File size: 3,053 Bytes
d456d17 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
license: other
license_name: databricks-open-model-license
license_link: https://www.databricks.com/legal/open-model-license
base_model: databricks/dbrx-base
---
# dbrx-base-FP8-KV
- ## Introduction
This model was created by applying [Quark](https://quark.docs.amd.com/latest/index.html) with calibration samples from Pile dataset.
- ## Quantization Stragegy
- ***Quantized Layers***: All linear layers excluding "lm_head" and "router.layer"
- ***Weight***: FP8 symmetric per-tensor
- ***Activation***: FP8 symmetric per-tensor
- ***KV Cache***: FP8 symmetric per-tensor
- ## Quick Start
1. [Download and install Quark](https://quark.docs.amd.com/latest/install.html)
2. Run the quantization script in the example folder using the following command line:
```sh
export MODEL_DIR = [local model checkpoint folder] or databricks/dbrx-base
# single GPU
python3 quantize_quark.py \
--model_dir $MODEL_DIR \
--output_dir dbrx-base-FP8-KV \
--quant_scheme w_fp8_a_fp8 \
--kv_cache_dtype fp8 \
--num_calib_data 128 \
--model_export quark_safetensors \
--no_weight_matrix_merge \
# If model size is too large for single GPU, please use multi GPU instead.
python3 quantize_quark.py
--model_dir $MODEL_DIR \
--output_dir dbrx-base-FP8-KV\
--quant_scheme w_fp8_a_fp8 \
--kv_cache_dtype fp8 \
--num_calib_data 128 \
--multi_gpu \
--model_export quark_safetensors \
--no_weight_matrix_merge \
--multi_gpu
```
## Deployment
Quark has its own export format and allows FP8 quantized models to be efficiently deployed using the vLLM backend(vLLM-compatible).
In the dbrx-base model, "transformer.blocks.\*.ffn.experts" modules can be divided into experts-num mlps, and if the shape of the weight of w1 in one of the mlps is [dim1, dim2],
then the shape of “transformer.blocks.\*.ffn.experts.mlp.w1.weight“ in the exported safetensors file is [dim1\*experts-num, dim2]. The shapes of "transformer.blocks.\*.ffn.experts.mlp.w1.weight_scale"
and "transformer.blocks.\*.ffn.experts.mlp.w1.input_scale" are [dim1]. Similarly, this also applies to the w2 and v1 of "transformer.blocks.\*.ffn.experts.mlp".
## Evaluation
Quark currently uses perplexity(PPL) as the evaluation metric for accuracy loss before and after quantization.The specific PPL algorithm can be referenced in the quantize_quark.py.
The quantization evaluation results are conducted in pseudo-quantization mode, which may slightly differ from the actual quantized inference accuracy. These results are provided for reference only.
#### Evaluation scores
<table>
<tr>
<td><strong>Benchmark</strong>
</td>
<td><strong>dbrx-base </strong>
</td>
<td><strong>dbrx-base-FP8-KV(this model)</strong>
</td>
</tr>
<tr>
<td>Perplexity-wikitext2
</td>
<td>3.9106
</td>
<td>3.9410
</td>
</tr>
</table>
#### License
Modifications copyright(c) 2024 Advanced Micro Devices,Inc. All rights reserved.
|