YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
Quantization made by Richard Erkhov.
bitnet_b1_58-3B - GGUF
- Model creator: https://huggingface.co/1bitLLM/
- Original model: https://huggingface.co/1bitLLM/bitnet_b1_58-3B/
| Name | Quant method | Size |
|---|---|---|
| bitnet_b1_58-3B.Q2_K.gguf | Q2_K | 1.79GB |
| bitnet_b1_58-3B.IQ3_XS.gguf | IQ3_XS | 1.79GB |
| bitnet_b1_58-3B.IQ3_S.gguf | IQ3_S | 1.79GB |
| bitnet_b1_58-3B.Q3_K_S.gguf | Q3_K_S | 1.79GB |
| bitnet_b1_58-3B.IQ3_M.gguf | IQ3_M | 1.86GB |
| bitnet_b1_58-3B.Q3_K.gguf | Q3_K | 1.94GB |
| bitnet_b1_58-3B.Q3_K_M.gguf | Q3_K_M | 1.94GB |
| bitnet_b1_58-3B.Q3_K_L.gguf | Q3_K_L | 2.01GB |
| bitnet_b1_58-3B.IQ4_XS.gguf | IQ4_XS | 1.81GB |
| bitnet_b1_58-3B.Q4_0.gguf | Q4_0 | 1.79GB |
| bitnet_b1_58-3B.IQ4_NL.gguf | IQ4_NL | 1.81GB |
| bitnet_b1_58-3B.Q4_K_S.gguf | Q4_K_S | 2.17GB |
| bitnet_b1_58-3B.Q4_K.gguf | Q4_K | 2.34GB |
| bitnet_b1_58-3B.Q4_K_M.gguf | Q4_K_M | 2.34GB |
| bitnet_b1_58-3B.Q4_1.gguf | Q4_1 | 1.98GB |
| bitnet_b1_58-3B.Q5_0.gguf | Q5_0 | 2.17GB |
| bitnet_b1_58-3B.Q5_K_S.gguf | Q5_K_S | 2.35GB |
| bitnet_b1_58-3B.Q5_K.gguf | Q5_K | 2.5GB |
| bitnet_b1_58-3B.Q5_K_M.gguf | Q5_K_M | 2.5GB |
| bitnet_b1_58-3B.Q5_1.gguf | Q5_1 | 2.35GB |
| bitnet_b1_58-3B.Q6_K.gguf | Q6_K | 3.29GB |
| bitnet_b1_58-3B.Q8_0.gguf | Q8_0 | 3.29GB |
Original model description:
license: mit
This is a reproduction of the BitNet b1.58 paper. The models are trained with RedPajama dataset for 100B tokens. The hypers, as well as two-stage LR and weight decay, are implemented as suggested in their following paper. All models are open-source in the repo. We will train larger models and/or more tokens when resource is available.
Results
PPL and zero-shot accuracy:
| Models | PPL | ARCe | ARCc | HS | BQ | OQ | PQ | WGe | Avg |
|---|---|---|---|---|---|---|---|---|---|
| FP16 700M (reported) | 12.33 | 54.7 | 23.0 | 37.0 | 60.0 | 20.2 | 68.9 | 54.8 | 45.5 |
| BitNet b1.58 700M (reported) | 12.87 | 51.8 | 21.4 | 35.1 | 58.2 | 20.0 | 68.1 | 55.2 | 44.3 |
| BitNet b1.58 700M (reproduced) | 12.78 | 51.4 | 21.8 | 35.0 | 59.6 | 20.6 | 67.5 | 55.4 | 44.5 |
| FP16 1.3B (reported) | 11.25 | 56.9 | 23.5 | 38.5 | 59.1 | 21.6 | 70.0 | 53.9 | 46.2 |
| BitNet b1.58 1.3B (reported) | 11.29 | 54.9 | 24.2 | 37.7 | 56.7 | 19.6 | 68.8 | 55.8 | 45.4 |
| BitNet b1.58 1.3B (reproduced) | 11.19 | 55.8 | 23.7 | 37.6 | 59.0 | 20.2 | 69.2 | 56.0 | 45.9 |
| FP16 3B (reported) | 10.04 | 62.1 | 25.6 | 43.3 | 61.8 | 24.6 | 72.1 | 58.2 | 49.7 |
| BitNet b1.58 3B (reported) | 9.91 | 61.4 | 28.3 | 42.9 | 61.5 | 26.6 | 71.5 | 59.3 | 50.2 |
| BitNet b1.58 3B (reproduced) | 9.88 | 60.9 | 28.0 | 42.3 | 58.3 | 26.0 | 71.4 | 60.3 | 49.6 |
The differences between the reported numbers and the reproduced results are possibly variances from the training data processing, seeds, or other random factors.
Evaluation
The evaluation pipelines are from the paper authors. Here is the commands to run the evaluation:
pip install lm-eval==0.3.0
python eval_ppl.py --hf_path 1bitLLM/bitnet_b1_58-3B --seqlen 2048
python eval_task.py --hf_path 1bitLLM/bitnet_b1_58-3B \
--batch_size 1 \
--tasks \
--output_path result.json \
--num_fewshot 0 \
--ctx_size 2048
- Downloads last month
- 13