add rwkv-5-3b model (#666)

* support rwkv5-3b learnboard

* update rwkv-5-3b config

* update config

* refine

* fix bug

* update config

* refine

* reduce batch size

* refine

* reduce batch size to avoid oom in special datasets

* Update huggingface.py

* Update huggingface.py

configs/eval_rwkv5_3b.py

+from mmengine.config import read_base
+
+with read_base():
+    from .models.rwkv.rwkv5_3b import models
+    from .datasets.collections.base_medium_llama import datasets
+    from .summarizers.leaderboard import summarizer

configs/models/rwkv/rwkv5_3b.py

+from opencompass.models import HuggingFaceCausalLM
+
+models = [
+    dict(
+        type=HuggingFaceCausalLM,
+        abbr='rwkv-5-3b',
+        path="RWKV/rwkv-5-world-3b",
+        tokenizer_path='RWKV/rwkv-5-world-3b',
+        model_kwargs=dict(
+            device_map='auto',
+            trust_remote_code=True,
+        ),
+        tokenizer_kwargs=dict(
+            padding_side='left',
+            truncation_side='left',
+            trust_remote_code=True,
+            use_fast=False,
+        ),
+        max_out_len=100,
+        max_seq_len=2048,
+        batch_padding=True,
+        batch_size=16,
+        run_cfg=dict(num_gpus=1, num_procs=1),
+    )
+]