amd
/

Llama2-7b-chat-awq-g128-int4-asym-bf16-onnx-ryzen-strix

Text Generation

Model card Files Files and versions Community

pooja-ganesh commited on 11 days ago

Commit

d99b822

•

1 Parent(s): 1db2bb3

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -18,9 +18,9 @@ tags:
 - ## Introduction
   This model was created using Quark Quantization, followed by OGA Model Builder, and finalized with post-processing for NPU deployment.
 - ## Quantization Strategy
-  - AWQ / Group 128 / Asymmetric / BF16 activations
 - ## Quick Start
-For quickstart, refer to AMD [RyzenAI-SW-EA](https://account.amd.com/en/member/ryzenai-sw-ea.html)
 #### Evaluation scores
 The perplexity measurement is run on the wikitext-2-raw-v1 (raw data) dataset provided by Hugging Face. Perplexity score measured for prompt length 2k is 7.153726.

 - ## Introduction
   This model was created using Quark Quantization, followed by OGA Model Builder, and finalized with post-processing for NPU deployment.
 - ## Quantization Strategy
+  - AWQ / Group 128 / Asymmetric / BF16 activations / UINT4 Weights
 - ## Quick Start
+  For quickstart, refer to npu-llm-artifacts_1.3.0.zip available in [RyzenAI-SW-EA](https://account.amd.com/en/member/ryzenai-sw-ea.html)
 #### Evaluation scores
 The perplexity measurement is run on the wikitext-2-raw-v1 (raw data) dataset provided by Hugging Face. Perplexity score measured for prompt length 2k is 7.153726.