Instructions to use Undi95/Dawn-v2-70B-2.55bpw-h6-exl2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Undi95/Dawn-v2-70B-2.55bpw-h6-exl2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Undi95/Dawn-v2-70B-2.55bpw-h6-exl2")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Undi95/Dawn-v2-70B-2.55bpw-h6-exl2") model = AutoModelForCausalLM.from_pretrained("Undi95/Dawn-v2-70B-2.55bpw-h6-exl2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Undi95/Dawn-v2-70B-2.55bpw-h6-exl2 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Undi95/Dawn-v2-70B-2.55bpw-h6-exl2" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Undi95/Dawn-v2-70B-2.55bpw-h6-exl2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Undi95/Dawn-v2-70B-2.55bpw-h6-exl2
- SGLang
How to use Undi95/Dawn-v2-70B-2.55bpw-h6-exl2 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Undi95/Dawn-v2-70B-2.55bpw-h6-exl2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Undi95/Dawn-v2-70B-2.55bpw-h6-exl2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Undi95/Dawn-v2-70B-2.55bpw-h6-exl2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Undi95/Dawn-v2-70B-2.55bpw-h6-exl2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Undi95/Dawn-v2-70B-2.55bpw-h6-exl2 with Docker Model Runner:
docker model run hf.co/Undi95/Dawn-v2-70B-2.55bpw-h6-exl2
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,417 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
tags:
|
| 4 |
+
- not-for-all-audiences
|
| 5 |
+
- nsfw
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
<center>[<a href="https://huggingface.co/Undi95/Dawn-v2-70B">fp16</a> - <a href="https://huggingface.co/Undi95/Dawn-v2-70B-GGUF">gguf</a> - exl2 : <a href="https://huggingface.co/Undi95/Dawn-v2-70B-2.55bpw-h6-exl2">2.55bpw</a>]</center>
|
| 9 |
+
</br>
|
| 10 |
+
<div style="width: 100%;">
|
| 11 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/63ab1241ad514ca8d1430003/Cxcfqi4WdtXCNLnaIqSRB.png" style="width: 75%; min-width: 200px; display: block; margin: auto;">
|
| 12 |
+
</div>
|
| 13 |
+
|
| 14 |
+
<!-- description start -->
|
| 15 |
+
## Description
|
| 16 |
+
|
| 17 |
+
This repo contains exl2 (2.55bpw, h6) files of Dawn-70B, a merge I have done with the new [layer shuffle](https://github.com/cg123/mergekit/blob/main/mergekit/scripts/layershuffle.py) method from mergekit.
|
| 18 |
+
|
| 19 |
+
[UtopiaXL](https://huggingface.co/Undi95/UtopiaXL-13B) was a huge success for me, I really liked it, so I took the same path to do this 70B: A good base, some psychologic data, some medical data, a little bit of this, of that, and LimaRP at the end as always.
|
| 20 |
+
|
| 21 |
+
NOTE: This repo contain the file [measurement.json](https://huggingface.co/Undi95/Dawn-v2-70B-2.4bpw-h6-exl2/blob/main/measurement.json) needed to do your own exl2 quant (I use [wikitext](https://huggingface.co/datasets/wikitext/resolve/refs%2Fconvert%2Fparquet/wikitext-2-raw-v1/train/0000.parquet)).
