FM / fine-tuning

/opt/canhelp/node_modules/openai/resources/fine-tuning/

/opt/canhelp/node_modules/openai/resources/fine-tuning

Name	Size	Mode	Actions
alpha/	-	0755	rm
checkpoints/	-	0755	rm
jobs/	-	0755	rm
alpha.d.mts	71	0644	edit dl rm
alpha.d.mts.map	130	0644	edit dl rm
alpha.d.ts	69	0644	edit dl rm
alpha.d.ts.map	129	0644	edit dl rm
alpha.js	309	0644	edit dl rm
alpha.js.map	154	0644	edit dl rm
alpha.mjs	156	0644	edit dl rm
alpha.mjs.map	139	0644	edit dl rm
checkpoints.d.mts	83	0644	edit dl rm
checkpoints.d.mts.map	142	0644	edit dl rm
checkpoints.d.ts	81	0644	edit dl rm
checkpoints.d.ts.map	141	0644	edit dl rm
checkpoints.js	321	0644	edit dl rm
checkpoints.js.map	166	0644	edit dl rm
checkpoints.mjs	168	0644	edit dl rm
checkpoints.mjs.map	151	0644	edit dl rm
fine-tuning.d.mts	2022	0644	edit dl rm
fine-tuning.d.mts.map	1346	0644	edit dl rm
fine-tuning.d.ts	2012	0644	edit dl rm
fine-tuning.d.ts.map	1345	0644	edit dl rm
fine-tuning.js	1354	0644	edit dl rm
fine-tuning.js.map	745	0644	edit dl rm
fine-tuning.mjs	1032	0644	edit dl rm
fine-tuning.mjs.map	777	0644	edit dl rm
index.d.mts	700	0644	edit dl rm
index.d.mts.map	508	0644	edit dl rm
index.d.ts	694	0644	edit dl rm
index.d.ts.map	507	0644	edit dl rm
index.js	1088	0644	edit dl rm
index.js.map	314	0644	edit dl rm
index.mjs	351	0644	edit dl rm
index.mjs.map	240	0644	edit dl rm
jobs.d.mts	69	0644	edit dl rm
jobs.d.mts.map	128	0644	edit dl rm
jobs.d.ts	67	0644	edit dl rm
jobs.d.ts.map	127	0644	edit dl rm
jobs.js	307	0644	edit dl rm
jobs.js.map	152	0644	edit dl rm
jobs.mjs	154	0644	edit dl rm
jobs.mjs.map	137	0644	edit dl rm
methods.d.mts	4147	0644	edit dl rm
methods.d.mts.map	1707	0644	edit dl rm
methods.d.ts	4144	0644	edit dl rm
methods.d.ts.map	1706	0644	edit dl rm
methods.js	355	0644	edit dl rm
methods.js.map	203	0644	edit dl rm
methods.mjs	223	0644	edit dl rm
methods.mjs.map	200	0644	edit dl rm

Edit: /opt/canhelp/node_modules/openai/resources/fine-tuning/methods.d.ts (4144B)

import { APIResource } from "../../core/resource.js";
import * as GraderModelsAPI from "../graders/grader-models.js";
export declare class Methods extends APIResource {
}
/**
 * The hyperparameters used for the DPO fine-tuning job.
 */
export interface DpoHyperparameters {
    /**
     * Number of examples in each batch. A larger batch size means that model
     * parameters are updated less frequently, but with lower variance.
     */
    batch_size?: 'auto' | number;
    /**
     * The beta value for the DPO method. A higher beta value will increase the weight
     * of the penalty between the policy and reference model.
     */
    beta?: 'auto' | number;
    /**
     * Scaling factor for the learning rate. A smaller learning rate may be useful to
     * avoid overfitting.
     */
    learning_rate_multiplier?: 'auto' | number;
    /**
     * The number of epochs to train the model for. An epoch refers to one full cycle
     * through the training dataset.
     */
    n_epochs?: 'auto' | number;
}
/**
 * Configuration for the DPO fine-tuning method.
 */
export interface DpoMethod {
    /**
     * The hyperparameters used for the DPO fine-tuning job.
     */
    hyperparameters?: DpoHyperparameters;
}
/**
 * The hyperparameters used for the reinforcement fine-tuning job.
 */
export interface ReinforcementHyperparameters {
    /**
     * Number of examples in each batch. A larger batch size means that model
     * parameters are updated less frequently, but with lower variance.
     */
    batch_size?: 'auto' | number;
    /**
     * Multiplier on amount of compute used for exploring search space during training.
     */
    compute_multiplier?: 'auto' | number;
    /**
     * The number of training steps between evaluation runs.
     */
    eval_interval?: 'auto' | number;
    /**
     * Number of evaluation samples to generate per training step.
     */
    eval_samples?: 'auto' | number;
    /**
     * Scaling factor for the learning rate. A smaller learning rate may be useful to
     * avoid overfitting.
     */
    learning_rate_multiplier?: 'auto' | number;
    /**
     * The number of epochs to train the model for. An epoch refers to one full cycle
     * through the training dataset.
     */
    n_epochs?: 'auto' | number;
    /**
     * Level of reasoning effort.
     */
    reasoning_effort?: 'default' | 'low' | 'medium' | 'high';
}
/**
 * Configuration for the reinforcement fine-tuning method.
 */
export interface ReinforcementMethod {
    /**
     * The grader used for the fine-tuning job.
     */
    grader: GraderModelsAPI.StringCheckGrader | GraderModelsAPI.TextSimilarityGrader | GraderModelsAPI.PythonGrader | GraderModelsAPI.ScoreModelGrader | GraderModelsAPI.MultiGrader;
    /**
     * The hyperparameters used for the reinforcement fine-tuning job.
     */
    hyperparameters?: ReinforcementHyperparameters;
}
/**
 * The hyperparameters used for the fine-tuning job.
 */
export interface SupervisedHyperparameters {
    /**
     * Number of examples in each batch. A larger batch size means that model
     * parameters are updated less frequently, but with lower variance.
     */
    batch_size?: 'auto' | number;
    /**
     * Scaling factor for the learning rate. A smaller learning rate may be useful to
     * avoid overfitting.
     */
    learning_rate_multiplier?: 'auto' | number;
    /**
     * The number of epochs to train the model for. An epoch refers to one full cycle
     * through the training dataset.
     */
    n_epochs?: 'auto' | number;
}
/**
 * Configuration for the supervised fine-tuning method.
 */
export interface SupervisedMethod {
    /**
     * The hyperparameters used for the fine-tuning job.
     */
    hyperparameters?: SupervisedHyperparameters;
}
export declare namespace Methods {
    export { type DpoHyperparameters as DpoHyperparameters, type DpoMethod as DpoMethod, type ReinforcementHyperparameters as ReinforcementHyperparameters, type ReinforcementMethod as ReinforcementMethod, type SupervisedHyperparameters as SupervisedHyperparameters, type SupervisedMethod as SupervisedMethod, };
}
//# sourceMappingURL=methods.d.ts.map

cmd: