# A10G not using VRAM after generating training split in AutoTrain

**URL:** <https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457>\
**Category:** 🤗AutoTrain\
**Created:** [August 12, 2025, 12:18am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457 "2025-08-12T00:18:06Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![SlayerL99](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/slayerl99/32/52448_2.png) [@SlayerL99](https://discuss.huggingface.co/u/SlayerL99)\
**Post date:** [August 12, 2025, 12:18am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457/1 "2025-08-12T00:18:06Z")

</div>

Hi everyone,

I recently purchased a Hugging Face AutoTrain space with an **NVIDIA A10G (24GB, reports ~22.49 GiB usable)** for fine-tuning `nothingiisreal/MN-12B-Celeste-V1.9` on `josecannete/large_spanish_corpus`.

When I start training, AutoTrain first generates the training split. During that step, VRAM usage is basically zero (around **2.88 MiB/22.49 GiB** ). After the split finishes, the process just stops — no training actually begins, and GPU usage never increases.

I expected VRAM usage to spike when training started, but it seems the job never reaches that stage.

Has anyone else experienced this with AutoTrain + A10G?  
Could this be an issue with:

- Dataset size or format?

- The LoRA/PEFT + quantization setup I’m using?

- Some AutoTrain pipeline bug for large models?

Any help would be appreciated — I just want to confirm if this is normal behavior for the split step, and why the actual training might not be starting.

Thanks in advance!

I want to train Celeste V1.9 to learn Spanish, then Spanish books with PleIAs/Spanish-PD-Books and then Argentine Spanish with ylacombe/google-argentinian-spanish. But I’m not that sure if my current JSON for Spanish Corpus is OK, or how to config the other JSON for next steps.

This is my JSON:

```json
{
  "model": "nothingiisreal/MN-12B-Celeste-V1.9",
  "data": "josecannete/large_spanish_corpus",
  "task": "text-generation",

  "hub_model_id": "SlayerL99/mn12b-celeste-espanol-stage1",

  "training_parameters": {
    "learning_rate": 0.00005,
    "num_train_epochs": 1,
    "per_device_train_batch_size": 1,
    "gradient_accumulation_steps": 16,
    "warmup_steps": 100,
    "max_seq_length": 2048,
    "weight_decay": 0.01,
    "lr_scheduler_type": "cosine",
    "seed": 42,
    "fp16": true, // switch bf16 to fp16 for better compatibility with 3060 downstream
    "bf16": false,
    "gradient_checkpointing": true,
    "dataloader_num_workers": 4, // boost data loading speed (assuming Linux/WSL)
    "push_to_hub": true,
    "save_total_limit": 2,
    "logging_steps": 25,
    "save_steps": 200,
    "eval_steps": 200,
    "evaluation_strategy": "steps",
    "load_best_model_at_end": true,
    "metric_for_best_model": "eval_loss",
    "greater_is_better": false,
    "report_to": "tensorboard"
  },

  "peft_parameters": {
    "use_peft": true,
    "lora_r": 16,
    "lora_alpha": 32,
    "lora_dropout": 0.1,
    "bias": "none",
    "task_type": "CAUSAL_LM",
    "target_modules": "all-linear"
  },

  "quantization_parameters": {
    "use_int4": true,
    "use_int8": false,
    "use_fp4": false, // disable fp4 for compatibility/stability on RTX 3060 GGUF
    "use_double_quant": true,
    "bnb_4bit_quant_type": "nf4"
  }
}

```

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\
**Post date:** [August 12, 2025, 4:19am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457/2 "2025-08-12T04:19:34Z")

</div>

> I just want to confirm if this is normal behavior for the split step, and why the actual training might not be starting.

Yeah. Maybe. I think [AutoTrain is designed to _stop as quickly as possible when any error occurs_](https://github.com/huggingface/autotrain-advanced/issues/670).