|
| 22 |
+
<!-- description end -->
|
| 23 |
+
<!-- description start -->
|
| 24 |
+
## Models and loras used
|
| 25 |
+
|
| 26 |
+
- [Sao10K/Euryale-1.3-L2-70B](https://huggingface.co/Sao10K/Euryale-1.3-L2-70B)
|
| 27 |
+
- [Xwin-LM/Xwin-LM-70B-V0.1](https://huggingface.co/Xwin-LM/Xwin-LM-70B-V0.1)
|
| 28 |
+
- [ehartford/Samantha-1.11-70b](https://huggingface.co/ehartford/Samantha-1.11-70b)
|
| 29 |
+
- [NousResearch/Nous-Hermes-Llama2-70b](https://huggingface.co/NousResearch/Nous-Hermes-Llama2-70b)
|
| 30 |
+
- [augtoma/qCammel-70-x](https://huggingface.co/augtoma/qCammel-70-x)
|
| 31 |
+
- [jondurbin/airoboros-l2-c70b-3.1.2](https://huggingface.co/jondurbin/airoboros-l2-c70b-3.1.2)
|
| 32 |
+
- [fangloveskari/ORCA_LLaMA_70B_QLoRA](https://huggingface.co/fangloveskari/ORCA_LLaMA_70B_QLoRA)
|
| 33 |
+
- [Doctor-Shotgun/limarpv3-llama2-70b-qlora](https://huggingface.co/Doctor-Shotgun/limarpv3-llama2-70b-qlora)
|
| 34 |
+
|
| 35 |
+
<!-- description end -->
|
| 36 |
+
## The sauce
|
| 37 |
+
```
|
| 38 |
+
!mergekit-layershuffle ./Dawn-v2-70B \
|
| 39 |
+
--model Sao10K/Euryale-1.3-L2-70B --weight 0.3 \
|
| 40 |
+
--model Xwin-LM/Xwin-LM-70B-V0.1 --weight 0.2 \
|
| 41 |
+
--model ehartford/Samantha-1.11-70b --weight 0.1 \
|
| 42 |
+
--model NousResearch/Nous-Hermes-Llama2-70b --weight 0.05 \
|
| 43 |
+
--model augtoma/qCammel-70-x --weight 0.05 \
|
| 44 |
+
--model jondurbin/airoboros-l2-c70b-3.1.2 --weight 0.2 \
|
| 45 |
+
--model fangloveskari/ORCA_LLaMA_70B_QLoRA --weight 0.1 \
|
| 46 |
+
--write-yaml Dawn-v2-70B.yaml
|
| 47 |
+
|
| 48 |
+
=========================
|
| 49 |
+
|
| 50 |
+
merge_method: passthrough
|
| 51 |
+
slices:
|
| 52 |
+
- sources:
|
| 53 |
+
- layer_range:
|
| 54 |
+
- 0
|
| 55 |
+
- 1
|
| 56 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 57 |
+
- sources:
|
| 58 |
+
- layer_range:
|
| 59 |
+
- 1
|
| 60 |
+
- 2
|
| 61 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 62 |
+
- sources:
|
| 63 |
+
- layer_range:
|
| 64 |
+
- 2
|
| 65 |
+
- 3
|
| 66 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 67 |
+
- sources:
|
| 68 |
+
- layer_range:
|
| 69 |
+
- 3
|
| 70 |
+
- 4
|
| 71 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 72 |
+
- sources:
|
| 73 |
+
- layer_range:
|
| 74 |
+
- 4
|
| 75 |
+
- 5
|
| 76 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 77 |
+
- sources:
|
| 78 |
+
- layer_range:
|
| 79 |
+
- 5
|
| 80 |
+
- 6
|
| 81 |
+
model: ehartford/Samantha-1.11-70b
|
| 82 |
+
- sources:
|
| 83 |
+
- layer_range:
|
| 84 |
+
- 6
|
| 85 |
+
- 8
|
| 86 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 87 |
+
- sources:
|
| 88 |
+
- layer_range:
|
| 89 |
+
- 8
|
| 90 |
+
- 9
|
| 91 |
+
model: ehartford/Samantha-1.11-70b
|
| 92 |
+
- sources:
|
| 93 |
+
- layer_range:
|
| 94 |
+
- 9
|
| 95 |
+
- 10
|
| 96 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 97 |
+
- sources:
|
| 98 |
+
- layer_range:
|
| 99 |
+
- 10
|
| 100 |
+
- 11
|
| 101 |
+
model: ehartford/Samantha-1.11-70b
|
| 102 |
+
