Log backflow

Log backflow converts SLS (Simple Log Service) inference logs into structured datasets that can be used for model fine-tuning or evaluation.

Feature overview

The log backflow feature backflows inference log data from SLS (Simple Log Service) to the Bailian platform, formats it, and generates structured datasets (in JSONL format) that can be used for model fine-tuning or model evaluation. The backflow produces structured data, rather than a direct copy of the raw logs.

Scope

Log backflow supports creating the following two types of datasets:

• Training set: The training scenario is text generation, and the training methods support SFT (Supervised Fine-Tuning), DPO (Direct Preference Optimization), and CPT (Continued Pre-Training).
• Evaluation set: Supports the text generation scenario.

Log backflow is currently available only in the China (Beijing) and Singapore regions. The log backflow entry is not displayed in other regions. The upper limit of a single backflow is 100,000 entries, and you can backflow multiple times to different versions of the same dataset to accumulate more data.

Log backflow supports two storage methods: Platform Storage (default) and object storage OSS Mount (requires additional authorization). For the differences between the storage methods and selection guidance, see Create a log backflow dataset.

The datasets generated by backflow can be directly used for downstream tasks: training sets can be used for model tuning, and evaluation sets can be used for model evaluation. The data also supports subsequent data cleansing.

Enable and authorize related services

Before using log backflow, confirm that the current region is China (Beijing) or Singapore (the log backflow entry is not displayed in other regions), and complete the following service activation and permission authorization on the Model monitoring page. After all steps are completed, the authorization configuration drawer automatically closes and you enter the log backflow form.

Enable audit logs and inference logs

Audit logs and inference logs each require three steps to be completed. You can use log backflow only after all six conditions are met. Audit logs are a prerequisite for inference logs, so you must complete all audit log steps first.

Audit logs

1. Authorize the SLS service-linked role: Click Authorize Now to authorize the AliyunServiceRoleForSFMAccessSLS role. When not authorized, a red Not Authorized label is displayed.
2. Activate SLS Log Service: When not activated, a Not Activated status and a redirect link are displayed. Click the link to go to the SLS console and complete activation.
3. Enable audit logs: Click Create and Enable Audit Logs, and the system creates a LogStore instance and polls until it is ready (up to 60 seconds).

Inference logs

1. Authorize the SLS service-linked role for inference logs.
2. Confirm that SLS Log Service is activated.
3. Enable inference logs: When audit logs are not enabled, this button is grayed out. You must enable audit logs first.

OSS multi-role authorization (OSS Mount mode only)

When you select the OSS Mount storage method, you must additionally complete the following authorization:

1. In the authorization pop-up, select the data access authorization agreement.

2. Click One-click authorization, and the system automatically authorizes the following two service-linked roles:

   • AliyunServiceRoleForAccessCusOss (OSS write role)
   • AliyunServiceRoleForSFMDataHubOSSImport (DataHub import role)

The following table summarizes the service-linked roles involved in log backflow.

Role name	Purpose	Authorization timing
AliyunServiceRoleForSFMAccessSLS	Bailian accesses SLS log data	Authorized once for audit logs and once for inference logs
AliyunServiceRoleForAccessCusOss	OSS data write	When OSS Mount storage is selected
AliyunServiceRoleForSFMDataHubOSSImport	DataHub data import	When OSS Mount storage is selected

Create a log backflow dataset

Log backflow provides three entry points, all of which lead to the configuration form to create a dataset:

Model monitoring list page

At the top of the Monitoring list page, click Log Ingestion. For first-time use, complete the authorization configuration first. After authorization is passed, the log backflow form is displayed automatically.

Model monitoring detail page

In the time selector area of the model monitoring detail page, click Log Ingestion. When entering from this entry point, the form automatically pre-fills the time range, API Key, and model of the current page, and the model cannot be modified.

Data management page

When creating a new dataset on the Data management page, select Log Backflow as the import method. This option is visible only for training sets (text generation + SFT) or evaluation sets (text generation).

Configure backflow parameters

After entering the log backflow form, configure the following parameters from top to bottom. Some parameters have prerequisites: API Key filtering requires a time range to be selected first, and model selection requires both a time range and an API Key to be selected first. Modifying the time range, data type, training scenario, or training method will cascade-reset other parameters, so it is recommended to fill in the form strictly in order. The parameter values are described in the table below.

Estimated backflow data: The system displays the estimated number of backflow data entries based on the filter conditions. When the number exceeds 100,000, a red warning is displayed, and the excess will not be backflowed. When the query results are too many, the OK button is disabled, and you need to narrow the filter range.

