llm-course

Форк
0
/
Fine_tune_LLMs_with_Axolotl.ipynb 
1555 строк · 76.6 Кб
1
{
2
  "cells": [
3
    {
4
      "cell_type": "markdown",
5
      "metadata": {
6
        "id": "view-in-github",
7
        "colab_type": "text"
8
      },
9
      "source": [
10
        "<a href=\"https://colab.research.google.com/github/mlabonne/llm-course/blob/main/Fine_tune_LLMs_with_Axolotl.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
11
      ]
12
    },
13
    {
14
      "cell_type": "markdown",
15
      "source": [
16
        "# Fine-tune LLMs with Axolotl\n",
17
        "\n",
18
        "> 🗣️ [Large Language Model Course](https://github.com/mlabonne/llm-course)\n",
19
        "\n",
20
        "❤️ Created by [@maximelabonne](https://twitter.com/maximelabonne), based on [Giorgio](https://github.com/g-i-o-r-g-i-o)'s notebook and Axolotl's [example](https://github.com/OpenAccess-AI-Collective/axolotl/blob/main/examples/colab-notebooks/colab-axolotl-example.ipynb)."
21
      ],
22
      "metadata": {
23
        "id": "mL-BPZyZ0gtV"
24
      }
25
    },
26
    {
27
      "cell_type": "code",
28
      "source": [
29
        "!git clone -q https://github.com/OpenAccess-AI-Collective/axolotl\n",
30
        "%cd axolotl\n",
31
        "!pip install -qqq packaging huggingface_hub --progress-bar off\n",
32
        "!pip install -qqq -e '.[flash-attn,deepspeed]' --progress-bar off"
33
      ],
34
      "metadata": {
35
        "id": "BI6B0Bfe0hxr"
36
      },
37
      "execution_count": null,
38
      "outputs": []
39
    },
40
    {
41
      "cell_type": "code",
42
      "source": [
43
        "import yaml\n",
44
        "\n",
45
        "new_model = \"mlabonne/TinyAlpaca\"\n",
46
        "yaml_string = \"\"\"\n",
47
        "base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T\n",
48
        "model_type: LlamaForCausalLM\n",
49
        "tokenizer_type: LlamaTokenizer\n",
50
        "is_llama_derived_model: true\n",
51
        "\n",
52
        "load_in_8bit: false\n",
53
        "load_in_4bit: true\n",
54
        "strict: false\n",
55
        "\n",
56
        "datasets:\n",
57
        "  - path: mhenrichsen/alpaca_2k_test\n",
58
        "    type: alpaca\n",
59
        "dataset_prepared_path:\n",
60
        "val_set_size: 0.05\n",
61
        "output_dir: ./qlora-out\n",
62
        "\n",
63
        "adapter: qlora\n",
64
        "lora_model_dir:\n",
65
        "\n",
66
        "sequence_len: 1096\n",
67
        "sample_packing: true\n",
68
        "pad_to_sequence_len: true\n",
69
        "\n",
70
        "lora_r: 32\n",
71
        "lora_alpha: 16\n",
72
        "lora_dropout: 0.05\n",
73
        "lora_target_modules:\n",
74
        "lora_target_linear: true\n",
75
        "lora_fan_in_fan_out:\n",
76
        "\n",
77
        "wandb_project:\n",
78
        "wandb_entity:\n",
79
        "wandb_watch:\n",
80
        "wandb_name:\n",
81
        "wandb_log_model:\n",
82
        "\n",
83
        "mlflow_experiment_name: colab-example\n",
84
        "\n",
85
        "gradient_accumulation_steps: 1\n",
86
        "micro_batch_size: 1\n",
87
        "num_epochs: 4\n",
88
        "max_steps: 20\n",
89
        "optimizer: paged_adamw_32bit\n",
90
        "lr_scheduler: cosine\n",
91
        "learning_rate: 0.0002\n",
92
        "\n",
93
        "train_on_inputs: false\n",
94
        "group_by_length: false\n",
95
        "bf16: false\n",
96
        "fp16: true\n",
97
        "tf32: false\n",
98
        "\n",
99
        "gradient_checkpointing: true\n",
100
        "early_stopping_patience:\n",
101
        "resume_from_checkpoint:\n",
102
        "local_rank:\n",
103
        "logging_steps: 1\n",
104
        "xformers_attention:\n",
105
        "flash_attention: false\n",
106
        "\n",
107
        "warmup_steps: 10\n",
108
        "evals_per_epoch:\n",
109
        "saves_per_epoch:\n",
110
        "debug:\n",
111
        "deepspeed:\n",
112
        "weight_decay: 0.0\n",
113
        "fsdp:\n",
114
        "fsdp_config:\n",
115
        "special_tokens:\n",
116
        "\n",
117
        "\"\"\"\n",
118
        "\n",
119
        "# Convert the YAML string to a Python dictionary\n",
120
        "yaml_dict = yaml.safe_load(yaml_string)\n",
121
        "\n",
122
        "# Specify your file path\n",
123
        "yaml_file = 'config.yaml'\n",
124
        "\n",
125
        "# Write the YAML file\n",
126
        "with open(yaml_file, 'w') as file:\n",
127
        "    yaml.dump(yaml_dict, file)"
128
      ],
129
      "metadata": {
130
        "id": "70zJf1hi0huQ"
131
      },
132
      "execution_count": 5,
133
      "outputs": []
134
    },
135
    {
136
      "cell_type": "code",
137
      "execution_count": 6,
138
      "metadata": {
139
        "colab": {
140
          "base_uri": "https://localhost:8080/"
141
        },
142
        "id": "ffcc0IB9Sr-z",
143
        "outputId": "f64291b0-0867-440b-b7b9-683fa66e3a08"
144
      },
145
      "outputs": [
146
        {
147
          "output_type": "stream",
148
          "name": "stdout",
149
          "text": [
150
            "The following values were not passed to `accelerate launch` and had defaults used instead:\n",
151
            "\t`--num_processes` was set to a value of `1`\n",
152
            "\t`--num_machines` was set to a value of `1`\n",
153
            "\t`--mixed_precision` was set to a value of `'no'`\n",
154
            "\t`--dynamo_backend` was set to a value of `'no'`\n",
155
            "To avoid this warning pass in values for each of the problematic parameters or run `accelerate config`.\n",
156
            "2024-01-27 22:05:58.426793: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
157
            "2024-01-27 22:05:58.426851: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
158
            "2024-01-27 22:05:58.428152: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
159
            "2024-01-27 22:05:59.454203: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
160
            "[2024-01-27 22:06:00,824] [INFO] [datasets.<module>:58] [PID:2797] PyTorch version 2.1.2 available.\n",
161
            "[2024-01-27 22:06:00,825] [INFO] [datasets.<module>:95] [PID:2797] TensorFlow version 2.15.0 available.\n",
162
            "[2024-01-27 22:06:00,827] [INFO] [datasets.<module>:108] [PID:2797] JAX version 0.4.23 available.\n",
163
            "[2024-01-27 22:06:01,844] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)\n",
164
            "config.json: 100% 560/560 [00:00<00:00, 3.84MB/s]\n",
165
            "[2024-01-27 22:06:03,638] [INFO] [axolotl.normalize_config:170] [PID:2797] [RANK:0] GPU memory usage baseline: 0.000GB (+0.255GB misc)\u001b[39m\n",
166
            "                                 dP            dP   dP \n",
167
            "                                 88            88   88 \n",
168
            "      .d8888b. dP.  .dP .d8888b. 88 .d8888b. d8888P 88 \n",
169
            "      88'  `88  `8bd8'  88'  `88 88 88'  `88   88   88 \n",
170
            "      88.  .88  .d88b.  88.  .88 88 88.  .88   88   88 \n",
171
            "      `88888P8 dP'  `dP `88888P' dP `88888P'   dP   dP \n",
172
            "                                                       \n",
173
            "                                                       \n",
174
            "\n",
175
            "\u001b[33m[2024-01-27 22:06:03,642] [WARNING] [axolotl.scripts.check_user_token:382] [PID:2797] [RANK:0] Error verifying HuggingFace token. Remember to log in using `huggingface-cli login` and get your access token from https://huggingface.co/settings/tokens if you want to use gated models or datasets.\u001b[39m\n",
176
            "tokenizer_config.json: 100% 776/776 [00:00<00:00, 4.70MB/s]\n",
177
            "tokenizer.model: 100% 500k/500k [00:00<00:00, 1.77MB/s]\n",
178
            "special_tokens_map.json: 100% 414/414 [00:00<00:00, 1.73MB/s]\n",
179
            "tokenizer.json: 100% 1.84M/1.84M [00:00<00:00, 4.52MB/s]\n",
180
            "[2024-01-27 22:06:05,850] [DEBUG] [axolotl.load_tokenizer:216] [PID:2797] [RANK:0] EOS: 2 / </s>\u001b[39m\n",
181
            "[2024-01-27 22:06:05,850] [DEBUG] [axolotl.load_tokenizer:217] [PID:2797] [RANK:0] BOS: 1 / <s>\u001b[39m\n",
182
            "[2024-01-27 22:06:05,850] [DEBUG] [axolotl.load_tokenizer:218] [PID:2797] [RANK:0] PAD: 2 / </s>\u001b[39m\n",
183
            "[2024-01-27 22:06:05,850] [DEBUG] [axolotl.load_tokenizer:219] [PID:2797] [RANK:0] UNK: 0 / <unk>\u001b[39m\n",
184
            "[2024-01-27 22:06:05,850] [INFO] [axolotl.load_tokenizer:230] [PID:2797] [RANK:0] No Chat template selected. Consider adding a chat template for easier inference.\u001b[39m\n",
185
            "[2024-01-27 22:06:05,850] [INFO] [axolotl.load_tokenized_prepared_datasets:182] [PID:2797] [RANK:0] Unable to find prepared dataset in last_run_prepared/c0112363192ff19da1e486577d4bf28b\u001b[39m\n",
186
            "[2024-01-27 22:06:05,850] [INFO] [axolotl.load_tokenized_prepared_datasets:183] [PID:2797] [RANK:0] Loading raw datasets...\u001b[39m\n",
187
            "\u001b[33m[2024-01-27 22:06:05,850] [WARNING] [axolotl.load_tokenized_prepared_datasets:185] [PID:2797] [RANK:0] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset.\u001b[39m\n",
188
            "[2024-01-27 22:06:05,850] [INFO] [axolotl.load_tokenized_prepared_datasets:192] [PID:2797] [RANK:0] No seed provided, using default seed of 42\u001b[39m\n",
189
            "Downloading readme: 100% 28.0/28.0 [00:00<00:00, 219kB/s]\n",
190
            "Downloading data: 100% 1.76M/1.76M [00:00<00:00, 2.45MB/s]\n",
191
            "Generating train split: 2000 examples [00:00, 37778.35 examples/s]\n",
192
            "Tokenizing Prompts (num_proc=2): 100% 2000/2000 [00:05<00:00, 352.09 examples/s]\n",
193
            "[2024-01-27 22:06:17,586] [INFO] [axolotl.load_tokenized_prepared_datasets:395] [PID:2797] [RANK:0] merging datasets\u001b[39m\n",
194
