> ## Documentation Index
> Fetch the complete documentation index at: https://training-docs.cerebras.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Troubleshooting

* [Cannot load Cerebras checkpoints in GPUs](./troubleshooting/cannot-load-cerebras-checkpoints-in-gpus)

  * [Work around](./troubleshooting/cannot-load-cerebras-checkpoints-in-gpus#work-around)

* [Custom PT training script spawns multiple compile jobs](./troubleshooting/custom-pt-training-script-spawns-multiple-compile-jobs)

  * [Observed Error](./troubleshooting/custom-pt-training-script-spawns-multiple-compile-jobs#observed-error)

  * [Explanation](./troubleshooting/custom-pt-training-script-spawns-multiple-compile-jobs#explanation)

  * [Work around](./troubleshooting/custom-pt-training-script-spawns-multiple-compile-jobs#work-around)

* [Loss compilation issues with Autogen](./troubleshooting/loss-compilation-issues-with-autogen)

  * [Custom loss functions with AutoGen](./troubleshooting/loss-compilation-issues-with-autogen#custom-loss-functions-with-autogen)

  * [Improving loss function performance](./troubleshooting/loss-compilation-issues-with-autogen#improving-loss-function-performance)

* [Error parsing metadata](./troubleshooting/error-parsing-metadata)

  * [Observed Error](./troubleshooting/error-parsing-metadata#observed-error)

  * [Explanation](./troubleshooting/error-parsing-metadata#explanation)

  * [Work around](./troubleshooting/error-parsing-metadata#work-around)

* [Error Receiving Activation](./troubleshooting/error-receiving-activation)

  * [cerebras.appliance.errors.ApplianceUnknownError: Ran into error while receiving activation tensor \<custom-call …>](./troubleshooting/error-receiving-activation#cerebras-appliance-errors-applianceunknownerror-ran-into-error-while-receiving-activation-tensor-custom-call)

* [Failed mount directory during execution](./troubleshooting/failed-mount-directory-during-execution)

  * [Observed Error](./troubleshooting/failed-mount-directory-during-execution#observed-error)

  * [Work around](./troubleshooting/failed-mount-directory-during-execution#work-around)

* [Failing to automatically load checkpoints](./troubleshooting/failing-to-automatically-load-checkpoints)

  * [Explanation](./troubleshooting/failing-to-automatically-load-checkpoints#explanation)

  * [Work around](./troubleshooting/failing-to-automatically-load-checkpoints#work-around)

* [Failure to trace due to functionalization error](./troubleshooting/failure-to-trace-due-to-functionalization-error)

  * [Observed Error](./troubleshooting/failure-to-trace-due-to-functionalization-error#observed-error)

  * [Explanation](./troubleshooting/failure-to-trace-due-to-functionalization-error#observed-error)

  * [Work around](./troubleshooting/failure-to-trace-due-to-functionalization-error#work-around)

* [Input Starvation](./troubleshooting/input-starvation)

* [Out of memory errors and system resources](./troubleshooting/out-of-memory-errors-and-system-resources)

  * [Determining if your job is queued](./troubleshooting/out-of-memory-errors-and-system-resources#determining-if-your-job-is-queued)

  * [Determining if job failed because of an OOM error](./troubleshooting/out-of-memory-errors-and-system-resources#determining-if-job-failed-because-of-an-oom-error)

  * [Determining if job failed because of system could not fit requested memory](./troubleshooting/out-of-memory-errors-and-system-resources#determining-if-job-failed-because-of-system-could-not-fit-requested-memory)

  * [Troubleshooting OOM errors](./troubleshooting/out-of-memory-errors-and-system-resources#troubleshooting-oom-errors)

* [Model is too large to fit in memory](./troubleshooting/model-is-too-large-to-fit-in-memory)

  * [Observed Error](./troubleshooting/model-is-too-large-to-fit-in-memory#observed-error)

  * [Causes and Possible Solutions](./troubleshooting/model-is-too-large-to-fit-in-memory#causes-and-possible-solutions)

* [ModuleNotFoundError](./troubleshooting/modulenotfounderror)

  * [ModuleNotFoundError: No module named \<’\_bz2’, ‘\_sqlite3’>](./troubleshooting/modulenotfounderror#modulenotfounderror-no-module-named-bz2-sqlite3)

  * [ModuleNotFoundError: No module named \<…>](./troubleshooting/modulenotfounderror#modulenotfounderror-no-module-named)

* [Numerical issues](./troubleshooting/numerical-issues)

  * [Observed Error](./troubleshooting/numerical-issues#observed-error)

  * [Explanation](./troubleshooting/numerical-issues#explanation)

  * [Work around](./troubleshooting/numerical-issues#work-around)

* [Throughput spike after saving checkpoints](./troubleshooting/throughput-spike-after-saving-checkpoints)

* [Training fails when logged-in as root](./troubleshooting/training-fails-when-logged-in-as-root)

  * [Observed Error](./troubleshooting/training-fails-when-logged-in-as-root#observed-error)

  * [Explanation](./troubleshooting/training-fails-when-logged-in-as-root#explanation)

* [Vocabulary Size Troubleshooting](./troubleshooting/vocabulary-size-troubleshooting)

  * [Large vocabulary size](./troubleshooting/vocabulary-size-troubleshooting#large-vocabulary-size)

  * [Small vocabulary size](./troubleshooting/vocabulary-size-troubleshooting#small-vocabulary-size)