BTW, that JSON settings may be for the parameters of the old version of Trainer. How about like this (in YAML)?

```yaml
task: llm
base_model: nothingiisreal/MN-12B-Celeste-V1.9
project_name: mn12b-celeste-espanol-stage1
log: tensorboard

data:
  path: josecannete/large_spanish_corpus
  train_split: train
  valid_split: null
  chat_template: null
  column_mapping:
    text_column: text

params:
  trainer: sft
  block_size: -1
  model_max_length: 4096
  epochs: 1
  batch_size: 1
  gradient_accumulation: 16
  lr: 5e-5
  warmup_ratio: 0.1
  optimizer: adamw_torch
  scheduler: linear
  weight_decay: 0.01
  logging_steps: 25
  eval_strategy: epoch
  save_total_limit: 2
  mixed_precision: fp16

  # QLoRA
  peft: true
  quantization: int4
  target_modules: all-linear
  lora_r: 16
  lora_alpha: 32
  lora_dropout: 0.10

  padding: right
  seed: 42

hub:
  username: SlayerL99
  push_to_hub: true

```

- [LLM Finetuning with AutoTrain Advanced](https://huggingface.co/docs/autotrain/en/tasks/llm_finetuning)
- [[BUG] Using `bitsandbytes` 8-bit quantization requires Accelerate: `pip install accelerate` and the latest version of bitsandbytes](https://github.com/huggingface/autotrain-advanced/issues/528) (An example of errors)

---

<div class="post-metadata">

**Author:** ![SlayerL99](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/slayerl99/32/52448_2.png) [@SlayerL99](https://discuss.huggingface.co/u/SlayerL99)\
**Post date:** [August 12, 2025, 5:27am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457/3 "2025-08-12T05:27:59Z")

</div>

Ah, that may be it:

> [@John6666](#):
>
> BTW, that JSON settings may be for the parameters of the old version of Trainer. How about like this (in YAML)?

I just transform that into JSON and paste it in the parameters and try?

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\
**Post date:** [August 12, 2025, 5:29am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457/4 "2025-08-12T05:29:38Z")

</div>

> I just transform that into JSON and paste it in the parameters and try?

Maybe okay. I think they were treated the same internally anyway…

---

<div class="post-metadata">

**Author:** ![SlayerL99](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/slayerl99/32/52448_2.png) [@SlayerL99](https://discuss.huggingface.co/u/SlayerL99)\
**Post date:** [August 12, 2025, 6:25am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457/5 "2025-08-12T06:25:17Z")

</div>

Heyo! I tried using that YAML into JSON, but the same happens. AutoTrainer gives “Error fertching trainer status” and just stops 😕

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\
**Post date:** [August 12, 2025, 6:41am UTC](https://discuss.huggingface.co/t/a10g-not-using-vram-after-generating-training-split-in-autotrain/166457/6 "2025-08-12T06:41:48Z")

</div>

> “Error fertching trainer status”

I think this error is caused by something wrong with the Accelerate library, but could it be that the version of the Accelerate library on the newly created space is outdated?🤔