            "Dropping Long Sequences (num_proc=2): 100% 2000/2000 [00:00<00:00, 2927.43 examples/s]\n",
195
            "Add position_id column (Sample Packing) (num_proc=2): 100% 2000/2000 [00:00<00:00, 3089.00 examples/s]\n",
196
            "[2024-01-27 22:06:19,041] [INFO] [axolotl.load_tokenized_prepared_datasets:405] [PID:2797] [RANK:0] Saving merged prepared dataset to disk... last_run_prepared/c0112363192ff19da1e486577d4bf28b\u001b[39m\n",
197
            "Saving the dataset (1/1 shards): 100% 2000/2000 [00:00<00:00, 142755.66 examples/s]\n",
198
            "[2024-01-27 22:06:19,067] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] total_num_tokens: 22777\u001b[39m\n",
199
            "[2024-01-27 22:06:19,069] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] `total_supervised_tokens: 16719`\u001b[39m\n",
200
            "[2024-01-27 22:06:24,965] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 1.0 total_num_tokens per device: 22777\u001b[39m\n",
201
            "[2024-01-27 22:06:24,965] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] data_loader_len: 19\u001b[39m\n",
202
            "[2024-01-27 22:06:24,966] [INFO] [axolotl.log:61] [PID:2797] [RANK:0] sample_packing_eff_est across ranks: [0.8659139294403893]\u001b[39m\n",
203
            "[2024-01-27 22:06:24,966] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] sample_packing_eff_est: None\u001b[39m\n",
204
            "[2024-01-27 22:06:24,966] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] total_num_steps: 76\u001b[39m\n",
205
            "[2024-01-27 22:06:24,970] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] total_num_tokens: 414041\u001b[39m\n",
206
            "[2024-01-27 22:06:25,001] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] `total_supervised_tokens: 294246`\u001b[39m\n",
207
            "[2024-01-27 22:06:25,009] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 1.0 total_num_tokens per device: 414041\u001b[39m\n",
208
            "[2024-01-27 22:06:25,009] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] data_loader_len: 372\u001b[39m\n",
209
            "[2024-01-27 22:06:25,009] [INFO] [axolotl.log:61] [PID:2797] [RANK:0] sample_packing_eff_est across ranks: [0.8624991667499917]\u001b[39m\n",
210
            "[2024-01-27 22:06:25,009] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] sample_packing_eff_est: 0.87\u001b[39m\n",
211
            "[2024-01-27 22:06:25,009] [DEBUG] [axolotl.log:61] [PID:2797] [RANK:0] total_num_steps: 1488\u001b[39m\n",
212
            "[2024-01-27 22:06:25,009] [INFO] [axolotl.prepare_dataset:115] [PID:2797] [RANK:0] Maximum number of steps set at 20\u001b[39m\n",
213
            "[2024-01-27 22:06:25,010] [DEBUG] [axolotl.train.log:61] [PID:2797] [RANK:0] loading tokenizer... TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T\u001b[39m\n",
214
            "[2024-01-27 22:06:25,574] [DEBUG] [axolotl.load_tokenizer:216] [PID:2797] [RANK:0] EOS: 2 / </s>\u001b[39m\n",
215
            "[2024-01-27 22:06:25,575] [DEBUG] [axolotl.load_tokenizer:217] [PID:2797] [RANK:0] BOS: 1 / <s>\u001b[39m\n",
216
            "[2024-01-27 22:06:25,575] [DEBUG] [axolotl.load_tokenizer:218] [PID:2797] [RANK:0] PAD: 2 / </s>\u001b[39m\n",
217
            "[2024-01-27 22:06:25,575] [DEBUG] [axolotl.load_tokenizer:219] [PID:2797] [RANK:0] UNK: 0 / <unk>\u001b[39m\n",
218
            "[2024-01-27 22:06:25,575] [INFO] [axolotl.load_tokenizer:230] [PID:2797] [RANK:0] No Chat template selected. Consider adding a chat template for easier inference.\u001b[39m\n",
219
            "[2024-01-27 22:06:25,575] [DEBUG] [axolotl.train.log:61] [PID:2797] [RANK:0] loading model and peft_config...\u001b[39m\n",
220
            "[2024-01-27 22:06:25,678] [INFO] [axolotl.load_model:372] [PID:2797] [RANK:0] patching _expand_mask\u001b[39m\n",
221
            "model.safetensors: 100% 4.40G/4.40G [01:42<00:00, 42.8MB/s]\n",
222
            "generation_config.json: 100% 129/129 [00:00<00:00, 758kB/s]\n",
223
            "[2024-01-27 22:08:13,799] [INFO] [axolotl.load_model:641] [PID:2797] [RANK:0] GPU memory usage after model load: 0.753GB (+0.022GB cache, +0.368GB misc)\u001b[39m\n",
224
            "[2024-01-27 22:08:13,825] [INFO] [axolotl.load_model:673] [PID:2797] [RANK:0] converting PEFT model w/ prepare_model_for_kbit_training\u001b[39m\n",
225
            "[2024-01-27 22:08:13,829] [INFO] [axolotl.load_model:685] [PID:2797] [RANK:0] converting modules to torch.float16 for flash attention\u001b[39m\n",
226
            "[2024-01-27 22:08:13,832] [INFO] [axolotl.load_lora:797] [PID:2797] [RANK:0] found linear modules: ['v_proj', 'k_proj', 'gate_proj', 'up_proj', 'o_proj', 'q_proj', 'down_proj']\u001b[39m\n",
227
            "trainable params: 25,231,360 || all params: 1,125,279,744 || trainable%: 2.2422299996542017\n",
228
            "[2024-01-27 22:08:14,263] [INFO] [axolotl.load_model:722] [PID:2797] [RANK:0] GPU memory usage after adapters: 0.847GB (+0.514GB cache, +0.368GB misc)\u001b[39m\n",
229
            "[2024-01-27 22:08:14,273] [INFO] [axolotl.train.log:61] [PID:2797] [RANK:0] Pre-saving adapter config to ./qlora-out\u001b[39m\n",
230
            "[2024-01-27 22:08:14,280] [INFO] [axolotl.train.log:61] [PID:2797] [RANK:0] Starting trainer...\u001b[39m\n",
231
            "[2024-01-27 22:08:14,686] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 414041\u001b[39m\n",
232
            "[2024-01-27 22:08:14,688] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 414041\u001b[39m\n",
233
            "[2024-01-27 22:08:14,850] [INFO] [axolotl.callbacks.on_train_begin:601] [PID:2797] [RANK:0] The Axolotl config has been saved to the MLflow artifacts.\u001b[39m\n",
234
            "  0% 0/20 [00:00<?, ?it/s][2024-01-27 22:08:14,853] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 414041\u001b[39m\n",
235
            "{'loss': 1.6463, 'learning_rate': 2e-05, 'epoch': 0.0}\n",
236
            "  5% 1/20 [00:02<00:45,  2.41s/it][2024-01-27 22:08:18,019] [INFO] [axolotl.callbacks.on_step_end:125] [PID:2797] [RANK:0] GPU memory usage while training: 0.963GB (+0.805GB cache, +0.638GB misc)\u001b[39m\n",
237
            "{'loss': 4.8805, 'learning_rate': 4e-05, 'epoch': 0.0}\n",
238
            "{'loss': 4.4957, 'learning_rate': 6e-05, 'epoch': 0.01}\n",
239
            "{'loss': 4.4627, 'learning_rate': 8e-05, 'epoch': 0.01}\n",
240
            "{'loss': 4.8849, 'learning_rate': 0.0001, 'epoch': 0.01}\n",
241
            "{'loss': 5.0871, 'learning_rate': 0.00012, 'epoch': 0.01}\n",
242
            "{'loss': 3.0099, 'learning_rate': 0.00014, 'epoch': 0.02}\n",
243
            "{'loss': 4.4522, 'learning_rate': 0.00016, 'epoch': 0.02}\n",
244
            "{'loss': 2.2085, 'learning_rate': 0.00018, 'epoch': 0.02}\n",
245
            "{'loss': 4.4377, 'learning_rate': 0.0002, 'epoch': 0.02}\n",
246
            "{'loss': 3.7593, 'learning_rate': 0.00019510565162951537, 'epoch': 0.03}\n",
247
            "{'loss': 3.5486, 'learning_rate': 0.00018090169943749476, 'epoch': 0.03}\n",
248
            "{'loss': 3.7615, 'learning_rate': 0.00015877852522924732, 'epoch': 0.03}\n",
249
            "{'loss': 3.5778, 'learning_rate': 0.00013090169943749476, 'epoch': 0.03}\n",
250
            "{'loss': 3.3562, 'learning_rate': 0.0001, 'epoch': 0.04}\n",
251
            "{'loss': 3.7881, 'learning_rate': 6.909830056250527e-05, 'epoch': 0.04}\n",
252
            "{'loss': 3.4147, 'learning_rate': 4.12214747707527e-05, 'epoch': 0.04}\n",
253
            "{'loss': 3.7467, 'learning_rate': 1.9098300562505266e-05, 'epoch': 0.04}\n",
254
            "{'loss': 3.6867, 'learning_rate': 4.8943483704846475e-06, 'epoch': 0.04}\n",
255
            "{'loss': 4.0308, 'learning_rate': 0.0, 'epoch': 0.05}\n",
256
            "100% 20/20 [00:17<00:00,  1.26it/s][2024-01-27 22:08:32,098] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
257
            "[2024-01-27 22:08:32,232] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
258
            "[2024-01-27 22:08:32,232] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
259
            "\n",
260
            "  0% 0/22 [00:00<?, ?it/s]\u001b[A[2024-01-27 22:08:32,451] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
261
            "\n",
262
            "  9% 2/22 [00:00<00:02,  9.16it/s]\u001b[A[2024-01-27 22:08:32,669] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
263
            "\n",
264
            " 14% 3/22 [00:00<00:02,  6.46it/s]\u001b[A[2024-01-27 22:08:32,890] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
265
            "\n",
266
            " 18% 4/22 [00:00<00:03,  5.58it/s]\u001b[A[2024-01-27 22:08:33,119] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
267
            "\n",
268
            " 23% 5/22 [00:00<00:03,  5.08it/s]\u001b[A[2024-01-27 22:08:33,339] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
269
            "\n",
270
            " 27% 6/22 [00:01<00:03,  4.89it/s]\u001b[A[2024-01-27 22:08:33,559] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
271
            "\n",
272
            " 32% 7/22 [00:01<00:03,  4.78it/s]\u001b[A[2024-01-27 22:08:33,779] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
273
            "\n",
274
            " 36% 8/22 [00:01<00:02,  4.70it/s]\u001b[A[2024-01-27 22:08:34,004] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
275
            "\n",
276
            " 41% 9/22 [00:01<00:02,  4.62it/s]\u001b[A[2024-01-27 22:08:34,224] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
277
            "\n",
278
            " 45% 10/22 [00:01<00:02,  4.59it/s]\u001b[A[2024-01-27 22:08:34,447] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
279
            "\n",
280
            " 50% 11/22 [00:02<00:02,  4.56it/s]\u001b[A[2024-01-27 22:08:34,670] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
281
            "\n",
282