Parameter	Description	Required	Value description
Backflow location	Dataset storage method	Yes	Platform storage (default)or OSS Mount. OSS Mount is disabled for evaluation sets. Cannot be changed after creation
Dataset name	Display name of the dataset in the list	Yes	Chinese, English, digits, underscores, slashes, and hyphens, up to 50 characters. Recommended naming format: function_scenario_model_name_time. Cannot be modified after creation
Dataset description	Supplementary description of the dataset purpose	No	Up to 200 characters
Type and format	Dataset purpose type	Yes	Training set or evaluation set. When evaluation set is selected, training scenario and training method are hidden. Cannot be changed after creation
Training scenario	Training scenario type (displayed only for training sets)	Yes	Currently only text generation is supported
Training method	Fine-tuning method (displayed only for training sets)	Yes	SFT, DPO, CPT; options are dynamically displayed by the system. Locked after creation
Time range	Time period of the backflow logs	Yes	Last 30 days (including today), precise to hours, minutes, and seconds. Modifying resets the API Key and model selection
API Key filter	Filter log data by API Key	Yes	All (no filter), Other (excludes listed Keys), or select specific Keys (multi-select)
Model selection	Target models for backflow	Yes	Up to 10. Models that do not match the capability type filter are grayed out
OSS data path	Target directory for data storage (displayed only in OSS Mount mode)	Yes	The Bucket must be in the same region

View backflow results

After submitting a log backflow task, view the dataset and import progress on the Data management list page.

On the data management list page, each dataset displays its name, storage location (Platform Storage / OSS Mount), import method (Log Backflow), import status, data volume, creation time, and other information. When import fails, you can view the specific failure reason.

List page display

Datasets created by log backflow display the import method as Log Ingestion on the list page, and the storage location is displayed as Platform Storage or OSS Mount based on the selection at creation.

You can view the import status of the task on the list page. When a task fails, you can view the specific failure reason returned by the system.

Under platform storage mode, the system automatically publishes the dataset version after import is complete, requiring no manual operation.

Detail page information

The information displayed on the dataset detail page varies depending on the storage method:

• OSS Mount: Displays release status, data volume, creation time, FileID, data type, import status, and OSS mount address.
• Platform storage (OSS import): Displays release status, data volume, creation time, FileID, data type, import status, and OSS import address.
• Other cases: Display release status, data volume, creation time, FileID, data type, and import status.

Append log backflow data

On the Data management page, you can append a new batch of log backflow data to an existing dataset in the following two ways:

Import data page

Enter the Import data page of an existing dataset and select Log Backflow as the import method. The form parameters are the same as those in the creation flow. For details, see Create a log backflow dataset. This method additionally supports filtering by workspace.

This method is applicable to all storage types of datasets, including OSS Mount datasets.

Add version dialog

On the dataset detail page, click New Version, and select the log backflow import method in the dialog. This method is available only for platform storage datasets.

NoteOSS Mount datasets do not support the New Version operation; the button is grayed out. Use the Import data page to append data.

Incremental backflow best practices

It is recommended to backflow in batches, targeting different time periods or different models separately, and gradually accumulating high-quality training sets. Each batch can precisely select models with good performance and data from business peak hours, ensuring data quality superior to a single large backflow.

FAQ

What requests does "Other" in API Key filter include?

Question: What requests does Other in the API Key filter include?

Other filters out requests from listed Keys, including deleted Keys or logs generated by other workspaces. All does not filter by Key. For a complete description of the three modes (All/Other/individual selection), see the API Key filter parameter in Create a log backflow dataset.

Is the 100,000-entry limit a total dataset limit?

Question: The single backflow limit is 100,000 entries. Does this mean the total dataset volume is also limited to 100,000 entries?

No. The 100,000-entry limit applies to a single backflow, not the total dataset volume. You can backflow multiple times to different versions of the same dataset, and the cumulative data volume is not subject to this limit. For example, by backflowing in batches for different time periods, each time within 100,000 entries, the final dataset can accumulate far more than 100,000 entries of data.

What is the difference between platform storage and OSS Mount?

Question: Besides the different data storage locations, what other differences exist between the two storage methods?

The core differences between the two storage methods are compared below.

Dimension	Platform storage	OSS Mount
Additional authorization	No additional authorization required	Requires authorizing two service-linked roles and selecting the data access agreement
Add version	Supported	Not supported (button grayed out); use the Import data page to append
Data access	Access via console	JSONL files can be viewed and managed directly in the OSS Bucket
Evaluation set	Supported	Not supported (option disabled)
Automatic release	Version automatically released after import	No automatic release

Are audit logs and inference logs the same operation?

Question: Enabling logs requires separate operations for audit logs and inference logs. What is the relationship between them?

They must be enabled separately, and audit logs must be completed before inference logs can be enabled. For operation steps, see Enable and authorize related services.

Why is the estimated data volume inconsistent with the actual backflow results?

Question: The estimated data volume displayed before submission differs from the actual number of entries after backflow is completed.

The estimated data volume is an approximate value based on filter conditions. The actual number of data entries after backflow may differ slightly, which is normal. The estimate is used to help determine whether filter conditions need to be adjusted (for example, when exceeding 100,000 entries, you are prompted to narrow the range) and does not represent an exact count.