- sources:
|
| 103 |
+
- layer_range:
|
| 104 |
+
- 11
|
| 105 |
+
- 12
|
| 106 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 107 |
+
- sources:
|
| 108 |
+
- layer_range:
|
| 109 |
+
- 12
|
| 110 |
+
- 13
|
| 111 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 112 |
+
- sources:
|
| 113 |
+
- layer_range:
|
| 114 |
+
- 13
|
| 115 |
+
- 14
|
| 116 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 117 |
+
- sources:
|
| 118 |
+
- layer_range:
|
| 119 |
+
- 14
|
| 120 |
+
- 15
|
| 121 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 122 |
+
- sources:
|
| 123 |
+
- layer_range:
|
| 124 |
+
- 15
|
| 125 |
+
- 16
|
| 126 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 127 |
+
- sources:
|
| 128 |
+
- layer_range:
|
| 129 |
+
- 16
|
| 130 |
+
- 17
|
| 131 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 132 |
+
- sources:
|
| 133 |
+
- layer_range:
|
| 134 |
+
- 17
|
| 135 |
+
- 18
|
| 136 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 137 |
+
- sources:
|
| 138 |
+
- layer_range:
|
| 139 |
+
- 18
|
| 140 |
+
- 19
|
| 141 |
+
model: NousResearch/Nous-Hermes-Llama2-70b
|
| 142 |
+
- sources:
|
| 143 |
+
- layer_range:
|
| 144 |
+
- 19
|
| 145 |
+
- 20
|
| 146 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 147 |
+
- sources:
|
| 148 |
+
- layer_range:
|
| 149 |
+
- 20
|
| 150 |
+
- 21
|
| 151 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 152 |
+
- sources:
|
| 153 |
+
- layer_range:
|
| 154 |
+
- 21
|
| 155 |
+
- 22
|
| 156 |
+
model: ehartford/Samantha-1.11-70b
|
| 157 |
+
- sources:
|
| 158 |
+
- layer_range:
|
| 159 |
+
- 22
|
| 160 |
+
- 23
|
| 161 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 162 |
+
- sources:
|
| 163 |
+
- layer_range:
|
| 164 |
+
- 23
|
| 165 |
+
- 24
|
| 166 |
+
model: augtoma/qCammel-70-x
|
| 167 |
+
- sources:
|
| 168 |
+
- layer_range:
|
| 169 |
+
- 24
|
| 170 |
+
- 25
|
| 171 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 172 |
+
- sources:
|
| 173 |
+
- layer_range:
|
| 174 |
+
- 25
|
| 175 |
+
- 27
|
| 176 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 177 |
+
- sources:
|
| 178 |
+
- layer_range:
|
| 179 |
+
- 27
|
| 180 |
+
- 28
|
| 181 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 182 |
+
- sources:
|
| 183 |
+
- layer_range:
|
| 184 |
+
- 28
|
| 185 |
+
- 29
|
| 186 |
+
model: ehartford/Samantha-1.11-70b
|
| 187 |
+
- sources:
|
| 188 |
+
- layer_range:
|
| 189 |
+
- 29
|
| 190 |
+
- 30
|
| 191 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 192 |
+
- sources:
|
| 193 |
+
- layer_range:
|
| 194 |
+
- 30
|
| 195 |
+
- 32
|
| 196 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 197 |
+
- sources:
|
| 198 |
+
- layer_range:
|
| 199 |
+
- 32
|
| 200 |
+
- 33
|
| 201 |
+
model: ehartford/Samantha-1.11-70b
|
| 202 |
+
- sources:
|
| 203 |
+
- layer_range:
|
| 204 |
+
- 33
|
| 205 |
+
- 34
|
| 206 |
+
model: augtoma/qCammel-70-x
|
| 207 |
+
- sources:
|
| 208 |
+
- layer_range:
|
| 209 |
+
- 34
|
| 210 |
+
- 35
|