            " 55% 12/22 [00:02<00:02,  4.54it/s]\u001b[A[2024-01-27 22:08:34,889] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
283
            "\n",
284
            " 59% 13/22 [00:02<00:01,  4.55it/s]\u001b[A[2024-01-27 22:08:35,115] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
285
            "\n",
286
            " 64% 14/22 [00:02<00:01,  4.51it/s]\u001b[A[2024-01-27 22:08:35,337] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
287
            "\n",
288
            " 68% 15/22 [00:03<00:01,  4.51it/s]\u001b[A[2024-01-27 22:08:35,559] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
289
            "\n",
290
            " 73% 16/22 [00:03<00:01,  4.51it/s]\u001b[A[2024-01-27 22:08:35,783] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
291
            "\n",
292
            " 77% 17/22 [00:03<00:01,  4.49it/s]\u001b[A[2024-01-27 22:08:36,005] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
293
            "\n",
294
            " 82% 18/22 [00:03<00:00,  4.50it/s]\u001b[A[2024-01-27 22:08:36,228] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
295
            "\n",
296
            " 86% 19/22 [00:03<00:00,  4.50it/s]\u001b[A[2024-01-27 22:08:36,449] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
297
            "\n",
298
            " 91% 20/22 [00:04<00:00,  4.50it/s]\u001b[A[2024-01-27 22:08:36,676] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
299
            "\n",
300
            " 95% 21/22 [00:04<00:00,  4.47it/s]\u001b[A[2024-01-27 22:08:36,896] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
301
            "\n",
302
            "100% 22/22 [00:04<00:00,  4.50it/s]\u001b[A[2024-01-27 22:08:37,116] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
303
            "\n",
304
            "23it [00:04,  4.51it/s]            \u001b[A[2024-01-27 22:08:37,336] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
305
            "\n",
306
            "24it [00:05,  4.52it/s]\u001b[A[2024-01-27 22:08:37,556] [INFO] [axolotl.utils.samplers.multipack._len_est:178] [PID:2797] [RANK:0] packing_efficiency_estimate: 0.87 total_num_tokens per device: 22777\u001b[39m\n",
307
            "\n",
308
            "                                   \n",
309
            "\u001b[A{'eval_loss': 3.513343572616577, 'eval_runtime': 5.5482, 'eval_samples_per_second': 18.024, 'eval_steps_per_second': 18.024, 'epoch': 0.05}\n",
310
            "100% 20/20 [00:22<00:00,  1.26it/s]\n",
311
            "25it [00:05,  4.53it/s]\u001b[A\n",
312
            "{'train_runtime': 24.8459, 'train_samples_per_second': 0.805, 'train_steps_per_second': 0.805, 'train_loss': 3.8117882788181303, 'epoch': 0.05}\n",
313
            "100% 20/20 [00:24<00:00,  1.24s/it]\n",
314
            "[2024-01-27 22:08:39,580] [INFO] [axolotl.train.log:61] [PID:2797] [RANK:0] Training Completed!!! Saving pre-trained model to ./qlora-out\u001b[39m\n",
315
            "(PeftModelForCausalLM(   (base_model): LoraModel(     (model): LlamaForCausalLM(       (model): LlamaModel(         (embed_tokens): Embedding(32000, 2048)         (layers): ModuleList(           (0-21): 22 x LlamaDecoderLayer(             (self_attn): LlamaSdpaAttention(               (q_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=2048, out_features=2048, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=2048, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=2048, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (k_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=2048, out_features=256, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=2048, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=256, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (v_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=2048, out_features=256, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=2048, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=256, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (o_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=2048, out_features=2048, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=2048, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=2048, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (rotary_emb): LlamaRotaryEmbedding()             )             (mlp): LlamaMLP(               (gate_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=2048, out_features=5632, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=2048, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=5632, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (up_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=2048, out_features=5632, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=2048, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=5632, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (down_proj): lora.Linear4bit(                 (base_layer): Linear4bit(in_features=5632, out_features=2048, bias=False)                 (lora_dropout): ModuleDict(                   (default): Dropout(p=0.05, inplace=False)                 )                 (lora_A): ModuleDict(                   (default): Linear(in_features=5632, out_features=32, bias=False)                 )                 (lora_B): ModuleDict(                   (default): Linear(in_features=32, out_features=2048, bias=False)                 )                 (lora_embedding_A): ParameterDict()                 (lora_embedding_B): ParameterDict()               )               (act_fn): SiLU()             )             (input_layernorm): LlamaRMSNorm()             (post_attention_layernorm): LlamaRMSNorm()           )         )         (norm): LlamaRMSNorm()       )       (lm_head): Linear(in_features=2048, out_features=32000, bias=False)     )   ) ), LlamaTokenizer(name_or_path='TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T', vocab_size=32000, model_max_length=1000000000000000019884624838656, is_fast=False, padding_side='right', truncation_side='right', special_tokens={'bos_token': '<s>', 'eos_token': '</s>', 'unk_token': '<unk>', 'pad_token': '</s>'}, clean_up_tokenization_spaces=False),  added_tokens_decoder={ \t0: AddedToken(\"<unk>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), \t1: AddedToken(\"<s>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), \t2: AddedToken(\"</s>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True), })\n",
316
            "\u001b[0m"
317
          ]
318
        }
319
      ],
320
      "source": [
321
        "!accelerate launch -m axolotl.cli.train config.yaml"
322
      ]
323
    },
324
    {
325
      "cell_type": "code",
326
      "source": [
327
        "!python3 -m axolotl.cli.merge_lora config.yaml --lora_model_dir=\"./qlora-out\""
328
      ],
329
      "metadata": {
330
        "colab": {
331
          "base_uri": "https://localhost:8080/"
332
        },
333
        "id": "5GENnc9Z8v-P",
334
        "outputId": "762e6290-9877-485c-c084-5370c3b3d7f3"
335
      },
336
      "execution_count": 10,
337
      "outputs": [
338
        {
339
          "output_type": "stream",
340
          "name": "stdout",
341
          "text": [
342
            "mkdir: cannot create directory ‘final_model’: File exists\n",
343
            "2024-01-27 22:10:36.876247: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
344
            "2024-01-27 22:10:36.876317: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
345
            "2024-01-27 22:10:36.878122: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
346
            "2024-01-27 22:10:38.390922: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
347
            "[2024-01-27 22:10:39,447] [INFO] [datasets.<module>:58] [PID:4079] PyTorch version 2.1.2 available.\n",
348
            "[2024-01-27 22:10:39,448] [INFO] [datasets.<module>:95] [PID:4079] TensorFlow version 2.15.0 available.\n",
349
            "[2024-01-27 22:10:39,449] [INFO] [datasets.<module>:108] [PID:4079] JAX version 0.4.23 available.\n",
350
            "[2024-01-27 22:10:40,350] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)\n",
351
            "                                 dP            dP   dP \n",
352
            "                                 88            88   88 \n",
353
            "      .d8888b. dP.  .dP .d8888b. 88 .d8888b. d8888P 88 \n",
354
            "      88'  `88  `8bd8'  88'  `88 88 88'  `88   88   88 \n",
355
            "      88.  .88  .d88b.  88.  .88 88 88.  .88   88   88 \n",
356
            "      `88888P8 dP'  `dP `88888P' dP `88888P'   dP   dP \n",
357
            "                                                       \n",
358
            "                                                       \n",
359
            "\n",
360
            "[2024-01-27 22:10:41,928] [INFO] [axolotl.normalize_config:170] [PID:4079] [RANK:0] GPU memory usage baseline: 0.000GB (+0.255GB misc)\u001b[39m\n",
361
            "[2024-01-27 22:10:41,928] [INFO] [axolotl.common.cli.load_model_and_tokenizer:49] [PID:4079] [RANK:0] loading tokenizer... TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T\u001b[39m\n",
362
            "[2024-01-27 22:10:42,234] [DEBUG] [axolotl.load_tokenizer:216] [PID:4079] [RANK:0] EOS: 2 / </s>\u001b[39m\n",
363
            "[2024-01-27 22:10:42,234] [DEBUG] [axolotl.load_tokenizer:217] [PID:4079] [RANK:0] BOS: 1 / <s>\u001b[39m\n",
364
            "[2024-01-27 22:10:42,234] [DEBUG] [axolotl.load_tokenizer:218] [PID:4079] [RANK:0] PAD: 2 / </s>\u001b[39m\n",
365
            "[2024-01-27 22:10:42,234] [DEBUG] [axolotl.load_tokenizer:219] [PID:4079] [RANK:0] UNK: 0 / <unk>\u001b[39m\n",
366
            "[2024-01-27 22:10:42,235] [INFO] [axolotl.load_tokenizer:230] [PID:4079] [RANK:0] No Chat template selected. Consider adding a chat template for easier inference.\u001b[39m\n",
367
            "[2024-01-27 22:10:42,235] [INFO] [axolotl.common.cli.load_model_and_tokenizer:51] [PID:4079] [RANK:0] loading model and (optionally) peft_config...\u001b[39m\n",
