measurement.json

README.md

@@ -0,0 +1,88 @@
+---
+language:
+- en
+library_name: transformers
+pipeline_tag: text-generation
+datasets:
+- jondurbin/airoboros-2.2
+- Open-Orca/OpenOrca
+- garage-bAInd/Open-Platypus
+- WizardLM/WizardLM_evol_instruct_V2_196k
+- TokenBender/python_eval_instruct_51k
+- codefuse-ai/Evol-Instruction-66k
+tags:
+- llama-2
+- code
+license: llama2
+model-index:
+- name: SpeechlessCoder
+  results:
+  - task:
+      type: text-generation
+    dataset:
+      type: openai_humaneval
+      name: HumanEval
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 
+      verified: false
+quantized_by: bartowski
+---
+
+## Exllama v2 Quantizations of speechless-sparsetral-mistral-16x7b-MoE
+
+Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.13">turboderp's ExLlamaV2 v0.0.13</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.
+
+Original model: https://huggingface.co/uukuguy/speechless-sparsetral-mistral-16x7b-MoE
+
+| Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
+| ----- | ---- | ------- | ------ | ------ | ------ | ------------ |
+| [8_0](https://huggingface.co/bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2/tree/8_0) | 8.0 | 8.0 | 8.3 GB | 9.7 GB | 11.8 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
+| [6_5](https://huggingface.co/bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2/tree/6_5) | 6.5 | 8.0 | 7.1 GB | 8.5 GB | 10.6 GB | Very similar to 8.0, good tradeoff of size vs performance, **recommended**. |
+| [5_0](https://huggingface.co/bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2/tree/5_0) | 5.0 | 6.0 | 5.7 GB | 7.1 GB | 9.2 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. |
+| [4_25](https://huggingface.co/bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2/tree/4_25) | 4.25 | 6.0 | 5.1 GB | 6.5 GB | 8.6 GB | GPTQ equivalent bits per weight, slightly higher quality. |
+| [3_5](https://huggingface.co/bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2/tree/3_5) | 3.5 | 6.0 | 4.4 GB | 5.8 GB | 7.9 GB | Lower quality, only use if you have to. |
+
+## Download instructions
+
+With git:
+
+```shell
+git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2 speechless-sparsetral-mistral-16x7b-MoE-exl2-6_5
+```
+
+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 `speechless-sparsetral-mistral-16x7b-MoE-exl2`:
+
+```shell
+mkdir speechless-sparsetral-mistral-16x7b-MoE-exl2
+huggingface-cli download bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2 --local-dir speechless-sparsetral-mistral-16x7b-MoE-exl2 --local-dir-use-symlinks False
+```
+
+To download from a different branch, add the `--revision` parameter:
+
+Linux:
+
+```shell
+mkdir speechless-sparsetral-mistral-16x7b-MoE-exl2-6_5
+huggingface-cli download bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2 --revision 6_5 --local-dir speechless-sparsetral-mistral-16x7b-MoE-exl2-6_5 --local-dir-use-symlinks False
+```
+
+Windows (which apparently doesn't like _ in folders sometimes?):
+
+```shell
+mkdir speechless-sparsetral-mistral-16x7b-MoE-exl2-6.5
+huggingface-cli download bartowski/speechless-sparsetral-mistral-16x7b-MoE-exl2 --revision 6_5 --local-dir speechless-sparsetral-mistral-16x7b-MoE-exl2-6.5 --local-dir-use-symlinks False
+```
+
+Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski