metadata
base_model:
- BlinkDL/rwkv-7-world
language:
- en
- zh
- ja
- ko
- fr
- ar
- es
- pt
license: apache-2.0
metrics:
- accuracy
pipeline_tag: text-generation
library_name: transformers
rwkv7-191M-world
This is RWKV-7 model under flash-linear attention format.
Model Details
Model Description
- Developed by: Bo Peng, Yu Zhang, Songlin Yang, Ruichong Zhang
- Funded by: RWKV Project (Under LF AI & Data Foundation)
- Model type: RWKV7
- Language(s) (NLP): English
- License: Apache-2.0
- Parameter count: 191M
- Tokenizer: RWKV World tokenizer
- Vocabulary size: 65,536
Model Sources
- Repository: https://github.com/fla-org/flash-linear-attention ; https://github.com/BlinkDL/RWKV-LM
- Paper: https://arxiv.org/abs/2503.14456
Uses
Install flash-linear-attention
and the latest version of transformers
before using this model:
pip install git+https://github.com/fla-org/flash-linear-attention
pip install 'transformers>=4.48.0'
Direct Use
You can use this model just as any other HuggingFace models:
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained('fla-hub/rwkv7-191M-world', trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained('fla-hub/rwkv7-191M-world', trust_remote_code=True)
model = model.cuda()
prompt = "What is a large language model?"
messages = [
{"role": "user", "content": "Who are you?"},
{"role": "assistant", "content": "I am a GPT-3 based model."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=1024,
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)[0]
print(response)
Training Details
Training Data
This model is trained on the World v2.8 with a total of 1.0 trillion tokens.
Training Hyperparameters
- Training regime: bfloat16, lr 8e-4 to 1e-5 "delayed" cosine decay, wd 0.1, bsz 8x30x4096 (Please contact Peng Bo for details)
Evaluation
Metrics
lambada_openai
: ppl 12.4 acc 49.1%
FAQ
Q: safetensors metadata is none.
A: upgrade transformers to >=4.48.0: pip install 'transformers>=4.48.0'