| 211 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 212 |
+
- sources:
|
| 213 |
+
- layer_range:
|
| 214 |
+
- 35
|
| 215 |
+
- 37
|
| 216 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 217 |
+
- sources:
|
| 218 |
+
- layer_range:
|
| 219 |
+
- 37
|
| 220 |
+
- 38
|
| 221 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 222 |
+
- sources:
|
| 223 |
+
- layer_range:
|
| 224 |
+
- 38
|
| 225 |
+
- 39
|
| 226 |
+
model: ehartford/Samantha-1.11-70b
|
| 227 |
+
- sources:
|
| 228 |
+
- layer_range:
|
| 229 |
+
- 39
|
| 230 |
+
- 40
|
| 231 |
+
model: augtoma/qCammel-70-x
|
| 232 |
+
- sources:
|
| 233 |
+
- layer_range:
|
| 234 |
+
- 40
|
| 235 |
+
- 41
|
| 236 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 237 |
+
- sources:
|
| 238 |
+
- layer_range:
|
| 239 |
+
- 41
|
| 240 |
+
- 42
|
| 241 |
+
model: ehartford/Samantha-1.11-70b
|
| 242 |
+
- sources:
|
| 243 |
+
- layer_range:
|
| 244 |
+
- 42
|
| 245 |
+
- 43
|
| 246 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 247 |
+
- sources:
|
| 248 |
+
- layer_range:
|
| 249 |
+
- 43
|
| 250 |
+
- 44
|
| 251 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 252 |
+
- sources:
|
| 253 |
+
- layer_range:
|
| 254 |
+
- 44
|
| 255 |
+
- 45
|
| 256 |
+
model: NousResearch/Nous-Hermes-Llama2-70b
|
| 257 |
+
- sources:
|
| 258 |
+
- layer_range:
|
| 259 |
+
- 45
|
| 260 |
+
- 46
|
| 261 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 262 |
+
- sources:
|
| 263 |
+
- layer_range:
|
| 264 |
+
- 46
|
| 265 |
+
- 48
|
| 266 |
+
model: ehartford/Samantha-1.11-70b
|
| 267 |
+
- sources:
|
| 268 |
+
- layer_range:
|
| 269 |
+
- 48
|
| 270 |
+
- 49
|
| 271 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 272 |
+
- sources:
|
| 273 |
+
- layer_range:
|
| 274 |
+
- 49
|
| 275 |
+
- 50
|
| 276 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 277 |
+
- sources:
|
| 278 |
+
- layer_range:
|
| 279 |
+
- 50
|
| 280 |
+
- 51
|
| 281 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 282 |
+
- sources:
|
| 283 |
+
- layer_range:
|
| 284 |
+
- 51
|
| 285 |
+
- 54
|
| 286 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 287 |
+
- sources:
|
| 288 |
+
- layer_range:
|
| 289 |
+
- 54
|
| 290 |
+
- 55
|
| 291 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 292 |
+
- sources:
|
| 293 |
+
- layer_range:
|
| 294 |
+
- 55
|
| 295 |
+
- 56
|
| 296 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 297 |
+
- sources:
|
| 298 |
+
- layer_range:
|
| 299 |
+
- 56
|
| 300 |
+
- 58
|
| 301 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 302 |
+
- sources:
|
| 303 |
+
- layer_range:
|
| 304 |
+
- 58
|
| 305 |
+
- 59
|
| 306 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 307 |
+
- sources:
|
| 308 |
+
- layer_range:
|
| 309 |
+
- 59
|
| 310 |
+
- 60
|
| 311 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 312 |
+
- sources:
|
| 313 |
+
- layer_range:
|
| 314 |
+
- 60
|
| 315 |
+
- 62
|
| 316 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 317 |
+
- sources:
|
| 318 |
+
- layer_range:
|
| 319 |
+
- 62
|
| 320 |
+
- 63
|
| 321 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 322 |
+
- sources:
|
| 323 |
+
- layer_range:
|
| 324 |
+
- 63
|
| 325 |
+
- 64
|
| 326 |
+
model: fangloveskari/ORCA_LLaMA_70B_QLoRA
|
| 327 |
+
- sources:
|
| 328 |
+
- layer_range:
|
| 329 |
+
- 64
|
| 330 |
+
- 65
|
| 331 |
+
model: NousResearch/Nous-Hermes-Llama2-70b
|
| 332 |
+
- sources:
|
| 333 |
+
- layer_range:
|
| 334 |
+
- 65
|
| 335 |
+
- 66
|
| 336 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 337 |
+
- sources:
|
| 338 |
+
- layer_range:
|
| 339 |
+
- 66
|
| 340 |
+
- 67
|
| 341 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 342 |
+
- sources:
|
| 343 |
+
- layer_range:
|
| 344 |
+
- 67
|
| 345 |
+
- 68
|
| 346 |
+
model: augtoma/qCammel-70-x
|
| 347 |
+
- sources:
|
| 348 |
+
- layer_range:
|
| 349 |
+
- 68
|
| 350 |
+
- 70
|
| 351 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 352 |
+
- sources:
|
| 353 |
+
- layer_range:
|
| 354 |
+
- 70
|
| 355 |
+
- 71
|
| 356 |
+
model: augtoma/qCammel-70-x
|
| 357 |
+
- sources:
|
| 358 |
+
- layer_range:
|
| 359 |
+
- 71
|
| 360 |
+
- 72
|
| 361 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 362 |
+
- sources:
|
| 363 |
+
- layer_range:
|
| 364 |
+
- 72
|
| 365 |
+
- 73
|
| 366 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 367 |
+
- sources:
|
| 368 |
+
- layer_range:
|
| 369 |
+
- 73
|
| 370 |
+
- 75
|
| 371 |
+
model: jondurbin/airoboros-l2-c70b-3.1.2
|
| 372 |
+
- sources:
|
| 373 |
+
- layer_range:
|
| 374 |
+
- 75
|
| 375 |
+
- 76
|
| 376 |
+
model: Sao10K/Euryale-1.3-L2-70B
|
| 377 |
+
- sources:
|
| 378 |
+
- layer_range:
|
| 379 |
+
- 76
|
| 380 |
+
- 77
|
| 381 |
+
model: augtoma/qCammel-70-x
|
| 382 |
+
- sources:
|
| 383 |
+
- layer_range:
|
| 384 |
+
- 77
|
| 385 |
+
- 78
|
| 386 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 387 |
+
- sources:
|
| 388 |
+
- layer_range:
|
| 389 |
+
- 78
|
| 390 |
+
- 79
|
| 391 |
+
model: NousResearch/Nous-Hermes-Llama2-70b
|
| 392 |
+
- sources:
|
| 393 |
+
- layer_range:
|
| 394 |
+
- 79
|
| 395 |
+
- 80
|
| 396 |
+
model: Xwin-LM/Xwin-LM-70B-V0.1
|
| 397 |
+
|
| 398 |
+
|
| 399 |
+
=========================
|
| 400 |
+
|
| 401 |
+
=> Applying Doctor-Shotgun/limarpv3-llama2-70b-qlora x 0.35
|
| 402 |
+
```
|
| 403 |
+
<!-- prompt-template start -->
|
| 404 |
+
## Prompt template: Alpaca
|
| 405 |
+
|
| 406 |
+
```
|
| 407 |
+
Below is an instruction that describes a task. Write a response that appropriately completes the request.
|
| 408 |
+
|
| 409 |
+
### Instruction:
|
| 410 |
+
{prompt}
|
| 411 |
+
|
| 412 |
+
### Response:
|
| 413 |
+
|
| 414 |
+
```
|
| 415 |
+
A big thanks to [Charles](https://huggingface.co/chargoddard) for adding the layer shuffle method to his tool [mergekit](https://github.com/cg123/mergekit/tree/main) and [Henky/KoboldAI](https://koboldai.org/) for the machine he let me use.
|
| 416 |
+
|
| 417 |
+
If you want to support me, you can [here](https://ko-fi.com/undiai).
|