368
            "[2024-01-27 22:10:42,320] [INFO] [axolotl.load_model:372] [PID:4079] [RANK:0] patching _expand_mask\u001b[39m\n",
369
            "[2024-01-27 22:10:45,770] [INFO] [axolotl.load_model:641] [PID:4079] [RANK:0] GPU memory usage after model load: 2.062GB (+0.087GB cache, +0.352GB misc)\u001b[39m\n",
370
            "[2024-01-27 22:10:45,787] [INFO] [axolotl.load_model:685] [PID:4079] [RANK:0] converting modules to torch.float16 for flash attention\u001b[39m\n",
371
            "[2024-01-27 22:10:45,791] [INFO] [axolotl.load_lora:797] [PID:4079] [RANK:0] found linear modules: ['q_proj', 'down_proj', 'v_proj', 'gate_proj', 'k_proj', 'up_proj', 'o_proj']\u001b[39m\n",
372
            "[2024-01-27 22:10:45,791] [DEBUG] [axolotl.load_lora:816] [PID:4079] [RANK:0] Loading pretained PEFT - LoRA\u001b[39m\n",
373
            "trainable params: 25,231,360 || all params: 1,125,279,744 || trainable%: 2.2422299996542017\n",
374
            "[2024-01-27 22:10:46,372] [INFO] [axolotl.load_model:722] [PID:4079] [RANK:0] GPU memory usage after adapters: 2.590GB (+1.060GB cache, +0.368GB misc)\u001b[39m\n",
375
            "[2024-01-27 22:10:46,372] [INFO] [axolotl.scripts.do_merge_lora:79] [PID:4079] [RANK:0] running merge of LoRA with base model\u001b[39m\n",
376
            "Unloading and merging model: 100% 468/468 [00:00<00:00, 4211.50it/s]\n",
377
            "[2024-01-27 22:10:46,490] [INFO] [axolotl.scripts.do_merge_lora:88] [PID:4079] [RANK:0] saving merged model to: qlora-out/merged\u001b[39m\n",
378
            "\u001b[0m"
379
          ]
380
        }
381
      ]
382
    },
383
    {
384
      "cell_type": "code",
385
      "source": [
386
        "from huggingface_hub import HfApi\n",
387
        "from google.colab import userdata\n",
388
        "\n",
389
        "new_model = \"mlabonne/TinyAlpaca\"\n",
390
        "\n",
391
        "# HF_TOKEN defined in the secrets tab in Google Colab\n",
392
        "api = HfApi()\n",
393
        "\n",
394
        "# Upload merge folder\n",
395
        "api.create_repo(\n",
396
        "    repo_id=new_model,\n",
397
        "    repo_type=\"model\",\n",
398
        "    exist_ok=True,\n",
399
        ")\n",
400
        "api.upload_folder(\n",
401
        "    repo_id=new_model,\n",
402
        "    folder_path=\"qlora-out/merged\",\n",
403
        ")"
404
      ],
405
      "metadata": {
406
        "colab": {
407
          "base_uri": "https://localhost:8080/",
408
          "height": 149,
409
          "referenced_widgets": [
410
            "76eab4e675b34a5486f28d16183658b6",
411
            "af7f38e810a64e5f98dd3f33672b8d57",
412
            "0e196628e7e64879bcc7184f67404127",
413
            "2ee0d001c05a4b89ab7a3cbea71c5960",
414
            "8a59f96ead334ba8b6848321fbde0811",
415
            "3db6a3ae89ec483783f97d77aef5e5a7",
416
            "f589d24de591428cb878405d1eb6034d",
417
            "e4520b22ca934c1fbe399131c021d0e8",
418
            "cb663cc62fc94108855d9b5509b573d8",
419
            "9926acc37862427d9fdd1ef37474d0c9",
420
            "215f84842ac5451a8f868e0dc3090a71",
421
            "0878f651a6b64ef495ebd5a2a9084b63",
422
            "eec99342bca04c1697ec43bc099c22c9",
423
            "cb7cc89f3edf4c6c8e38043ef0d697ba",
424
            "7a9497ed06004a82b2893e0f5ab03736",
425
            "f50ea1d3c5874926a6db2af3d9d795ab",
426
            "7851c929f5f64e31856d3793b963130e",
427
            "49f8123c3c5e43ababed7e09e97cb7d2",
428
            "c4dfec3f687241d6a69cf30252ada077",
429
            "a1c9ecf2bc124658a08f585b1bf309e6",
430
            "101d0b228b1943ff9c9d69fd6f4847ee",
431
            "b03c7fa74c774a39946636803e086218",
432
            "90726b0049c344c19fb5c9ef1dcf0628",
433
            "bebef38ee9e24586a7f894ffaf1ae43e",
434
            "9d90c5ea423e4586bc3bb303bcaa4b03",
435
            "d417d5bf00294e3f9cb7e78b5d83ee03",
436
            "68eac408135d4d048de89fa674b701da",
437
            "34c11d3b1a3441bf8533f7a6f057d0cf",
438
            "6270a81b7f5749d7a656c1edaaa63118",
439
            "e1de7a96086c4351bba99bbd4d905802",
440
            "6d28893d0a054d6aa2726ee94ee0d880",
441
            "91c9822b8ef14074bd360bb5d2998267",
442
            "8673edee8ab5430588daba8d422999eb"
443
          ]
444
        },
445
        "id": "NsvZw3B7-gfm",
446
        "outputId": "68c10dec-ee7a-4c0d-dd8d-c6631b53f4b3"
447
      },
448
      "execution_count": 13,
449
      "outputs": [
450
        {
451
          "output_type": "display_data",
452
          "data": {
453
            "text/plain": [
454
              "pytorch_model.bin:   0%|          | 0.00/2.20G [00:00<?, ?B/s]"
455
            ],
456
            "application/vnd.jupyter.widget-view+json": {
457
              "version_major": 2,
458
              "version_minor": 0,
459
              "model_id": "76eab4e675b34a5486f28d16183658b6"
460
            }
461
          },
462
          "metadata": {}
463
        },
464
        {
465
          "output_type": "display_data",
466
          "data": {
467
            "text/plain": [
468
              "Upload 2 LFS files:   0%|          | 0/2 [00:00<?, ?it/s]"
469
            ],
470
            "application/vnd.jupyter.widget-view+json": {
471
              "version_major": 2,
472
              "version_minor": 0,
473
              "model_id": "0878f651a6b64ef495ebd5a2a9084b63"
474
            }
475
          },
476
          "metadata": {}
477
        },
478
        {
479
          "output_type": "display_data",
480
          "data": {
481
            "text/plain": [
482
              "tokenizer.model:   0%|          | 0.00/500k [00:00<?, ?B/s]"
483
            ],
484
            "application/vnd.jupyter.widget-view+json": {
485
              "version_major": 2,
486
              "version_minor": 0,
487
              "model_id": "90726b0049c344c19fb5c9ef1dcf0628"
488
            }
489
          },
490
          "metadata": {}
491
        },
492
        {
493
          "output_type": "execute_result",
494
          "data": {
495
            "text/plain": [
496
              "CommitInfo(commit_url='https://huggingface.co/mlabonne/TinyAlpaca/commit/0428c0eb7a6c18dfb7ce7a4cd86ecf4e397048f9', commit_message='Upload folder using huggingface_hub', commit_description='', oid='0428c0eb7a6c18dfb7ce7a4cd86ecf4e397048f9', pr_url=None, pr_revision=None, pr_num=None)"
497
            ],
498
            "application/vnd.google.colaboratory.intrinsic+json": {
499
              "type": "string"
500
            }
501
          },
502
          "metadata": {},
503
          "execution_count": 13
504
        }
505
      ]
506
    }
507
  ],
508
  "metadata": {
509
    "accelerator": "GPU",
510
    "colab": {
511
      "provenance": [],
512
      "gpuType": "T4",
513
      "include_colab_link": true
514
    },
515
    "kernelspec": {
516
      "display_name": "Python 3",
517
      "name": "python3"
518
    },
519
    "language_info": {
520
      "name": "python"
521
    },
522
    "widgets": {
523
      "application/vnd.jupyter.widget-state+json": {
524
        "76eab4e675b34a5486f28d16183658b6": {
525
          "model_module": "@jupyter-widgets/controls",
526
          "model_name": "HBoxModel",
527
          "model_module_version": "1.5.0",
528
          "state": {
529
            "_dom_classes": [],
530
            "_model_module": "@jupyter-widgets/controls",
531
            "_model_module_version": "1.5.0",
532
            "_model_name": "HBoxModel",
533
            "_view_count": null,
534
            "_view_module": "@jupyter-widgets/controls",
535
            "_view_module_version": "1.5.0",
536
            "_view_name": "HBoxView",
537
            "box_style": "",
538
            "children": [
539
              "IPY_MODEL_af7f38e810a64e5f98dd3f33672b8d57",
540
              "IPY_MODEL_0e196628e7e64879bcc7184f67404127",
541
              "IPY_MODEL_2ee0d001c05a4b89ab7a3cbea71c5960"
542
            ],
543
            "layout": "IPY_MODEL_8a59f96ead334ba8b6848321fbde0811"
544
          }
545
        },
546
        "af7f38e810a64e5f98dd3f33672b8d57": {
547
          "model_module": "@jupyter-widgets/controls",
548
          "model_name": "HTMLModel",
549
          "model_module_version": "1.5.0",
550
          "state": {
551
            "_dom_classes": [],
552
            "_model_module": "@jupyter-widgets/controls",
553
            "_model_module_version": "1.5.0",
554
            "_model_name": "HTMLModel",
555
            "_view_count": null,
556
            "_view_module": "@jupyter-widgets/controls",
557
            "_view_module_version": "1.5.0",
558
            "_view_name": "HTMLView",
559
            "description": "",
560
            "description_tooltip": null,
561
            "layout": "IPY_MODEL_3db6a3ae89ec483783f97d77aef5e5a7",
562
            "placeholder": "​",
563
            "style": "IPY_MODEL_f589d24de591428cb878405d1eb6034d",
564
            "value": "pytorch_model.bin: 100%"
565
          }
566
        },
567
        "0e196628e7e64879bcc7184f67404127": {
568
          "model_module": "@jupyter-widgets/controls",
569
          "model_name": "FloatProgressModel",
570
          "model_module_version": "1.5.0",
571
          "state": {
572
            "_dom_classes": [],
573
            "_model_module": "@jupyter-widgets/controls",
574
            "_model_module_version": "1.5.0",
575
            "_model_name": "FloatProgressModel",
576
            "_view_count": null,
577
            "_view_module": "@jupyter-widgets/controls",
578
            "_view_module_version": "1.5.0",
579
            "_view_name": "ProgressView",
580
            "bar_style": "success",
581
            "description": "",
582
            "description_tooltip": null,
583
            "layout": "IPY_MODEL_e4520b22ca934c1fbe399131c021d0e8",
584
            "max": 2200164718,
585
            "min": 0,
