Datum

A single training datum containing model input and optional loss function inputs.

modelInput

The tokenized input to feed to the model.

Required: Yes

lossFnInputs

The tensor inputs for the loss function, keyed by name. Unsupported keys are rejected. Supported keys: target_tokens, weights, advantages, logprobs, ref_log_prob, temperature. Required keys depend on the loss function type: SFT (cross_entropy) requires target_tokens and weights; RL (ppo, cispo, importance_sampling) requires target_tokens, weights, advantages, and logprobs.

Type: String to TensorData object map

Required: No

Datum

Contents

See Also