Upload README.md

README.md

@@ -7,7 +7,7 @@ language:
 - en
 library_name: transformers
 license: apache-2.0
-model_creator: Jet Davis
+model_creator: Devin Gulliver
 model_name: OpenInstruct Mistral 7B
 model_type: mistral
 pipeline_tag: text-generation
@@ -45,13 +45,13 @@ quantized_by: TheBloke
 <!-- header end -->
 
 # OpenInstruct Mistral 7B - GPTQ
-- Model creator: [Jet Davis](https://huggingface.co/monology)
+- Model creator: [Devin Gulliver](https://huggingface.co/monology)
 - Original model: [OpenInstruct Mistral 7B](https://huggingface.co/monology/openinstruct-mistral-7b)
 
 <!-- description start -->
 # Description
 
-This repo contains GPTQ model files for [Jet Davis's OpenInstruct Mistral 7B](https://huggingface.co/monology/openinstruct-mistral-7b).
+This repo contains GPTQ model files for [Devin Gulliver's OpenInstruct Mistral 7B](https://huggingface.co/monology/openinstruct-mistral-7b).
 
 Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
 
@@ -64,7 +64,7 @@ These files were quantised using hardware kindly provided by [Massed Compute](ht
 * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/openinstruct-mistral-7B-AWQ)
 * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ)
 * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GGUF)
-* [Jet Davis's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/monology/openinstruct-mistral-7b)
+* [Devin Gulliver's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/monology/openinstruct-mistral-7b)
 <!-- repositories-available end -->
 
 <!-- prompt-template start -->
@@ -121,12 +121,12 @@ Most GPTQ files are made with AutoGPTQ. Mistral models are currently made with T
 
 | Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
 | ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
-| [main](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/main) | 4 | 128 | Yes | 0.1 | [open-instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.16 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. | 
-| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [open-instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.57 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. | 
-| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [open-instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 7.52 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. | 
-| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [open-instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 7.68 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. | 
-| [gptq-8bit-32g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-8bit-32g-actorder_True) | 8 | 32 | Yes | 0.1 | [open-instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 8.17 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. | 
-| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [open-instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.29 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
+| [main](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/main) | 4 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.16 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. | 
+| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.57 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. | 
+| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 7.52 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. | 
+| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 7.68 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. | 
+| [gptq-8bit-32g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-8bit-32g-actorder_True) | 8 | 32 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 8.17 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. | 
+| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/openinstruct-mistral-7B-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.29 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
 
 <!-- README_GPTQ.md-provided-files end -->
 
@@ -384,7 +384,7 @@ And thank you again to a16z for their generous grant.
 
 <!-- footer end -->
 
-# Original model card: Jet Davis's OpenInstruct Mistral 7B
+# Original model card: Devin Gulliver's OpenInstruct Mistral 7B
 
 
 # OpenInstruct Mistral-7B
@@ -395,7 +395,7 @@ Quantized to FP16 and released under the [Apache-2.0](https://choosealicense.com
 Compute generously provided by [Higgsfield AI](https://higgsfield.ai/model/655559e6b5777dab620095e0).  
 
 
-Prompt format: Alpaca
+## Prompt format: Alpaca
 ```
 Below is an instruction that describes a task. Write a response that appropriately completes the request.
 
@@ -405,4 +405,10 @@ Below is an instruction that describes a task. Write a response that appropriate
 ### Response:
 ```
 
+## Recommended preset:
+- temperature: 0.2
+- top_k: 50
+- top_p 0.95
+- repetition_penalty: 1.1
+
 \*as of 21 Nov 2023. "commercially-usable" includes both an open-source base model and a *non-synthetic* open-source finetune dataset. updated leaderboard results available [here](https://huggingfaceh4-open-llm-leaderboard.hf.space).