586
            "orientation": "horizontal",
587
            "style": "IPY_MODEL_cb663cc62fc94108855d9b5509b573d8",
588
            "value": 2200164718
589
          }
590
        },
591
        "2ee0d001c05a4b89ab7a3cbea71c5960": {
592
          "model_module": "@jupyter-widgets/controls",
593
          "model_name": "HTMLModel",
594
          "model_module_version": "1.5.0",
595
          "state": {
596
            "_dom_classes": [],
597
            "_model_module": "@jupyter-widgets/controls",
598
            "_model_module_version": "1.5.0",
599
            "_model_name": "HTMLModel",
600
            "_view_count": null,
601
            "_view_module": "@jupyter-widgets/controls",
602
            "_view_module_version": "1.5.0",
603
            "_view_name": "HTMLView",
604
            "description": "",
605
            "description_tooltip": null,
606
            "layout": "IPY_MODEL_9926acc37862427d9fdd1ef37474d0c9",
607
            "placeholder": "​",
608
            "style": "IPY_MODEL_215f84842ac5451a8f868e0dc3090a71",
609
            "value": " 2.20G/2.20G [01:14&lt;00:00, 34.5MB/s]"
610
          }
611
        },
612
        "8a59f96ead334ba8b6848321fbde0811": {
613
          "model_module": "@jupyter-widgets/base",
614
          "model_name": "LayoutModel",
615
          "model_module_version": "1.2.0",
616
          "state": {
617
            "_model_module": "@jupyter-widgets/base",
618
            "_model_module_version": "1.2.0",
619
            "_model_name": "LayoutModel",
620
            "_view_count": null,
621
            "_view_module": "@jupyter-widgets/base",
622
            "_view_module_version": "1.2.0",
623
            "_view_name": "LayoutView",
624
            "align_content": null,
625
            "align_items": null,
626
            "align_self": null,
627
            "border": null,
628
            "bottom": null,
629
            "display": null,
630
            "flex": null,
631
            "flex_flow": null,
632
            "grid_area": null,
633
            "grid_auto_columns": null,
634
            "grid_auto_flow": null,
635
            "grid_auto_rows": null,
636
            "grid_column": null,
637
            "grid_gap": null,
638
            "grid_row": null,
639
            "grid_template_areas": null,
640
            "grid_template_columns": null,
641
            "grid_template_rows": null,
642
            "height": null,
643
            "justify_content": null,
644
            "justify_items": null,
645
            "left": null,
646
            "margin": null,
647
            "max_height": null,
648
            "max_width": null,
649
            "min_height": null,
650
            "min_width": null,
651
            "object_fit": null,
652
            "object_position": null,
653
            "order": null,
654
            "overflow": null,
655
            "overflow_x": null,
656
            "overflow_y": null,
657
            "padding": null,
658
            "right": null,
659
            "top": null,
660
            "visibility": null,
661
            "width": null
662
          }
663
        },
664
        "3db6a3ae89ec483783f97d77aef5e5a7": {
665
          "model_module": "@jupyter-widgets/base",
666
          "model_name": "LayoutModel",
667
          "model_module_version": "1.2.0",
668
          "state": {
669
            "_model_module": "@jupyter-widgets/base",
670
            "_model_module_version": "1.2.0",
671
            "_model_name": "LayoutModel",
672
            "_view_count": null,
673
            "_view_module": "@jupyter-widgets/base",
674
            "_view_module_version": "1.2.0",
675
            "_view_name": "LayoutView",
676
            "align_content": null,
677
            "align_items": null,
678
            "align_self": null,
679
            "border": null,
680
            "bottom": null,
681
            "display": null,
682
            "flex": null,
683
            "flex_flow": null,
684
            "grid_area": null,
685
            "grid_auto_columns": null,
686
            "grid_auto_flow": null,
687
            "grid_auto_rows": null,
688
            "grid_column": null,
689
            "grid_gap": null,
690
            "grid_row": null,
691
            "grid_template_areas": null,
692
            "grid_template_columns": null,
693
            "grid_template_rows": null,
694
            "height": null,
695
            "justify_content": null,
696
            "justify_items": null,
697
            "left": null,
698
            "margin": null,
699
            "max_height": null,
700
            "max_width": null,
701
            "min_height": null,
702
            "min_width": null,
703
            "object_fit": null,
704
            "object_position": null,
705
            "order": null,
706
            "overflow": null,
707
            "overflow_x": null,
708
            "overflow_y": null,
709
            "padding": null,
710
            "right": null,
711
            "top": null,
712
            "visibility": null,
713
            "width": null
714
          }
715
        },
716
        "f589d24de591428cb878405d1eb6034d": {
717
          "model_module": "@jupyter-widgets/controls",
718
          "model_name": "DescriptionStyleModel",
719
          "model_module_version": "1.5.0",
720
          "state": {
721
            "_model_module": "@jupyter-widgets/controls",
722
            "_model_module_version": "1.5.0",
723
            "_model_name": "DescriptionStyleModel",
724
            "_view_count": null,
725
            "_view_module": "@jupyter-widgets/base",
726
            "_view_module_version": "1.2.0",
727
            "_view_name": "StyleView",
728
            "description_width": ""
729
          }
730
        },
731
        "e4520b22ca934c1fbe399131c021d0e8": {
732
          "model_module": "@jupyter-widgets/base",
733
          "model_name": "LayoutModel",
734
          "model_module_version": "1.2.0",
735
          "state": {
736
            "_model_module": "@jupyter-widgets/base",
737
            "_model_module_version": "1.2.0",
738
            "_model_name": "LayoutModel",
739
            "_view_count": null,
740
            "_view_module": "@jupyter-widgets/base",
741
            "_view_module_version": "1.2.0",
742
            "_view_name": "LayoutView",
743
            "align_content": null,
744
            "align_items": null,
745
            "align_self": null,
746
            "border": null,
747
            "bottom": null,
748
            "display": null,
749
            "flex": null,
750
            "flex_flow": null,
751
            "grid_area": null,
752
            "grid_auto_columns": null,
753
            "grid_auto_flow": null,
754
            "grid_auto_rows": null,
755
            "grid_column": null,
756
            "grid_gap": null,
757
            "grid_row": null,
758
            "grid_template_areas": null,
759
            "grid_template_columns": null,
760
            "grid_template_rows": null,
761
            "height": null,
762
            "justify_content": null,
763
            "justify_items": null,
764
            "left": null,
765
            "margin": null,
766
            "max_height": null,
767
            "max_width": null,
768
            "min_height": null,
769
            "min_width": null,
770
            "object_fit": null,
771
            "object_position": null,
772
            "order": null,
773
            "overflow": null,
774
            "overflow_x": null,
775
            "overflow_y": null,
776
            "padding": null,
777
            "right": null,
778
            "top": null,
779
            "visibility": null,
780
            "width": null
781
          }
782
        },
783
        "cb663cc62fc94108855d9b5509b573d8": {
784
          "model_module": "@jupyter-widgets/controls",
785
          "model_name": "ProgressStyleModel",
786
          "model_module_version": "1.5.0",
787
          "state": {
788
            "_model_module": "@jupyter-widgets/controls",
789
            "_model_module_version": "1.5.0",
790
            "_model_name": "ProgressStyleModel",
791
            "_view_count": null,
792
            "_view_module": "@jupyter-widgets/base",
793
            "_view_module_version": "1.2.0",
794
            "_view_name": "StyleView",
795
            "bar_color": null,
796
            "description_width": ""
797
          }
798
        },
799
        "9926acc37862427d9fdd1ef37474d0c9": {
800
          "model_module": "@jupyter-widgets/base",
801
          "model_name": "LayoutModel",
802
          "model_module_version": "1.2.0",
803
          "state": {
804
            "_model_module": "@jupyter-widgets/base",
805
            "_model_module_version": "1.2.0",
806
            "_model_name": "LayoutModel",
807
            "_view_count": null,
808
            "_view_module": "@jupyter-widgets/base",
809
            "_view_module_version": "1.2.0",
810
            "_view_name": "LayoutView",
811
            "align_content": null,
812
            "align_items": null,
813
            "align_self": null,
814
            "border": null,
815
            "bottom": null,
816
            "display": null,
817
            "flex": null,
818
            "flex_flow": null,
819
            "grid_area": null,
820
            "grid_auto_columns": null,
821
            "grid_auto_flow": null,
822
            "grid_auto_rows": null,
823
            "grid_column": null,
824
            "grid_gap": null,
825
            "grid_row": null,
826
            "grid_template_areas": null,
827
            "grid_template_columns": null,
828
            "grid_template_rows": null,
829
            "height": null,
830
            "justify_content": null,
831
            "justify_items": null,
832
            "left": null,
833
            "margin": null,
834
            "max_height": null,
835
            "max_width": null,
836
            "min_height": null,
837