> <https://github.com/huggingface/autotrain-advanced/issues/786>
>
> \### Prerequisites
> 
> \- \[X\] I have read the \[documentation\](https://hf.co/docs/auto…train).
> \- \[X\] I have checked other issues for similar problems.
> 
> \### Backend
> 
> Hugging Face Space/Endpoints
> 
> \### Interface Used
> 
> UI
> 
> \### CLI Command
> 
> \_No response\_
> 
> \### UI Screenshots & Parameters
> 
> Issue with AutoTrain Advanced:
> 
> Project: Fine-tuning LLM (Legacy-Ledger-Fine-Tune-v1)
> Base Model: meta-llama/Llama-3-1-8B-Instruct
> Dataset: mb7419/legal-advice-reddit (Hugging Face Hub)
> 
> Error Description:
> 
> Initial error: ValueError regarding unknown split "0.8"
> Subsequent error: subprocess.CalledProcessError when launching training
> 
> Key Error Details:
> 
> Command execution failed with non-zero exit status 2
> Unrecognized arguments: repeated "-m autotrain.trainers.clm"
> Training process (PID 354) terminated unexpectedly
> 
> Additional Notes:
> 
> Accelerate configuration warnings present
> Issue persists after addressing initial dataset split problem
> 
> The errors suggest a potential bug in the AutoTrain backend, possibly related to command argument handling or training process initialization. This prevents the training from starting successfully.
> \<img width="1196" alt="Screenshot 2024-10-04 at 1 14 44 PM" src="https://github.com/user-attachments/assets/922597d6-2eb9-40f1-a528-1caf669d51cb"\>
> 
> 
> \### Error Logs
> 
> INFO | 2024-10-04 20:09:15 | autotrain.app.utils:kill\_process\_by\_pid:52 - Sent SIGTERM to process with PID 354
> 
> INFO | 2024-10-04 20:09:15 | autotrain.app.utils:get\_running\_jobs:26 - Killing PID: 354
> 
> subprocess.CalledProcessError: Command '\['/app/env/bin/python', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json'\]' returned non-zero exit status 2.
> 
> raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
> 
> File "/app/env/lib/python3.10/site-packages/accelerate/commands/launch.py", line 769, in simple\_launcher
> 
> simple\_launcher(args)
> 
> File "/app/env/lib/python3.10/site-packages/accelerate/commands/launch.py", line 1174, in launch\_command
> 
> args.func(args)
> 
> File "/app/env/lib/python3.10/site-packages/accelerate/commands/accelerate\_cli.py", line 48, in main
> 
> sys.exit(main())
> 
> File "/app/env/bin/accelerate", line 8, in \<module\>
> 
> Traceback (most recent call last):
> 
> \_\_main\_\_.py: error: unrecognized arguments: -m autotrain.trainers.clm -m autotrain.trainers.clm
> 
> usage: \_\_main\_\_.py \[-h\] --training\_config TRAINING\_CONFIG
> 
> To avoid this warning pass in values for each of the problematic parameters or run \`accelerate config\`.
> 
> \`--dynamo\_backend\` was set to a value of \`'no'\`
> 
> \`--mixed\_precision\` was set to a value of \`'no'\`
> 
> \`--num\_machines\` was set to a value of \`1\`
> 
> \`--num\_processes\` was set to a value of \`0\`
> 
> The following values were not passed to \`accelerate launch\` and had defaults used instead:
> 
> INFO | 2024-10-04 20:09:10 | autotrain.backends.local:create:13 - Training PID: 354
> 
