explain_model
The factors behind a trained model as structured data: each one with its direction and its size, the quality summary, and, once the model has retrained at least once, what moved since the last run. It reads what the run already produced and computes nothing new. No language model is involved.
What it needs
| Argument | Type | Required | Description |
|---|---|---|---|
model_id | string | Yes | |
compare_to_previous | boolean | No |
What it returns
The model explained from what it already produced - the outcome in plain words, the top factors with the direction and size of each, the model-quality summary, and what changed since the previous run when there has been one.
Credential
An app session token, from an authenticated app.ax1om.ai session. Call it with only an API key and you get a plain refusal saying so, rather than a confusing 401.
Network call: yes. The full credential split is on authentication.
Example call
Most readers never write the protocol themselves: the agent does, once the server is connected. Both forms are the same call.
Use the ax1om MCP server.Call explain_model: the factors behind a trained model as structured data: each one with its direction and its size, the quality summary, and, once the model has retrained at least once, what moved since the last run.{ "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "explain_model", "arguments": {} }}The arguments object is whatever this tool's own input schema declares. An MCP client reads that schema from the server, so it is the server that answers for the exact shape rather than this page.