ReinforcementHyperparameters
Example Usage
typescript
import { ReinforcementHyperparameters } from "@meetkai/mka1/models/components";
let value: ReinforcementHyperparameters = {};Fields
| Field | Type | Required | Description |
|---|---|---|---|
batchSize | components.ReinforcementHyperparametersBatchSize | ➖ | Number of examples in each batch |
computeMultiplier | components.ComputeMultiplier | ➖ | Scales compute for exploration during training |
evalInterval | components.EvalInterval | ➖ | Number of steps between evaluations |
evalSamples | components.EvalSamples | ➖ | Number of samples generated per eval step |
learningRateMultiplier | components.ReinforcementHyperparametersLearningRateMultiplier | ➖ | Scaling factor for learning rate |
nEpochs | components.ReinforcementHyperparametersNEpochs | ➖ | Number of epochs to train for |
reasoningEffort | components.ReinforcementHyperparametersReasoningEffort | ➖ | Reasoning effort level |