            "min_width": null,
838
            "object_fit": null,
839
            "object_position": null,
840
            "order": null,
841
            "overflow": null,
842
            "overflow_x": null,
843
            "overflow_y": null,
844
            "padding": null,
845
            "right": null,
846
            "top": null,
847
            "visibility": null,
848
            "width": null
849
          }
850
        },
851
        "215f84842ac5451a8f868e0dc3090a71": {
852
          "model_module": "@jupyter-widgets/controls",
853
          "model_name": "DescriptionStyleModel",
854
          "model_module_version": "1.5.0",
855
          "state": {
856
            "_model_module": "@jupyter-widgets/controls",
857
            "_model_module_version": "1.5.0",
858
            "_model_name": "DescriptionStyleModel",
859
            "_view_count": null,
860
            "_view_module": "@jupyter-widgets/base",
861
            "_view_module_version": "1.2.0",
862
            "_view_name": "StyleView",
863
            "description_width": ""
864
          }
865
        },
866
        "0878f651a6b64ef495ebd5a2a9084b63": {
867
          "model_module": "@jupyter-widgets/controls",
868
          "model_name": "HBoxModel",
869
          "model_module_version": "1.5.0",
870
          "state": {
871
            "_dom_classes": [],
872
            "_model_module": "@jupyter-widgets/controls",
873
            "_model_module_version": "1.5.0",
874
            "_model_name": "HBoxModel",
875
            "_view_count": null,
876
            "_view_module": "@jupyter-widgets/controls",
877
            "_view_module_version": "1.5.0",
878
            "_view_name": "HBoxView",
879
            "box_style": "",
880
            "children": [
881
              "IPY_MODEL_eec99342bca04c1697ec43bc099c22c9",
882
              "IPY_MODEL_cb7cc89f3edf4c6c8e38043ef0d697ba",
883
              "IPY_MODEL_7a9497ed06004a82b2893e0f5ab03736"
884
            ],
885
            "layout": "IPY_MODEL_f50ea1d3c5874926a6db2af3d9d795ab"
886
          }
887
        },
888
        "eec99342bca04c1697ec43bc099c22c9": {
889
          "model_module": "@jupyter-widgets/controls",
890
          "model_name": "HTMLModel",
891
          "model_module_version": "1.5.0",
892
          "state": {
893
            "_dom_classes": [],
894
            "_model_module": "@jupyter-widgets/controls",
895
            "_model_module_version": "1.5.0",
896
            "_model_name": "HTMLModel",
897
            "_view_count": null,
898
            "_view_module": "@jupyter-widgets/controls",
899
            "_view_module_version": "1.5.0",
900
            "_view_name": "HTMLView",
901
            "description": "",
902
            "description_tooltip": null,
903
            "layout": "IPY_MODEL_7851c929f5f64e31856d3793b963130e",
904
            "placeholder": "​",
905
            "style": "IPY_MODEL_49f8123c3c5e43ababed7e09e97cb7d2",
906
            "value": "Upload 2 LFS files: 100%"
907
          }
908
        },
909
        "cb7cc89f3edf4c6c8e38043ef0d697ba": {
910
          "model_module": "@jupyter-widgets/controls",
911
          "model_name": "FloatProgressModel",
912
          "model_module_version": "1.5.0",
913
          "state": {
914
            "_dom_classes": [],
915
            "_model_module": "@jupyter-widgets/controls",
916
            "_model_module_version": "1.5.0",
917
            "_model_name": "FloatProgressModel",
918
            "_view_count": null,
919
            "_view_module": "@jupyter-widgets/controls",
920
            "_view_module_version": "1.5.0",
921
            "_view_name": "ProgressView",
922
            "bar_style": "success",
923
            "description": "",
924
            "description_tooltip": null,
925
            "layout": "IPY_MODEL_c4dfec3f687241d6a69cf30252ada077",
926
            "max": 2,
927
            "min": 0,
928
            "orientation": "horizontal",
929
            "style": "IPY_MODEL_a1c9ecf2bc124658a08f585b1bf309e6",
930
            "value": 2
931
          }
932
        },
933
        "7a9497ed06004a82b2893e0f5ab03736": {
934
          "model_module": "@jupyter-widgets/controls",
935
          "model_name": "HTMLModel",
936
          "model_module_version": "1.5.0",
937
          "state": {
938
            "_dom_classes": [],
939
            "_model_module": "@jupyter-widgets/controls",
940
            "_model_module_version": "1.5.0",
941
            "_model_name": "HTMLModel",
942
            "_view_count": null,
943
            "_view_module": "@jupyter-widgets/controls",
944
            "_view_module_version": "1.5.0",
945
            "_view_name": "HTMLView",
946
            "description": "",
947
            "description_tooltip": null,
948
            "layout": "IPY_MODEL_101d0b228b1943ff9c9d69fd6f4847ee",
949
            "placeholder": "​",
950
            "style": "IPY_MODEL_b03c7fa74c774a39946636803e086218",
951
            "value": " 2/2 [01:14&lt;00:00, 74.58s/it]"
952
          }
953
        },
954
        "f50ea1d3c5874926a6db2af3d9d795ab": {
955
          "model_module": "@jupyter-widgets/base",
956
          "model_name": "LayoutModel",
957
          "model_module_version": "1.2.0",
958
          "state": {
959
            "_model_module": "@jupyter-widgets/base",
960
            "_model_module_version": "1.2.0",
961
            "_model_name": "LayoutModel",
962
            "_view_count": null,
963
            "_view_module": "@jupyter-widgets/base",
964
            "_view_module_version": "1.2.0",
965
            "_view_name": "LayoutView",
966
            "align_content": null,
967
            "align_items": null,
968
            "align_self": null,
969
            "border": null,
970
            "bottom": null,
971
            "display": null,
972
            "flex": null,
973
            "flex_flow": null,
974
            "grid_area": null,
975
            "grid_auto_columns": null,
976
            "grid_auto_flow": null,
977
            "grid_auto_rows": null,
978
            "grid_column": null,
979
            "grid_gap": null,
980
            "grid_row": null,
981
            "grid_template_areas": null,
982
            "grid_template_columns": null,
983
            "grid_template_rows": null,
984
            "height": null,
985
            "justify_content": null,
986
            "justify_items": null,
987
            "left": null,
988
            "margin": null,
989
            "max_height": null,
990
            "max_width": null,
991
            "min_height": null,
992
            "min_width": null,
993
            "object_fit": null,
994
            "object_position": null,
995
            "order": null,
996
            "overflow": null,
997
            "overflow_x": null,
998
            "overflow_y": null,
999
            "padding": null,
1000
            "right": null,
1001
            "top": null,
1002
            "visibility": null,
1003
            "width": null
1004
          }
1005
        },
1006
        "7851c929f5f64e31856d3793b963130e": {
1007
          "model_module": "@jupyter-widgets/base",
1008
          "model_name": "LayoutModel",
1009
          "model_module_version": "1.2.0",
1010
          "state": {
1011
            "_model_module": "@jupyter-widgets/base",
1012
            "_model_module_version": "1.2.0",
1013
            "_model_name": "LayoutModel",
1014
            "_view_count": null,
1015
            "_view_module": "@jupyter-widgets/base",
1016
            "_view_module_version": "1.2.0",
1017
            "_view_name": "LayoutView",
1018
            "align_content": null,
1019
            "align_items": null,
1020
            "align_self": null,
1021
            "border": null,
1022
            "bottom": null,
1023
            "display": null,
1024
            "flex": null,
1025
            "flex_flow": null,
1026
            "grid_area": null,
1027
            "grid_auto_columns": null,
1028
            "grid_auto_flow": null,
1029
            "grid_auto_rows": null,
1030
            "grid_column": null,
1031
            "grid_gap": null,
1032
            "grid_row": null,
1033
            "grid_template_areas": null,
1034
            "grid_template_columns": null,
1035
            "grid_template_rows": null,
1036
            "height": null,
1037
            "justify_content": null,
1038
            "justify_items": null,
1039
            "left": null,
1040
            "margin": null,
1041
            "max_height": null,
1042
            "max_width": null,
1043
            "min_height": null,
1044
            "min_width": null,
1045
            "object_fit": null,
1046
            "object_position": null,
1047
            "order": null,
1048
            "overflow": null,
1049
            "overflow_x": null,
1050
            "overflow_y": null,
1051
            "padding": null,
1052
            "right": null,
1053
            "top": null,
1054
            "visibility": null,
1055
            "width": null
1056
          }
1057
        },
1058
        "49f8123c3c5e43ababed7e09e97cb7d2": {
1059
          "model_module": "@jupyter-widgets/controls",
1060
          "model_name": "DescriptionStyleModel",
1061
          "model_module_version": "1.5.0",
1062
          "state": {
1063
            "_model_module": "@jupyter-widgets/controls",
1064
            "_model_module_version": "1.5.0",
1065
            "_model_name": "DescriptionStyleModel",
1066
            "_view_count": null,
1067
            "_view_module": "@jupyter-widgets/base",
1068
            "_view_module_version": "1.2.0",
1069
            "_view_name": "StyleView",
1070
            "description_width": ""
1071
          }
1072
        },
1073
        "c4dfec3f687241d6a69cf30252ada077": {
1074
          "model_module": "@jupyter-widgets/base",
1075
          "model_name": "LayoutModel",
1076
          "model_module_version": "1.2.0",
1077
          "state": {
1078
