Datasets:

---

Den4ikAI

/

russian_dialogues_2

like 16

all samples was in string form of list
fixed:

from datasets import load_dataset
import ast
dataset = load_dataset("Den4ikAI/russian_dialogues_2")
def prepare_dataset(batch):
batch["sample"] = ast.literal_eval(batch["sample"])
return batch
dataset = dataset.map(prepare_dataset, num_proc=8)