> INFO | 2024-10-04 20:09:10 | autotrain.commands:launch\_command:502 - {'model': 'meta-llama/Llama-3.2-3B-Instruct', 'project\_name': 'legacy-ledger-at-v1', 'data\_path': 'bunny0702/Legal\_Research', 'train\_split': '0.8', 'valid\_split': None, 'add\_eos\_token': True, 'block\_size': 1024, 'model\_max\_length': 2048, 'padding': 'right', 'trainer': 'sft', 'use\_flash\_attention\_2': False, 'log': 'tensorboard', 'disable\_gradient\_checkpointing': False, 'logging\_steps': -1, 'eval\_strategy': 'epoch', 'save\_total\_limit': 1, 'auto\_find\_batch\_size': False, 'mixed\_precision': 'bf16', 'lr': 3e-05, 'epochs': 3, 'batch\_size': 8, 'warmup\_ratio': 0.1, 'gradient\_accumulation': 4, 'optimizer': 'adamw\_torch', 'scheduler': 'linear', 'weight\_decay': 0.0, 'max\_grad\_norm': 1.0, 'seed': 42, 'chat\_template': 'none', 'quantization': 'int4', 'target\_modules': 'all-linear', 'merge\_adapter': False, 'peft': True, 'lora\_r': 16, 'lora\_alpha': 32, 'lora\_dropout': 0.05, 'model\_ref': None, 'dpo\_beta': 0.1, 'max\_prompt\_length': 128, 'max\_completion\_length': None, 'prompt\_text\_column': 'prompt', 'text\_column': 'text', 'rejected\_text\_column': 'rejected\_text', 'push\_to\_hub': True, 'username': 'Gladystoper', 'token': '\*\*\*\*\*', 'unsloth': False, 'distributed\_backend': None}
> 
> INFO | 2024-10-04 20:09:10 | autotrain.commands:launch\_command:501 - \['accelerate', 'launch', '--cpu', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json'\]
> 
> WARNING | 2024-10-04 20:09:10 | autotrain.commands:get\_accelerate\_command:52 - No GPU found. Forcing training on CPU. This will be super slow!
> 
> INFO | 2024-10-04 20:09:10 | autotrain.backends.local:create:8 - Starting local training...
> 
> INFO | 2024-10-04 20:09:10 | autotrain.app.ui\_routes:handle\_form:500 - hardware: local-ui
> 
> INFO | 2024-10-04 20:08:05 | autotrain.app.utils:kill\_process\_by\_pid:52 - Sent SIGTERM to process with PID 352
> 
> INFO | 2024-10-04 20:08:05 | autotrain.app.utils:get\_running\_jobs:26 - Killing PID: 352
> 
> subprocess.CalledProcessError: Command '\['/app/env/bin/python', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json'\]' returned non-zero exit status 2.
> 
> raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
> 
> File "/app/env/lib/python3.10/site-packages/accelerate/commands/launch.py", line 769, in simple\_launcher
> 
> simple\_launcher(args)
> 
> File "/app/env/lib/python3.10/site-packages/accelerate/commands/launch.py", line 1174, in launch\_command
> 
> args.func(args)
> 
> File "/app/env/lib/python3.10/site-packages/accelerate/commands/accelerate\_cli.py", line 48, in main
> 
> sys.exit(main())
> 
> File "/app/env/bin/accelerate", line 8, in \<module\>
> 
> Traceback (most recent call last):
> 
> \_\_main\_\_.py: error: unrecognized arguments: -m autotrain.trainers.clm
> 
> usage: \_\_main\_\_.py \[-h\] --training\_config TRAINING\_CONFIG
> 
> To avoid this warning pass in values for each of the problematic parameters or run \`accelerate config\`.
> 
> \`--dynamo\_backend\` was set to a value of \`'no'\`
> 
> \`--mixed\_precision\` was set to a value of \`'no'\`
> 
> \`--num\_machines\` was set to a value of \`1\`
> 
> \`--num\_processes\` was set to a value of \`0\`
> 
> The following values were not passed to \`accelerate launch\` and had defaults used instead:
> 
> INFO | 2024-10-04 20:07:59 | autotrain.backends.local:create:13 - Training PID: 352
> 