            "_model_module": "@jupyter-widgets/base",
1079
            "_model_module_version": "1.2.0",
1080
            "_model_name": "LayoutModel",
1081
            "_view_count": null,
1082
            "_view_module": "@jupyter-widgets/base",
1083
            "_view_module_version": "1.2.0",
1084
            "_view_name": "LayoutView",
1085
            "align_content": null,
1086
            "align_items": null,
1087
            "align_self": null,
1088
            "border": null,
1089
            "bottom": null,
1090
            "display": null,
1091
            "flex": null,
1092
            "flex_flow": null,
1093
            "grid_area": null,
1094
            "grid_auto_columns": null,
1095
            "grid_auto_flow": null,
1096
            "grid_auto_rows": null,
1097
            "grid_column": null,
1098
            "grid_gap": null,
1099
            "grid_row": null,
1100
            "grid_template_areas": null,
1101
            "grid_template_columns": null,
1102
            "grid_template_rows": null,
1103
            "height": null,
1104
            "justify_content": null,
1105
            "justify_items": null,
1106
            "left": null,
1107
            "margin": null,
1108
            "max_height": null,
1109
            "max_width": null,
1110
            "min_height": null,
1111
            "min_width": null,
1112
            "object_fit": null,
1113
            "object_position": null,
1114
            "order": null,
1115
            "overflow": null,
1116
            "overflow_x": null,
1117
            "overflow_y": null,
1118
            "padding": null,
1119
            "right": null,
1120
            "top": null,
1121
            "visibility": null,
1122
            "width": null
1123
          }
1124
        },
1125
        "a1c9ecf2bc124658a08f585b1bf309e6": {
1126
          "model_module": "@jupyter-widgets/controls",
1127
          "model_name": "ProgressStyleModel",
1128
          "model_module_version": "1.5.0",
1129
          "state": {
1130
            "_model_module": "@jupyter-widgets/controls",
1131
            "_model_module_version": "1.5.0",
1132
            "_model_name": "ProgressStyleModel",
1133
            "_view_count": null,
1134
            "_view_module": "@jupyter-widgets/base",
1135
            "_view_module_version": "1.2.0",
1136
            "_view_name": "StyleView",
1137
            "bar_color": null,
1138
            "description_width": ""
1139
          }
1140
        },
1141
        "101d0b228b1943ff9c9d69fd6f4847ee": {
1142
          "model_module": "@jupyter-widgets/base",
1143
          "model_name": "LayoutModel",
1144
          "model_module_version": "1.2.0",
1145
          "state": {
1146
            "_model_module": "@jupyter-widgets/base",
1147
            "_model_module_version": "1.2.0",
1148
            "_model_name": "LayoutModel",
1149
            "_view_count": null,
1150
            "_view_module": "@jupyter-widgets/base",
1151
            "_view_module_version": "1.2.0",
1152
            "_view_name": "LayoutView",
1153
            "align_content": null,
1154
            "align_items": null,
1155
            "align_self": null,
1156
            "border": null,
1157
            "bottom": null,
1158
            "display": null,
1159
            "flex": null,
1160
            "flex_flow": null,
1161
            "grid_area": null,
1162
            "grid_auto_columns": null,
1163
            "grid_auto_flow": null,
1164
            "grid_auto_rows": null,
1165
            "grid_column": null,
1166
            "grid_gap": null,
1167
            "grid_row": null,
1168
            "grid_template_areas": null,
1169
            "grid_template_columns": null,
1170
            "grid_template_rows": null,
1171
            "height": null,
1172
            "justify_content": null,
1173
            "justify_items": null,
1174
            "left": null,
1175
            "margin": null,
1176
            "max_height": null,
1177
            "max_width": null,
1178
            "min_height": null,
1179
            "min_width": null,
1180
            "object_fit": null,
1181
            "object_position": null,
1182
            "order": null,
1183
            "overflow": null,
1184
            "overflow_x": null,
1185
            "overflow_y": null,
1186
            "padding": null,
1187
            "right": null,
1188
            "top": null,
1189
            "visibility": null,
1190
            "width": null
1191
          }
1192
        },
1193
        "b03c7fa74c774a39946636803e086218": {
1194
          "model_module": "@jupyter-widgets/controls",
1195
          "model_name": "DescriptionStyleModel",
1196
          "model_module_version": "1.5.0",
1197
          "state": {
1198
            "_model_module": "@jupyter-widgets/controls",
1199
            "_model_module_version": "1.5.0",
1200
            "_model_name": "DescriptionStyleModel",
1201
            "_view_count": null,
1202
            "_view_module": "@jupyter-widgets/base",
1203
            "_view_module_version": "1.2.0",
1204
            "_view_name": "StyleView",
1205
            "description_width": ""
1206
          }
1207
        },
1208
        "90726b0049c344c19fb5c9ef1dcf0628": {
1209
          "model_module": "@jupyter-widgets/controls",
1210
          "model_name": "HBoxModel",
1211
          "model_module_version": "1.5.0",
1212
          "state": {
1213
            "_dom_classes": [],
1214
            "_model_module": "@jupyter-widgets/controls",
1215
            "_model_module_version": "1.5.0",
1216
            "_model_name": "HBoxModel",
1217
            "_view_count": null,
1218
            "_view_module": "@jupyter-widgets/controls",
1219
            "_view_module_version": "1.5.0",
1220
            "_view_name": "HBoxView",
1221
            "box_style": "",
1222
            "children": [
1223
              "IPY_MODEL_bebef38ee9e24586a7f894ffaf1ae43e",
1224
              "IPY_MODEL_9d90c5ea423e4586bc3bb303bcaa4b03",
1225
              "IPY_MODEL_d417d5bf00294e3f9cb7e78b5d83ee03"
1226
            ],
1227
            "layout": "IPY_MODEL_68eac408135d4d048de89fa674b701da"
1228
          }
1229
        },
1230
        "bebef38ee9e24586a7f894ffaf1ae43e": {
1231
          "model_module": "@jupyter-widgets/controls",
1232
          "model_name": "HTMLModel",
1233
          "model_module_version": "1.5.0",
1234
          "state": {
1235
            "_dom_classes": [],
1236
            "_model_module": "@jupyter-widgets/controls",
1237
            "_model_module_version": "1.5.0",
1238
            "_model_name": "HTMLModel",
1239
            "_view_count": null,
1240
            "_view_module": "@jupyter-widgets/controls",
1241
            "_view_module_version": "1.5.0",
1242
            "_view_name": "HTMLView",
1243
            "description": "",
1244
            "description_tooltip": null,
1245
            "layout": "IPY_MODEL_34c11d3b1a3441bf8533f7a6f057d0cf",
1246
            "placeholder": "​",
1247
            "style": "IPY_MODEL_6270a81b7f5749d7a656c1edaaa63118",
1248
            "value": "tokenizer.model: 100%"
1249
          }
1250
        },
1251
        "9d90c5ea423e4586bc3bb303bcaa4b03": {
1252
          "model_module": "@jupyter-widgets/controls",
1253
          "model_name": "FloatProgressModel",
1254
          "model_module_version": "1.5.0",
1255
          "state": {
1256
            "_dom_classes": [],
1257
            "_model_module": "@jupyter-widgets/controls",
1258
            "_model_module_version": "1.5.0",
1259
            "_model_name": "FloatProgressModel",
1260
            "_view_count": null,
1261
            "_view_module": "@jupyter-widgets/controls",
1262
            "_view_module_version": "1.5.0",
1263
            "_view_name": "ProgressView",
1264
            "bar_style": "success",
1265
            "description": "",
1266
            "description_tooltip": null,
1267
            "layout": "IPY_MODEL_e1de7a96086c4351bba99bbd4d905802",
1268
            "max": 499723,
1269
            "min": 0,
1270
            "orientation": "horizontal",
1271
            "style": "IPY_MODEL_6d28893d0a054d6aa2726ee94ee0d880",
1272
            "value": 499723
1273
          }
1274
        },
1275
        "d417d5bf00294e3f9cb7e78b5d83ee03": {
1276
          "model_module": "@jupyter-widgets/controls",
1277
          "model_name": "HTMLModel",
1278
          "model_module_version": "1.5.0",
1279
          "state": {
1280
            "_dom_classes": [],
1281
            "_model_module": "@jupyter-widgets/controls",
1282
            "_model_module_version": "1.5.0",
1283
            "_model_name": "HTMLModel",
1284
            "_view_count": null,
1285
            "_view_module": "@jupyter-widgets/controls",
1286
            "_view_module_version": "1.5.0",
1287
            "_view_name": "HTMLView",
1288
            "description": "",
1289
            "description_tooltip": null,
1290
            "layout": "IPY_MODEL_91c9822b8ef14074bd360bb5d2998267",
1291
            "placeholder": "​",
1292
            "style": "IPY_MODEL_8673edee8ab5430588daba8d422999eb",
1293
            "value": " 500k/500k [00:00&lt;00:00, 48.1kB/s]"
1294
          }
1295
        },
1296
        "68eac408135d4d048de89fa674b701da": {
1297
          "model_module": "@jupyter-widgets/base",
1298
          "model_name": "LayoutModel",
1299
          "model_module_version": "1.2.0",
1300
          "state": {
1301
            "_model_module": "@jupyter-widgets/base",
1302
            "_model_module_version": "1.2.0",
1303
            "_model_name": "LayoutModel",
1304
            "_view_count": null,
1305
            "_view_module": "@jupyter-widgets/base",
1306
            "_view_module_version": "1.2.0",
1307
            "_view_name": "LayoutView",
1308
            "align_content": null,
1309
            "align_items": null,
1310
            "align_self": null,
1311
            "border": null,
