measurement.json

README.md

@@ -0,0 +1,83 @@
+---
+license: llama3.1
+language:
+- en
+inference: false
+fine-tuning: false
+tags:
+- nvidia
+- llama3.1
+datasets:
+- nvidia/HelpSteer2
+base_model: meta-llama/Llama-3.1-70B-Instruct
+pipeline_tag: text-generation
+library_name: transformers
+quantized_by: bartowski
+---
+
+## Exllama v2 Quantizations of Llama-3.1-Nemotron-70B-Instruct-HF
+
+Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.2.3">turboderp's ExLlamaV2 v0.2.3</a> for quantization.
+
+<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>
+
+Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
+
+Conversion was done using the default calibration dataset.
+
+Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.
+
+Original model: https://huggingface.co/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
+
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/8_0">8.0 bits per weight</a>
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/6_5">6.5 bits per weight</a>
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/5_0">5.0 bits per weight</a>
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/4_25">4.25 bits per weight</a>
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/3_5">3.5 bits per weight</a>
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/3_0">3.0 bits per weight</a>
+
+<a href="https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2/tree/2_2">2.2 bits per weight</a>
+
+
+## Download instructions
+
+With git:
+
+```shell
+git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2
+```
+
+With huggingface hub (credit to TheBloke for instructions):
+
+```shell
+pip3 install huggingface-hub
+```
+
+To download the `main` (only useful if you only care about measurement.json) branch to a folder called `Llama-3.1-Nemotron-70B-Instruct-HF-exl2`:
+
+```shell
+mkdir Llama-3.1-Nemotron-70B-Instruct-HF-exl2
+huggingface-cli download bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2 --local-dir Llama-3.1-Nemotron-70B-Instruct-HF-exl2
+```
+
+To download from a different branch, add the `--revision` parameter:
+
+Linux:
+
+```shell
+mkdir Llama-3.1-Nemotron-70B-Instruct-HF-exl2-6_5
+huggingface-cli download bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2 --revision 6_5 --local-dir Llama-3.1-Nemotron-70B-Instruct-HF-exl2-6_5
+```
+
+Windows (which apparently doesn't like _ in folders sometimes?):
+
+```shell
+mkdir Llama-3.1-Nemotron-70B-Instruct-HF-exl2-6.5
+huggingface-cli download bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-exl2 --revision 6_5 --local-dir Llama-3.1-Nemotron-70B-Instruct-HF-exl2-6.5
+```