> INFO | 2024-10-04 20:07:59 | autotrain.commands:launch\_command:502 - {'model': 'meta-llama/Llama-3.2-3B-Instruct', 'project\_name': 'legacy-ledger-at-v1', 'data\_path': 'bunny0702/Legal\_Research', 'train\_split': '0.8', 'valid\_split': None, 'add\_eos\_token': True, 'block\_size': 1024, 'model\_max\_length': 2048, 'padding': 'right', 'trainer': 'sft', 'use\_flash\_attention\_2': False, 'log': 'tensorboard', 'disable\_gradient\_checkpointing': False, 'logging\_steps': -1, 'eval\_strategy': 'epoch', 'save\_total\_limit': 1, 'auto\_find\_batch\_size': False, 'mixed\_precision': 'fp16', 'lr': 3e-05, 'epochs': 3, 'batch\_size': 2, 'warmup\_ratio': 0.1, 'gradient\_accumulation': 4, 'optimizer': 'adamw\_torch', 'scheduler': 'linear', 'weight\_decay': 0.0, 'max\_grad\_norm': 1.0, 'seed': 42, 'chat\_template': 'none', 'quantization': 'int4', 'target\_modules': 'all-linear', 'merge\_adapter': False, 'peft': True, 'lora\_r': 16, 'lora\_alpha': 32, 'lora\_dropout': 0.05, 'model\_ref': None, 'dpo\_beta': 0.1, 'max\_prompt\_length': 128, 'max\_completion\_length': None, 'prompt\_text\_column': 'prompt', 'text\_column': 'text', 'rejected\_text\_column': 'rejected\_text', 'push\_to\_hub': True, 'username': 'Gladystoper', 'token': '\*\*\*\*\*', 'unsloth': False, 'distributed\_backend': None}
> 
> INFO | 2024-10-04 20:07:59 | autotrain.commands:launch\_command:501 - \['accelerate', 'launch', '--cpu', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json'\]
> 
> WARNING | 2024-10-04 20:07:59 | autotrain.commands:get\_accelerate\_command:52 - No GPU found. Forcing training on CPU. This will be super slow!
> 
> INFO | 2024-10-04 20:07:59 | autotrain.backends.local:create:8 - Starting local training...
> 
> INFO | 2024-10-04 20:07:59 | autotrain.app.ui\_routes:handle\_form:500 - hardware: local-ui
> 
> INFO | 2024-10-04 20:07:29 | autotrain.app.ui\_routes:handle\_form:500 - hardware: local-ui
> 
> INFO | 2024-10-04 20:06:30 | autotrain.app.utils:kill\_process\_by\_pid:52 - Sent SIGTERM to process with PID 343
> 
> INFO | 2024-10-04 20:06:30 | autotrain.app.utils:get\_running\_jobs:26 - Killing PID: 343
> 
> ERROR | 2024-10-04 20:06:26 | autotrain.trainers.common:wrapper:121 - Unknown split "0.8". Should be one of \['train'\].
> 
> ValueError: Unknown split "0.8". Should be one of \['train'\].
> 
> raise ValueError(f'Unknown split "{split}". Should be one of {list(name2len)}.')
> 
> File "/app/env/lib/python3.10/site-packages/datasets/arrow\_reader.py", line 480, in \_rel\_to\_abs\_instr
> 
> return \[\_rel\_to\_abs\_instr(rel\_instr, name2len) for rel\_instr in self.\_relative\_instructions\]
> 
> File "/app/env/lib/python3.10/site-packages/datasets/arrow\_reader.py", line 663, in \<listcomp\>
> 
> return \[\_rel\_to\_abs\_instr(rel\_instr, name2len) for rel\_instr in self.\_relative\_instructions\]
> 
> File "/app/env/lib/python3.10/site-packages/datasets/arrow\_reader.py", line 663, in to\_absolute
> 
> absolute\_instructions = instruction.to\_absolute(name2len)
> 
> File "/app/env/lib/python3.10/site-packages/datasets/arrow\_reader.py", line 134, in make\_file\_instructions
> 
> file\_instructions = make\_file\_instructions(
> 
> File "/app/env/lib/python3.10/site-packages/datasets/arrow\_reader.py", line 225, in get\_file\_instructions
> 
> files = self.get\_file\_instructions(name, instructions, split\_infos)
> 
> File "/app/env/lib/python3.10/site-packages/datasets/arrow\_reader.py", line 252, in read
> 
> dataset\_kwargs = ArrowReader(cache\_dir, self.info).read(
> 
> File "/app/env/lib/python3.10/site-packages/datasets/builder.py", line 1370, in \_as\_dataset
> 
> ds = self.\_as\_dataset(
> 
> File "/app/env/lib/python3.10/site-packages/datasets/builder.py", line 1296, in \_build\_single\_dataset
> 
> mapped = function(data\_struct)
> 
> File "/app/env/lib/python3.10/site-packages/datasets/utils/py\_utils.py", line 484, in map\_nested
> 
> datasets = map\_nested(
> 
> File "/app/env/lib/python3.10/site-packages/datasets/builder.py", line 1266, in as\_dataset
> 
> ds = builder\_instance.as\_dataset(split=split, verification\_mode=verification\_mode, in\_memory=keep\_in\_memory)
> 
> File "/app/env/lib/python3.10/site-packages/datasets/load.py", line 2621, in load\_dataset
> 
> train\_data = load\_dataset(
> 
> File "/app/env/lib/python3.10/site-packages/autotrain/trainers/clm/utils.py", line 351, in process\_input\_data
> 
> train\_data, valid\_data = utils.process\_input\_data(config)
> 
> File "/app/env/lib/python3.10/site-packages/autotrain/trainers/clm/train\_clm\_sft.py", line 14, in train
> 
> train\_sft(config)
> 
> File "/app/env/lib/python3.10/site-packages/autotrain/trainers/clm/\_\_main\_\_.py", line 28, in train
> 
> return func(\*args, \*\*kwargs)
> 
> File "/app/env/lib/python3.10/site-packages/autotrain/trainers/common.py", line 117, in wrapper
> 
> ERROR | 2024-10-04 20:06:26 | autotrain.trainers.common:wrapper:120 - train has failed due to an exception: Traceback (most recent call last):
> 
> Generating train split: 100%|██████████| 155703/155703 \[00:00\<00:00, 3644055.24 examples/s\]
> 
> Generating train split: 0%| | 0/155703 \[00:00\<?, ? examples/s\]
> 
> Downloading data: 100%|██████████| 180k/180k \[00:00\<00:00, 1.35MB/s\]
> 
> Downloading data: 100%|██████████| 180k/180k \[00:00\<00:00, 1.36MB/s\]
> 
> Downloading data: 100%|██████████| 2.83M/2.83M \[00:00\<00:00, 8.18MB/s\]
> 
> Downloading data: 100%|██████████| 2.83M/2.83M \[00:00\<00:00, 8.23MB/s\]
> 
> Downloading data: 100%|██████████| 5.84k/5.84k \[00:00\<00:00, 50.1kB/s\]
> 
> Downloading data: 100%|██████████| 5.84k/5.84k \[00:00\<00:00, 50.2kB/s\]
> 
> Downloading data: 0%| | 0.00/5.84k \[00:00\<?, ?B/s\]
> 
> Downloading data: 100%|██████████| 152k/152k \[00:00\<00:00, 764kB/s\]
> 
> Downloading data: 100%|██████████| 152k/152k \[00:00\<00:00, 767kB/s\]
> 
> Downloading data: 0%| | 0.00/152k \[00:00\<?, ?B/s\]
> 
> INFO | 2024-10-04 20:06:22 | autotrain.trainers.clm.train\_clm\_sft:train:11 - Starting SFT training...
> 
> To avoid this warning pass in values for each of the problematic parameters or run \`accelerate config\`.
> 
> \`--dynamo\_backend\` was set to a value of \`'no'\`
> 
> \`--mixed\_precision\` was set to a value of \`'no'\`
> 
> \`--num\_machines\` was set to a value of \`1\`
> 
> \`--num\_processes\` was set to a value of \`0\`
> 
> The following values were not passed to \`accelerate launch\` and had defaults used instead:
> 
> INFO | 2024-10-04 20:06:17 | autotrain.backends.local:create:13 - Training PID: 343
> 
> INFO | 2024-10-04 20:06:17 | autotrain.commands:launch\_command:502 - {'model': 'meta-llama/Llama-3.2-3B-Instruct', 'project\_name': 'legacy-ledger-at-v1', 'data\_path': 'bunny0702/Legal\_Research', 'train\_split': '0.8', 'valid\_split': None, 'add\_eos\_token': True, 'block\_size': 1024, 'model\_max\_length': 2048, 'padding': 'right', 'trainer': 'sft', 'use\_flash\_attention\_2': False, 'log': 'tensorboard', 'disable\_gradient\_checkpointing': False, 'logging\_steps': -1, 'eval\_strategy': 'epoch', 'save\_total\_limit': 1, 'auto\_find\_batch\_size': False, 'mixed\_precision': 'fp16', 'lr': 3e-05, 'epochs': 3, 'batch\_size': 2, 'warmup\_ratio': 0.1, 'gradient\_accumulation': 4, 'optimizer': 'adamw\_torch', 'scheduler': 'linear', 'weight\_decay': 0.0, 'max\_grad\_norm': 1.0, 'seed': 42, 'chat\_template': 'none', 'quantization': 'int4', 'target\_modules': 'all-linear', 'merge\_adapter': False, 'peft': True, 'lora\_r': 16, 'lora\_alpha': 32, 'lora\_dropout': 0.05, 'model\_ref': None, 'dpo\_beta': 0.1, 'max\_prompt\_length': 128, 'max\_completion\_length': None, 'prompt\_text\_column': 'prompt', 'text\_column': 'text', 'rejected\_text\_column': 'rejected\_text', 'push\_to\_hub': True, 'username': 'Gladystoper', 'token': '\*\*\*\*\*', 'unsloth': False, 'distributed\_backend': None}
