Quantization made by Richard Erkhov.

TinyAlpaca-1.1B - GGUF

Model creator: https://huggingface.co/luckychao/
Original model: https://huggingface.co/luckychao/TinyAlpaca-1.1B/

Name	Quant method	Size
TinyAlpaca-1.1B.Q2_K.gguf	Q2_K	0.4GB
TinyAlpaca-1.1B.IQ3_XS.gguf	IQ3_XS	0.44GB
TinyAlpaca-1.1B.IQ3_S.gguf	IQ3_S	0.47GB
TinyAlpaca-1.1B.Q3_K_S.gguf	Q3_K_S	0.47GB
TinyAlpaca-1.1B.IQ3_M.gguf	IQ3_M	0.48GB
TinyAlpaca-1.1B.Q3_K.gguf	Q3_K	0.51GB
TinyAlpaca-1.1B.Q3_K_M.gguf	Q3_K_M	0.51GB
TinyAlpaca-1.1B.Q3_K_L.gguf	Q3_K_L	0.55GB
TinyAlpaca-1.1B.IQ4_XS.gguf	IQ4_XS	0.57GB
TinyAlpaca-1.1B.Q4_0.gguf	Q4_0	0.59GB
TinyAlpaca-1.1B.IQ4_NL.gguf	IQ4_NL	0.6GB
TinyAlpaca-1.1B.Q4_K_S.gguf	Q4_K_S	0.6GB
TinyAlpaca-1.1B.Q4_K.gguf	Q4_K	0.62GB
TinyAlpaca-1.1B.Q4_K_M.gguf	Q4_K_M	0.62GB
TinyAlpaca-1.1B.Q4_1.gguf	Q4_1	0.65GB
TinyAlpaca-1.1B.Q5_0.gguf	Q5_0	0.71GB
TinyAlpaca-1.1B.Q5_K_S.gguf	Q5_K_S	0.71GB
TinyAlpaca-1.1B.Q5_K.gguf	Q5_K	0.73GB
TinyAlpaca-1.1B.Q5_K_M.gguf	Q5_K_M	0.73GB
TinyAlpaca-1.1B.Q5_1.gguf	Q5_1	0.77GB
TinyAlpaca-1.1B.Q6_K.gguf	Q6_K	0.84GB
TinyAlpaca-1.1B.Q8_0.gguf	Q8_0	1.09GB

Original model description:

language: - en datasets: - tatsu-lab/alpaca

Model Card for Model ID

This model checkpoint is the TinyLlama-1.1B fine-tuned on alpaca dataset.

Model Details

Model Sources

Repository: https://github.com/jzhang38/TinyLlama
Paper: [https://arxiv.org/abs/2404.02406]

Uses

The use of this model should comply with the restrictions from TinyLlama-1.1b and Stanford Alpaca.

How to Get Started with the Model

Use the code below to get started with the model.

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("luckychao/TinyAlpaca-1.1B")
model = AutoModelForCausalLM.from_pretrained("luckychao/TinyAlpaca-1.1B")

Training Details

Training Data

We use the alpaca dataset, which is created by researchers from Stanford University.

Training Procedure

We follow the same training procedure and mostly same hyper-parameters to fine-tune the original Alpaca model on Llama. The procedure can be found in stanford_alpaca project.

Training Hyperparameters

--num_train_epochs 3 \
--per_device_train_batch_size 2 \
--per_device_eval_batch_size 2 \
--gradient_accumulation_steps 4 \
--evaluation_strategy "no" \
--save_strategy "steps" \
--save_steps 1000 \
--save_total_limit 1 \
--learning_rate 2e-5 \
--weight_decay 0. \
--warmup_ratio 0.03 \
--lr_scheduler_type "cosine" \
--logging_steps 1 \
--bf16 True \
--fsdp "full_shard auto_wrap" \
--fsdp_transformer_layer_cls_to_wrap 'LlamaDecoderLayer' \
--model_max_length 2048

Citation

The model is mostly developed for the paper below. Please cite it if you find the repository helpful.

BibTeX:

@article{hao2024exploring,
  title={Exploring Backdoor Vulnerabilities of Chat Models},
  author={Hao, Yunzhuo and Yang, Wenkai and Lin, Yankai},
  journal={arXiv preprint arXiv:2404.02406},
  year={2024}
}