b[0m         chebyshevt_poly, chebyshevu_poly, hermite_poly, hermite_prob_poly,\n\u001b[1;32m    105\u001b[0m         legendre_poly, laguerre_poly, apart, apart_list, assemble_partfrac_list,\n\u001b[1;32m    106\u001b[0m         Options, ring, xring, vring, sring, field, xfield, vfield, sfield)\n\u001b[1;32m    108\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mseries\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (Order, O, limit, Limit, gruntz, series, approximants,\n\u001b[1;32m    109\u001b[0m         residue, EmptySequence, SeqPer, SeqFormula, sequence, SeqAdd, SeqMul,\n\u001b[1;32m    110\u001b[0m         fourier_series, fps, difference_delta, limit_seq)\n\u001b[1;32m    112\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (factorial, factorial2, rf, ff, binomial,\n\u001b[1;32m    113\u001b[0m         RisingFactorial, FallingFactorial, subfactorial, carmichael,\n\u001b[1;32m    114\u001b[0m         fibonacci, lucas, motzkin, tribonacci, harmonic, bernoulli, bell, euler,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    133\u001b[0m         Znm, elliptic_k, elliptic_f, elliptic_e, elliptic_pi, beta, mathieus,\n\u001b[1;32m    134\u001b[0m         mathieuc, mathieusprime, mathieucprime, riemann_xi, betainc, betainc_regularized)\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/polys/__init__.py:78\u001b[0m\n\u001b[1;32m      3\u001b[0m __all__ \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m      4\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mPoly\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mPurePoly\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpoly_from_expr\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mparallel_poly_from_expr\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mdegree\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m      5\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtotal_degree\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mdegree_list\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mLC\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mLM\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mLT\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpdiv\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mprem\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpquo\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     65\u001b[0m     \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mfield\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mxfield\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mvfield\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124msfield\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     66\u001b[0m ]\n\u001b[1;32m     68\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolytools\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (Poly, PurePoly, poly_from_expr,\n\u001b[1;32m     69\u001b[0m         parallel_poly_from_expr, degree, total_degree, degree_list, LC, LM,\n\u001b[1;32m     70\u001b[0m         LT, pdiv, prem, pquo, pexquo, div, rem, quo, exquo, half_gcdex, gcdex,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     75\u001b[0m         count_roots, real_roots, nroots, ground_roots, nth_power_roots_poly,\n\u001b[1;32m     76\u001b[0m         cancel, reduced, groebner, is_zero_dimensional, GroebnerBasis, poly)\n\u001b[0;32m---> 78\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolyfuncs\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (symmetrize, horner, interpolate,\n\u001b[1;32m     79\u001b[0m         rational_interpolate, viete)\n\u001b[1;32m     81\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mrationaltools\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m together\n\u001b[1;32m     83\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolyerrors\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (BasePolynomialError, ExactQuotientFailed,\n\u001b[1;32m     84\u001b[0m         PolynomialDivisionFailed, OperationNotSupported, HeuristicGCDFailed,\n\u001b[1;32m     85\u001b[0m         HomomorphismFailed, IsomorphismFailed, ExtraneousFactors,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     90\u001b[0m         MultivariatePolynomialError, PolificationFailed, OptionError,\n\u001b[1;32m     91\u001b[0m         FlagError)\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/polys/polyfuncs.py:10\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolyoptions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m allowed_flags, build_options\n\u001b[1;32m      9\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolytools\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m poly_from_expr, Poly\n\u001b[0;32m---> 10\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mspecialpolys\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (\n\u001b[1;32m     11\u001b[0m     symmetric_poly, interpolating_poly)\n\u001b[1;32m     12\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mrings\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m sring\n\u001b[1;32m     13\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutilities\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m numbered_symbols, take, public\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/polys/specialpolys.py:298\u001b[0m\n\u001b[1;32m    294\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m dmp_mul(f, h, n, K), dmp_mul(g, h, n, K), h\n\u001b[1;32m    296\u001b[0m \u001b[38;5;66;03m# A few useful polynomials from Wang's paper ('78).