Upload README.md with huggingface_hub
Browse files
README.md
ADDED
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
-license: other
|
3 |
+
-license_link: https://huggingface.co/microsoft/phi-2/resolve/main/LICENSE
|
4 |
+
-license_name: microsoft-research-license
|
5 |
+
base_model: Yhyu13/LMCocktail-phi-2-v1
|
6 |
+
inference: false
|
7 |
+
model_creator: Yhyu13
|
8 |
+
model_name: LMCocktail-phi-2-v1
|
9 |
+
pipeline_tag: text-generation
|
10 |
+
quantized_by: afrideva
|
11 |
+
tags:
|
12 |
+
- gguf
|
13 |
+
- ggml
|
14 |
+
- quantized
|
15 |
+
- q2_k
|
16 |
+
- q3_k_m
|
17 |
+
- q4_k_m
|
18 |
+
- q5_k_m
|
19 |
+
- q6_k
|
20 |
+
- q8_0
|
21 |
+
---
|
22 |
+
# Yhyu13/LMCocktail-phi-2-v1-GGUF
|
23 |
+
|
24 |
+
Quantized GGUF model files for [LMCocktail-phi-2-v1](https://huggingface.co/Yhyu13/LMCocktail-phi-2-v1) from [Yhyu13](https://huggingface.co/Yhyu13)
|
25 |
+
|
26 |
+
|
27 |
+
| Name | Quant method | Size |
|
28 |
+
| ---- | ---- | ---- |
|
29 |
+
| [lmcocktail-phi-2-v1.fp16.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.fp16.gguf) | fp16 | 5.56 GB |
|
30 |
+
| [lmcocktail-phi-2-v1.q2_k.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.q2_k.gguf) | q2_k | 1.17 GB |
|
31 |
+
| [lmcocktail-phi-2-v1.q3_k_m.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.q3_k_m.gguf) | q3_k_m | 1.48 GB |
|
32 |
+
| [lmcocktail-phi-2-v1.q4_k_m.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.q4_k_m.gguf) | q4_k_m | 1.79 GB |
|
33 |
+
| [lmcocktail-phi-2-v1.q5_k_m.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.q5_k_m.gguf) | q5_k_m | 2.07 GB |
|
34 |
+
| [lmcocktail-phi-2-v1.q6_k.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.q6_k.gguf) | q6_k | 2.29 GB |
|
35 |
+
| [lmcocktail-phi-2-v1.q8_0.gguf](https://huggingface.co/afrideva/LMCocktail-phi-2-v1-GGUF/resolve/main/lmcocktail-phi-2-v1.q8_0.gguf) | q8_0 | 2.96 GB |
|
36 |
+
|
37 |
+
|
38 |
+
|
39 |
+
## Original Model Card:
|
40 |
+
# LM-cocktail phi-2 v1
|
41 |
+
|
42 |
+
|
43 |
+
This is a 50%-50% model of the phi2 alpca gpt4 and phi2 ultrachat200K models.
|
44 |
+
|
45 |
+
https://huggingface.co/Yhyu13/phi-2-sft-alpaca_gpt4_en-ep1
|
46 |
+
|
47 |
+
https://huggingface.co/venkycs/phi-2-ultrachat200k
|
48 |
+
|
49 |
+
|
50 |
+
|
51 |
+
# Code
|
52 |
+
|
53 |
+
The LM-cocktail is novel technique for merging multiple models https://arxiv.org/abs/2311.13534
|
54 |
+
|
55 |
+
Code is backed up by this repo https://github.com/FlagOpen/FlagEmbedding.git
|
56 |
+
|
57 |
+
Merging scripts available under the [./scripts](./scripts) folder
|