> 
> INFO | 2024-10-04 20:06:17 | autotrain.commands:launch\_command:501 - \['accelerate', 'launch', '--cpu', '-m', 'autotrain.trainers.clm', '--training\_config', 'legacy-ledger-at-v1/training\_params.json'\]
> 
> WARNING | 2024-10-04 20:06:17 | autotrain.commands:get\_accelerate\_command:52 - No GPU found. Forcing training on CPU. This will be super slow!
> 
> INFO | 2024-10-04 20:06:17 | autotrain.backends.local:create:8 - Starting local training...
> 
> INFO | 2024-10-04 20:06:17 | autotrain.app.ui\_routes:handle\_form:500 - hardware: local-ui
> 
> INFO | 2024-10-04 20:05:25 | autotrain.app.ui\_routes:fetch\_params:391 - Param distributed\_backend not found in UI\_PARAMS
> 
> INFO | 2024-10-04 20:05:25 | autotrain.app.ui\_routes:fetch\_params:381 - Task: llm:sft
> 
> INFO: 10.203.0.7:58006 - "GET /?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMzIzLCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODcyMywiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.f01ZyphHQ-H1f0hqNduC6oLd-R1\_oqQqrJOE4IGNDVV1ESg\_RZRun38N4lLoWEPOn91dfw6jvrJg\_Ilg42nlAA HTTP/1.1" 307 Temporary Redirect
> 
> INFO: 10.203.4.41:54838 - "GET /?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMzIzLCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODcyMywiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.f01ZyphHQ-H1f0hqNduC6oLd-R1\_oqQqrJOE4IGNDVV1ESg\_RZRun38N4lLoWEPOn91dfw6jvrJg\_Ilg42nlAA HTTP/1.1" 307 Temporary Redirect
> 
> INFO | 2024-10-04 20:02:27 | autotrain.app.ui\_routes:fetch\_params:391 - Param distributed\_backend not found in UI\_PARAMS
> 
> INFO | 2024-10-04 20:02:27 | autotrain.app.ui\_routes:fetch\_params:381 - Task: llm:sft
> 
> INFO: 10.203.0.7:39860 - "GET /?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMTQ2LCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODU0NiwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.mmzj3\_JYvkG4SsyTvlNl9oxBl8jfpuK0wlbQMjZeBfjEK28ijS6sSYk6hPH9PS3oKIfdqJSof2omk9KDpPrCDg HTTP/1.1" 307 Temporary Redirect
> 
> INFO: 10.203.4.41:52548 - "GET /?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMTQ2LCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODU0NiwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.mmzj3\_JYvkG4SsyTvlNl9oxBl8jfpuK0wlbQMjZeBfjEK28ijS6sSYk6hPH9PS3oKIfdqJSof2omk9KDpPrCDg HTTP/1.1" 307 Temporary Redirect
> 
> INFO | 2024-10-04 20:01:49 | autotrain.app.ui\_routes:fetch\_params:391 - Param distributed\_backend not found in UI\_PARAMS
> 
> INFO | 2024-10-04 20:01:49 | autotrain.app.ui\_routes:fetch\_params:381 - Task: llm:sft
> 
> INFO: 10.203.0.7:53992 - "GET / HTTP/1.1" 307 Temporary Redirect
> 
> INFO: 10.203.0.7:53992 - "GET /auth?code=pCIvjXNYbHDvcDZN&state=rLaqKVN6FpwF0ETcx5Ga0HWfrCqb1t HTTP/1.1" 307 Temporary Redirect
> 
> INFO: 10.203.0.7:53992 - "GET /login/huggingface?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMTA0LCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODUwNCwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.DDPA7r37oUjUrnhV4tKSThbbemp4bIlcRJ1jITIKI4kUp9EnwopGjjP1bqvWjVOQ7DpzZaI94JfFeCIcK-mVBQ HTTP/1.1" 302 Found
> 
> ERROR | 2024-10-04 20:01:45 | autotrain.app.ui\_routes:load\_index:347 - Failed to get user and orgs: object of type '\_TemplateResponse' has no len()