\u001b[39;00m\n\u001b[0;32m--> 298\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mrings\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m ring\n\u001b[1;32m    300\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_f_0\u001b[39m():\n\u001b[1;32m    301\u001b[0m     R, x, y, z \u001b[38;5;241m=\u001b[39m ring(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mx,y,z\u001b[39m\u001b[38;5;124m\"\u001b[39m, ZZ)\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/polys/rings.py:30\u001b[0m\n\u001b[1;32m     26\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolyoptions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (Domain \u001b[38;5;28;01mas\u001b[39;00m DomainOpt,\n\u001b[1;32m     27\u001b[0m                                      Order \u001b[38;5;28;01mas\u001b[39;00m OrderOpt, build_options)\n\u001b[1;32m     28\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolys\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpolyutils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (expr_from_dict, _dict_reorder,\n\u001b[1;32m     29\u001b[0m                                    _parallel_dict_from_expr)\n\u001b[0;32m---> 30\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mprinting\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdefaults\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m DefaultPrinting\n\u001b[1;32m     31\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutilities\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m public, subsets\n\u001b[1;32m     32\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutilities\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01miterables\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m is_sequence\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/printing/__init__.py:5\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;124;03m\"\"\"Printing subsystem\"\"\"\u001b[39;00m\n\u001b[1;32m      3\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpretty\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m pager_print, pretty, pretty_print, pprint, pprint_use_unicode, pprint_try_use_unicode\n\u001b[0;32m----> 5\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlatex\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m latex, print_latex, multiline_latex\n\u001b[1;32m      7\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mmathml\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m mathml, print_mathml\n\u001b[1;32m      9\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m python, print_python\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/printing/latex.py:18\u001b[0m\n\u001b[1;32m     16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01msympify\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m SympifyError\n\u001b[1;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlogic\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mboolalg\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m true, BooleanTrue, BooleanFalse\n\u001b[0;32m---> 18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtensor\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01marray\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m NDimArray\n\u001b[1;32m     20\u001b[0m \u001b[38;5;66;03m# sympy.printing imports\u001b[39;00m\n\u001b[1;32m     21\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mprinting\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mprecedence\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m precedence_traditional\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/tensor/__init__.py:4\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;124;03m\"\"\"A module to manipulate symbolic objects with indices including tensors\u001b[39;00m\n\u001b[1;32m      2\u001b[0m \n\u001b[1;32m      3\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m----> 4\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mindexed\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m IndexedBase, Idx, Indexed\n\u001b[1;32m      5\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mindex_methods\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m get_contraction_structure, get_indices\n\u001b[1;32m      6\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m shape\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/tensor/indexed.py:114\u001b[0m\n\u001b[1;32m    112\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlogic\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m fuzzy_bool, fuzzy_not\n\u001b[1;32m    113\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01msympify\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m _sympify\n\u001b[0;32m--> 114\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mspecial\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtensor_functions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m KroneckerDelta\n\u001b[1;32m    115\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mmultipledispatch\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m dispatch\n\u001b[1;32m    116\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutilities\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01miterables\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m is_sequence, NotIterable\n",
-      "File \u001b[0;32m~/anaconda3/envs/peft/lib/python3.9/site-packages/sympy/functions/__init__.py:21\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01melementary\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtrigonometric\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (sin, cos, tan,\n\u001b[1;32m     18\u001b[0m         sec, csc, cot, sinc, asin, acos, atan, asec, acsc, acot, atan2)\n\u001b[1;32m     19\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01melementary\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mexponential\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (exp_polar, exp, log,\n\u001b[1;32m     20\u001b[0m         LambertW)\n\u001b[0;32m---> 21\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01melementary\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mhyperbolic\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (sinh, cosh, tanh, coth,\n\u001b[1;32m     22\u001b[0m         sech, csch, asinh, acosh, atanh, acoth, asech, acsch)\n\u001b[1;32m     23\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01melementary\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mintegers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m floor, ceiling, frac\n\u001b[1;32m     24\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msympy\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mfunctions\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01melementary\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpiecewise\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m (Piecewise, piecewise_fold,\n\u001b[1;32m     25\u001b[0m                                                   piecewise_exclusive)\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap>:1007\u001b[0m, in \u001b[0;36m_find_and_load\u001b[0;34m(name, import_)\u001b[0m\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap>:986\u001b[0m, in \u001b[0;36m_find_and_load_unlocked\u001b[0;34m(name, import_)\u001b[0m\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap>:680\u001b[0m, in \u001b[0;36m_load_unlocked\u001b[0;34m(spec)\u001b[0m\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap_external>:846\u001b[0m, in \u001b[0;36mexec_module\u001b[0;34m(self, module)\u001b[0m\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap_external>:978\u001b[0m, in \u001b[0;36mget_code\u001b[0;34m(self, fullname)\u001b[0m\n",
-      "File \u001b[0;32m<frozen importlib._bootstrap_external>:647\u001b[0m, in \u001b[0;36m_compile_bytecode\u001b[0;34m(data, name, bytecode_path, source_path)\u001b[0m\n",
-      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import os\n",
     "\n",
@@ -69,10 +26,9 @@
     "from peft import get_peft_model, PromptTuningInit, PromptTuningConfig, TaskType\n",
     "from datasets import load_dataset\n",
     "\n",
-    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"0\"\n",
     "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
     "\n",
-    "device = \"cuda\"\n",
+    "device = torch.accelerator.current_accelerator().type if hasattr(torch, \"accelerator\") else \"cuda\"\n",
     "model_name_or_path = \"t5-large\"\n",
     "tokenizer_name_or_path = \"t5-large\"\n",
     "\n",
@@ -100,7 +56,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "trainable params: 40960 || all params: 737709056 || trainable%: 0.005552324411210698\n"
+      "trainable params: 40,960 || all params: 737,709,056 || trainable%: 0.0056\n"
      ]
     },
     {
@@ -289,27 +245,14 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Found cached dataset financial_phrasebank (/data/proxem/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141)\n"
+      "Using the latest cached version of the dataset since financial_phrasebank couldn't be found on the Hugging Face Hub\n",
+      "Found the latest cached dataset configuration 'sentences_allagree' at /root/.cache/huggingface/datasets/financial_phrasebank/sentences_allagree/1.0.0/550bde12e6c30e2674da973a55f57edde5181d53f5a5a34c1531c53f93b7e141 (last modified on Thu Jul 31 06:43:45 2025).\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "d3a799c64a2c43258dc6166c90e2e49f",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/1 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "79ef90cbad2f4c2088f01102cadb8a3b",
        "version_major": 2,
        "version_minor": 0
       },
@@ -323,7 +266,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "0f5b177b658646cfa90b3a2801138807",
        "version_major": 2,
        "version_minor": 0
       },
@@ -337,9 +280,9 @@
     {
      "data": {
       "text/plain": [
-       "{'sentence': 'The price of the 10,000 kroon par value bonds was 9663,51 kroons in the primary issue .',\n",
-       " 'label': 1,\n",
-       " 'text_label': 'neutral'}"
+       "{'sentence': 'This new partnership agreement represents a significant milestone for both parties .',\n",
+       " 'label': 2,\n",
+       " 'text_label': 'positive'}"
       ]
      },
      "execution_count": 3,
@@ -375,22 +318,10 @@
     }
    },
    "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/models/t5/tokenization_t5_fast.py:155: FutureWarning: This tokenizer was incorrectly instantiated with a model max length of 512 which will be corrected in Transformers v5.\n",
-      "For now, this behavior is kept to avoid breaking backwards compatibility when padding/encoding with `truncation is True`.\n",
-      "- Be aware that you SHOULD NOT rely on t5-large automatically truncating your input to 512 when padding/encoding.\n",
-      "- If you want to encode/pad to sequences longer than 512 you can either instantiate this tokenizer with `model_max_length` or pass `max_length` when encoding/padding.\n",
-      "- To avoid this warning, please instantiate this tokenizer with `model_max_length` set to your preferred value.\n",
-      "  warnings.warn(\n"
-     ]
-    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "0a5f7b5967704fab97f11bc07813625c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -404,7 +335,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "",
+       "model_id": "1ff9578c074e4736a8812f6ffc8138b5",
        "version_major": 2,
        "version_minor": 0
       },
@@ -447,7 +378,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 5,
    "id": "6b3a4090",
    "metadata": {
     "ExecuteTime": {
@@ -460,8 +391,42 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/udir/tschilla/anaconda3/envs/peft/lib/python3.9/site-packages/transformers/optimization.py:407: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning\n",
-      "  warnings.warn(\n"
+      "[W731 07:06:51.135038656 OperatorEntry.cpp:217] Warning: Warning only once for all operators,  other operators may also be overridden.\n",
+      "  Overriding a previously registered kernel for the same operator and the same dispatch key\n",
+      "  operator: aten::geometric_(Tensor(a!) self, float p, *, Generator? generator=None) -> Tensor(a!)\n",
