This model is a fine-tuned version of google/gemma-7b on the chansung/mentalhealthcounselingconversationsmerged dataset. It achieves the following results on the evaluation set: — Loss: 0.8529 The following hyperparameters were used during training: — learningrate: 0.0002 — trainbatchsize: 2 — evalbatchsize: 2 — seed: 42 — distributedtype: multi-GPU — numdevices: 6 — gradientaccumulationsteps: 2 — totaltrainbatchsize: 24 — totalevalbatchsize: 12 — optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 — lrschedulertype: cosine — lrschedulerwarmupratio: 0.1 — num_epochs: 5 — Transformers 4.41.
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: chansung
Теги: tensorboard, gemma, alignment-handbook, trl, sft, generated_from_trainer, conversational, text-generation-inference
Лайков: 3 | Загрузок: 30
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
