Downloads · 30 days
89
0% of all-time downloads
uukuguy/speechless-mistral-dolphin-orca-platypus-samantha-7b
speechless-mistral-dolphin-orca-platypus-samantha-7b is a text generation model from uukuguy. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as llama2.
<p<h1 speechless-mistral-dolphin-orca-platypus-samantha-7b </h1</p
Downloads · 30 days
89
0% of all-time downloads
All-time downloads
38.4K
Public
Parameters
7.2B
15 GB on disk
Likes
17
Public
Click a slice to open those files.
.safetensors15 GB · 100%
How the weights are stored.
BF167B · 96%
From the Hugging Face model README
This model is a merge of ehartford/dolphin-2.1-mistral-7b, Open-Orca/Mistral-7B-OpenOrca, bhenrym14/mistral-7b-platypus-fp16 and ehartford/samantha-1.2-mistral-7b.
I'm very sorry for giving such a long and peculiar name. Originally, it was just my lazy behavior during the process of making models to easily distinguish various model and dataset combinations. I didn't expect the previous model (Thebloke GPTQ Version) to be so popular. This time, based on some guys's request, I am releasing a model based on Mistral, and I have also inherited the style of the super long name along with it. Welcome to try the model, please refrain from harsh criticism if you don't like it.
Code: https://github.com/uukuguy/speechless
| Metric | Value |
|---|---|
| humaneval-python | 34.146 |
CodeLlama-34B-Python: 53.29
CodeLlama-34B-Instruct: 50.79
CodeLlama-13B-Instruct: 50.6
CodeLlama-34B: 45.11
CodeLlama-13B-Python: 42.89
CodeLlama-13B: 35.07
Mistral-7B-v0.1: 30.488
| Metric | Value |
|---|---|
| ARC | 64.33 |
| HellaSwag | 84.4 |
| MMLU | 63.72 |
| TruthfulQA | 52.52 |
| Winogrande | 78.37 |
| GSM8K | 21.38 |
| DROP | 8.66 |
| Average | 53.34 |
The Mistral-7B-v0.1 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters. Mistral-7B-v0.1 outperforms Llama 2 13B on all benchmarks we tested.
For full details of this model please read our paper and release blog post.
Mistral-7B-v0.1 is a transformer model, with the following architecture choices:
KeyError: 'mistral'NotImplementedError: Cannot copy out of meta tensor; no data!Ensure you are utilizing a stable version of Transformers, 4.34.0 or newer.
Mistral 7B is a pretrained base model and therefore does not have any moderation mechanisms.
Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.`
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 53.34 |
| ARC (25-shot) | 64.33 |
| HellaSwag (10-shot) | 84.4 |
| MMLU (5-shot) | 63.72 |
| TruthfulQA (0-shot) | 52.52 |
| Winogrande (5-shot) | 78.37 |
| GSM8K (5-shot) | 21.38 |
| DROP (3-shot) | 8.66 |