+      "    registered at /pytorch/build/aten/src/ATen/RegisterSchema.cpp:6\n",
+      "  dispatch key: XPU\n",
+      "  previous kernel: registered at /pytorch/aten/src/ATen/VmapModeRegistrations.cpp:37\n",
+      "       new kernel: registered at /build/intel-pytorch-extension/build/Release/csrc/gpu/csrc/gpu/xpu/ATen/RegisterXPU_0.cpp:172 (function operator())\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[2025-07-31 07:06:51,984] [INFO] [real_accelerator.py:254:get_accelerator] Setting ds_accelerator to xpu (auto detect)\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/usr/bin/ld: cannot find -laio: No such file or directory\n",
+      "collect2: error: ld returned 1 exit status\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[2025-07-31 07:06:52,955] [INFO] [logging.py:107:log_dist] [Rank -1] [TorchCheckpointEngine] Initialized with serialization = False\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "No label_names provided for model class `PeftModelForSeq2SeqLM`. Since `PeftModel` hides base models input arguments, if label_names is not given, label_names can't be set automatically within `Trainer`. Note that empty label_names list will be used instead.\n"
      ]
     },
     {
@@ -471,7 +436,7 @@
        "    <div>\n",
        "      \n",
        "      <progress value='1275' max='1275' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
-       "      [1275/1275 02:52, Epoch 5/5]\n",
+       "      [1275/1275 03:31, Epoch 5/5]\n",
        "    </div>\n",
        "    <table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
@@ -485,33 +450,33 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <td>1</td>\n",
-       "      <td>4.784800</td>\n",
-       "      <td>0.576933</td>\n",
-       "      <td>0.559471</td>\n",
+       "      <td>2.169900</td>\n",
+       "      <td>0.507156</td>\n",
+       "      <td>0.621145</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <td>2</td>\n",
-       "      <td>0.648200</td>\n",
-       "      <td>0.437575</td>\n",
-       "      <td>0.577093</td>\n",
+       "      <td>0.537700</td>\n",
+       "      <td>0.430996</td>\n",
+       "      <td>0.651982</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <td>3</td>\n",
-       "      <td>0.536200</td>\n",
-       "      <td>0.397857</td>\n",
-       "      <td>0.625551</td>\n",
+       "      <td>0.482200</td>\n",
+       "      <td>0.426718</td>\n",
+       "      <td>0.696035</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <td>4</td>\n",
-       "      <td>0.472200</td>\n",
-       "      <td>0.373160</td>\n",
-       "      <td>0.643172</td>\n",
+       "      <td>0.459700</td>\n",
+       "      <td>0.470894</td>\n",
+       "      <td>0.682819</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <td>5</td>\n",
-       "      <td>0.452500</td>\n",
-       "      <td>0.370234</td>\n",
-       "      <td>0.656388</td>\n",
+       "      <td>0.436000</td>\n",
+       "      <td>0.409604</td>\n",
+       "      <td>0.718062</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table><p>"
@@ -526,7 +491,7 @@
     {
      "data": {
       "text/plain": [
-       "TrainOutput(global_step=1275, training_loss=1.3787811279296875, metrics={'train_runtime': 173.3699, 'train_samples_per_second': 58.747, 'train_steps_per_second': 7.354, 'total_flos': 344546979840000.0, 'train_loss': 1.3787811279296875, 'epoch': 5.0})"
+       "TrainOutput(global_step=1275, training_loss=0.8170911183076747, metrics={'train_runtime': 213.5513, 'train_samples_per_second': 47.693, 'train_steps_per_second': 5.97, 'total_flos': 344546979840000.0, 'train_loss': 0.8170911183076747, 'epoch': 5.0})"
       ]
      },
      "execution_count": 5,
@@ -596,7 +561,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "id": "bd20cd4c",
    "metadata": {
     "ExecuteTime": {
@@ -604,17 +569,9 @@
      "start_time": "2023-05-30T09:53:15.059304Z"
     }
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "164K\tt5-large_PROMPT_TUNING_SEQ_2_SEQ_LM/adapter_model.bin\r\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "ckpt = f\"{peft_model_id}/adapter_model.bin\"\n",
+    "ckpt = f\"{peft_model_id}/adapter_model.safetensors\"\n",
     "!du -h $ckpt"
    ]
   },
@@ -654,13 +611,11 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Aspocomp Group , headquartered in Helsinki , Finland , develops interconnection solutions for the electronics industry .\n",
-      "{'input_ids': tensor([[   71,  7990,  7699,  1531,     3,     6,     3, 27630,    16, 29763,\n",
-      "             3,     6, 16458,     3,     6,  1344,     7,  1413, 28102,  1275,\n",
-      "            21,     8, 12800,   681,     3,     5,     1]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
-      "         1, 1, 1]])}\n",
-      "tensor([[   0, 7163,    1]])\n",
-      "['neutral']\n"
+      "EPS grew to 0.04 eur from 0.02 eur .\n",
+      "{'input_ids': tensor([[    3, 24935,     3,  4774,    12,  4097,  6348,     3,  1238,    45,\n",
+      "          4097,  4305,     3,  1238,     3,     5,     1]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]])}\n",
+      "tensor([[   0, 1465,    1]])\n",
+      "['positive']\n"
      ]
     }
    ],
@@ -688,9 +643,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "peft",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
-   "name": "peft"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -702,7 +657,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.16"
+   "version": "3.11.13"
   },
   "toc": {
    "base_numbering": 1,
diff --git a/examples/conditional_generation/requirements.txt b/examples/conditional_generation/requirements.txt
index 8dcf534a99..9571ec3501 100644
--- a/examples/conditional_generation/requirements.txt
+++ b/examples/conditional_generation/requirements.txt
@@ -3,4 +3,6 @@ accelerate
 evaluate
 deepspeed
 tqdm
-datasets
\ No newline at end of file
+datasets
+safetensors
+scikit-learn
\ No newline at end of file