> 
> INFO: 10.203.0.7:53992 - "GET /?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMTA0LCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODUwNCwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.DDPA7r37oUjUrnhV4tKSThbbemp4bIlcRJ1jITIKI4kUp9EnwopGjjP1bqvWjVOQ7DpzZaI94JfFeCIcK-mVBQ HTTP/1.1" 307 Temporary Redirect
> 
> INFO: 10.203.4.41:36212 - "GET /?\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDcyMTA0LCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODE1ODUwNCwiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.DDPA7r37oUjUrnhV4tKSThbbemp4bIlcRJ1jITIKI4kUp9EnwopGjjP1bqvWjVOQ7DpzZaI94JfFeCIcK-mVBQ HTTP/1.1" 307 Temporary Redirect
> 
> ERROR | 2024-10-04 02:03:05 | autotrain.app.ui\_routes:user\_authentication:324 - Failed to verify token: Invalid token (/oauth/userinfo). Please login with a write token.
> 
> ERROR | 2024-10-04 02:03:05 | autotrain.app.utils:token\_verification:84 - Failed to request /oauth/userinfo - 504
> 
> ERROR | 2024-10-04 02:03:05 | autotrain.app.ui\_routes:user\_authentication:324 - Failed to verify token: Invalid token (/oauth/userinfo). Please login with a write token.
> 
> ERROR | 2024-10-04 02:03:05 | autotrain.app.utils:token\_verification:84 - Failed to request /oauth/userinfo - 504
> 
> INFO | 2024-10-04 01:42:36 | autotrain.app.ui\_routes:fetch\_params:391 - Param distributed\_backend not found in UI\_PARAMS
> 
> INFO | 2024-10-04 01:42:36 | autotrain.app.ui\_routes:fetch\_params:381 - Task: llm:sft
> 
> INFO: 10.203.14.122:50602 - "GET /?logs=build&\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDA2MTUzLCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODA5MjU1MywiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.dnUKIVelGNJw5CMS1JkW3wgFATQbdXScm-1aELkG\_U5msHucvhHlC4R7445LYEB\_HWfk1M0LXep7fk1A8mWhDg HTTP/1.1" 307 Temporary Redirect
> 
> INFO: 10.203.14.157:50772 - "GET /?logs=build&\_\_sign=eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sIm9uQmVoYWxmT2YiOnsia2luZCI6InVzZXIiLCJfaWQiOiI2NTEwNTY4YThmMzIyOGQ4MDdhNzFkNjkiLCJ1c2VyIjoiR2xhZHlzdG9wZXIifSwiaWF0IjoxNzI4MDA2MTUzLCJzdWIiOiIvc3BhY2VzL0dsYWR5c3RvcGVyL2F1dG90cmFpbi1hZHZhbmNlZCIsImV4cCI6MTcyODA5MjU1MywiaXNzIjoiaHR0cHM6Ly9odWdnaW5nZmFjZS5jbyJ9.dnUKIVelGNJw5CMS1JkW3wgFATQbdXScm-1aELkG\_U5msHucvhHlC4R7445LYEB\_HWfk1M0LXep7fk1A8mWhDg HTTP/1.1" 307 Temporary Redirect
> 
> INFO: Uvicorn running on http://0.0.0.0:7860 (Press CTRL+C to quit)
> 
> INFO: Application startup complete.
> 
> INFO: Waiting for application startup.
> 
> INFO: Started server process \[121\]
> 
> INFO | 2024-10-04 01:42:32 | autotrain.app.app:\<module\>:24 - AutoTrain started successfully
> 
> INFO | 2024-10-04 01:42:32 | autotrain.app.app:\<module\>:23 - AutoTrain version: 0.8.21
> 
> INFO | 2024-10-04 01:42:32 | autotrain.app.app:\<module\>:13 - Starting AutoTrain...
> 
> INFO | 2024-10-04 01:42:32 | autotrain.app.ui\_routes:\<module\>:298 - AutoTrain started successfully
> 
> INFO | 2024-10-04 01:42:27 | autotrain.app.ui\_routes:\<module\>:32 - Starting AutoTrain...
> 
> \### Additional Information
> 
> \_No response\_

> <https://github.com/huggingface/autotrain-advanced/issues/881>
>
> Getting this error message when trying to train my model using Autotrain
> 
> 
> Accel…erators: Error fetching data
> Error fetching training status
> 
> 
> My data file is a csv & correctly formatted. 
> What are possible ways to troubleshoot this problem?
> I'm new to fine-tuning so would love any assistance