1312
            "bottom": null,
1313
            "display": null,
1314
            "flex": null,
1315
            "flex_flow": null,
1316
            "grid_area": null,
1317
            "grid_auto_columns": null,
1318
            "grid_auto_flow": null,
1319
            "grid_auto_rows": null,
1320
            "grid_column": null,
1321
            "grid_gap": null,
1322
            "grid_row": null,
1323
            "grid_template_areas": null,
1324
            "grid_template_columns": null,
1325
            "grid_template_rows": null,
1326
            "height": null,
1327
            "justify_content": null,
1328
            "justify_items": null,
1329
            "left": null,
1330
            "margin": null,
1331
            "max_height": null,
1332
            "max_width": null,
1333
            "min_height": null,
1334
            "min_width": null,
1335
            "object_fit": null,
1336
            "object_position": null,
1337
            "order": null,
1338
            "overflow": null,
1339
            "overflow_x": null,
1340
            "overflow_y": null,
1341
            "padding": null,
1342
            "right": null,
1343
            "top": null,
1344
            "visibility": null,
1345
            "width": null
1346
          }
1347
        },
1348
        "34c11d3b1a3441bf8533f7a6f057d0cf": {
1349
          "model_module": "@jupyter-widgets/base",
1350
          "model_name": "LayoutModel",
1351
          "model_module_version": "1.2.0",
1352
          "state": {
1353
            "_model_module": "@jupyter-widgets/base",
1354
            "_model_module_version": "1.2.0",
1355
            "_model_name": "LayoutModel",
1356
            "_view_count": null,
1357
            "_view_module": "@jupyter-widgets/base",
1358
            "_view_module_version": "1.2.0",
1359
            "_view_name": "LayoutView",
1360
            "align_content": null,
1361
            "align_items": null,
1362
            "align_self": null,
1363
            "border": null,
1364
            "bottom": null,
1365
            "display": null,
1366
            "flex": null,
1367
            "flex_flow": null,
1368
            "grid_area": null,
1369
            "grid_auto_columns": null,
1370
            "grid_auto_flow": null,
1371
            "grid_auto_rows": null,
1372
            "grid_column": null,
1373
            "grid_gap": null,
1374
            "grid_row": null,
1375
            "grid_template_areas": null,
1376
            "grid_template_columns": null,
1377
            "grid_template_rows": null,
1378
            "height": null,
1379
            "justify_content": null,
1380
            "justify_items": null,
1381
            "left": null,
1382
            "margin": null,
1383
            "max_height": null,
1384
            "max_width": null,
1385
            "min_height": null,
1386
            "min_width": null,
1387
            "object_fit": null,
1388
            "object_position": null,
1389
            "order": null,
1390
            "overflow": null,
1391
            "overflow_x": null,
1392
            "overflow_y": null,
1393
            "padding": null,
1394
            "right": null,
1395
            "top": null,
1396
            "visibility": null,
1397
            "width": null
1398
          }
1399
        },
1400
        "6270a81b7f5749d7a656c1edaaa63118": {
1401
          "model_module": "@jupyter-widgets/controls",
1402
          "model_name": "DescriptionStyleModel",
1403
          "model_module_version": "1.5.0",
1404
          "state": {
1405
            "_model_module": "@jupyter-widgets/controls",
1406
            "_model_module_version": "1.5.0",
1407
            "_model_name": "DescriptionStyleModel",
1408
            "_view_count": null,
1409
            "_view_module": "@jupyter-widgets/base",
1410
            "_view_module_version": "1.2.0",
1411
            "_view_name": "StyleView",
1412
            "description_width": ""
1413
          }
1414
        },
1415
        "e1de7a96086c4351bba99bbd4d905802": {
1416
          "model_module": "@jupyter-widgets/base",
1417
          "model_name": "LayoutModel",
1418
          "model_module_version": "1.2.0",
1419
          "state": {
1420
            "_model_module": "@jupyter-widgets/base",
1421
            "_model_module_version": "1.2.0",
1422
            "_model_name": "LayoutModel",
1423
            "_view_count": null,
1424
            "_view_module": "@jupyter-widgets/base",
1425
            "_view_module_version": "1.2.0",
1426
            "_view_name": "LayoutView",
1427
            "align_content": null,
1428
            "align_items": null,
1429
            "align_self": null,
1430
            "border": null,
1431
            "bottom": null,
1432
            "display": null,
1433
            "flex": null,
1434
            "flex_flow": null,
1435
            "grid_area": null,
1436
            "grid_auto_columns": null,
1437
            "grid_auto_flow": null,
1438
            "grid_auto_rows": null,
1439
            "grid_column": null,
1440
            "grid_gap": null,
1441
            "grid_row": null,
1442
            "grid_template_areas": null,
1443
            "grid_template_columns": null,
1444
            "grid_template_rows": null,
1445
            "height": null,
1446
            "justify_content": null,
1447
            "justify_items": null,
1448
            "left": null,
1449
            "margin": null,
1450
            "max_height": null,
1451
            "max_width": null,
1452
            "min_height": null,
1453
            "min_width": null,
1454
            "object_fit": null,
1455
            "object_position": null,
1456
            "order": null,
1457
            "overflow": null,
1458
            "overflow_x": null,
1459
            "overflow_y": null,
1460
            "padding": null,
1461
            "right": null,
1462
            "top": null,
1463
            "visibility": null,
1464
            "width": null
1465
          }
1466
        },
1467
        "6d28893d0a054d6aa2726ee94ee0d880": {
1468
          "model_module": "@jupyter-widgets/controls",
1469
          "model_name": "ProgressStyleModel",
1470
          "model_module_version": "1.5.0",
1471
          "state": {
1472
            "_model_module": "@jupyter-widgets/controls",
1473
            "_model_module_version": "1.5.0",
1474
            "_model_name": "ProgressStyleModel",
1475
            "_view_count": null,
1476
            "_view_module": "@jupyter-widgets/base",
1477
            "_view_module_version": "1.2.0",
1478
            "_view_name": "StyleView",
1479
            "bar_color": null,
1480
            "description_width": ""
1481
          }
1482
        },
1483
        "91c9822b8ef14074bd360bb5d2998267": {
1484
          "model_module": "@jupyter-widgets/base",
1485
          "model_name": "LayoutModel",
1486
          "model_module_version": "1.2.0",
1487
          "state": {
1488
            "_model_module": "@jupyter-widgets/base",
1489
            "_model_module_version": "1.2.0",
1490
            "_model_name": "LayoutModel",
1491
            "_view_count": null,
1492
            "_view_module": "@jupyter-widgets/base",
1493
            "_view_module_version": "1.2.0",
1494
            "_view_name": "LayoutView",
1495
            "align_content": null,
1496
            "align_items": null,
1497
            "align_self": null,
1498
            "border": null,
1499
            "bottom": null,
1500
            "display": null,
1501
            "flex": null,
1502
            "flex_flow": null,
1503
            "grid_area": null,
1504
            "grid_auto_columns": null,
1505
            "grid_auto_flow": null,
1506
            "grid_auto_rows": null,
1507
            "grid_column": null,
1508
            "grid_gap": null,
1509
            "grid_row": null,
1510
            "grid_template_areas": null,
1511
            "grid_template_columns": null,
1512
            "grid_template_rows": null,
1513
            "height": null,
1514
            "justify_content": null,
1515
            "justify_items": null,
1516
            "left": null,
1517
            "margin": null,
1518
            "max_height": null,
1519
            "max_width": null,
1520
            "min_height": null,
1521
            "min_width": null,
1522
            "object_fit": null,
1523
            "object_position": null,
1524
            "order": null,
1525
            "overflow": null,
1526
            "overflow_x": null,
1527
            "overflow_y": null,
1528
            "padding": null,
1529
            "right": null,
1530
            "top": null,
1531
            "visibility": null,
1532
            "width": null
1533
          }
1534
        },
1535
        "8673edee8ab5430588daba8d422999eb": {
1536
          "model_module": "@jupyter-widgets/controls",
1537
          "model_name": "DescriptionStyleModel",
1538
          "model_module_version": "1.5.0",
1539
          "state": {
1540
            "_model_module": "@jupyter-widgets/controls",
1541
            "_model_module_version": "1.5.0",
1542
            "_model_name": "DescriptionStyleModel",
1543
            "_view_count": null,
1544
            "_view_module": "@jupyter-widgets/base",
1545
            "_view_module_version": "1.2.0",
1546
            "_view_name": "StyleView",
1547
            "description_width": ""
1548
          }
1549
        }
1550
      }
1551
    }
1552
  },
1553
  "nbformat": 4,
1554
  "nbformat_minor": 0
1555
}

Использование cookies

Мы используем файлы cookie в соответствии с Политикой конфиденциальности и Политикой использования cookies.

Нажимая кнопку «Принимаю», Вы даете АО «СберТех» согласие на обработку Ваших персональных данных в целях совершенствования нашего веб-сайта и Сервиса GitVerse, а также повышения удобства их использования.

Запретить использование cookies Вы можете самостоятельно в настройках Вашего браузера.