Update README.md
Browse files
README.md
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
model-index:
|
| 5 |
-
- name:
|
| 6 |
results: []
|
| 7 |
---
|
| 8 |
|
|
@@ -10,21 +10,63 @@ model-index:
|
|
| 10 |
should probably proofread and complete it, then remove this comment. -->
|
| 11 |
|
| 12 |
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
| 13 |
-
# lora
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
|
| 15 |
-
This model was trained from scratch on the None dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
- Loss: 1.1214
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
| 21 |
-
|
| 22 |
|
| 23 |
-
|
| 24 |
|
| 25 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
|
| 27 |
-
|
| 28 |
|
| 29 |
More information needed
|
| 30 |
|
|
|
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
model-index:
|
| 5 |
+
- name: Leyley-13B-LoRA
|
| 6 |
results: []
|
| 7 |
---
|
| 8 |
|
|
|
|
| 10 |
should probably proofread and complete it, then remove this comment. -->
|
| 11 |
|
| 12 |
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
| 13 |
+
# Leyley-13B-lora
|
| 14 |
+
Train required to find an usable one (Oh brother...):
|
| 15 |
+
|
| 16 |
+
https://wandb.ai/undis95/leyleytest?workspace=user-undis95
|
| 17 |
+
|
| 18 |
+
https://wandb.ai/undis95/leyleytest2?workspace=user-undis95
|
| 19 |
+
|
| 20 |
+
https://wandb.ai/undis95/leyleytest2-noro?workspace=user-undis95
|
| 21 |
+
|
| 22 |
+
https://wandb.ai/undis95/leyleytest3-noro?workspace=user-undis95
|
| 23 |
+
|
| 24 |
+
https://wandb.ai/undis95/leyleytest4-noro?workspace=user-undis95
|
| 25 |
+
|
| 26 |
+
This LoRA was trained on [Noromaid](https://huggingface.co/NeverSleep/Noromaid-13b-v0.1.1) from scratch using a [custom dataset](https://github.com/Undi95/somethingdata) of the game "The Coffin of Andy and Leyley".
|
| 27 |
|
|
|
|
| 28 |
It achieves the following results on the evaluation set:
|
| 29 |
- Loss: 1.1214
|
| 30 |
|
| 31 |
## Model description
|
| 32 |
|
| 33 |
+
LoRA of Andrew and Ashley from the game.
|
| 34 |
|
| 35 |
+
Only conversation between them is in the dataset, the AI reply in the name of Ashley.
|
| 36 |
|
| 37 |
+
It was trained in a way that you speak as her brother, but it can be changed with lower weight, custom system prompt or custom card.
|
| 38 |
+
|
| 39 |
+
## Prompt template
|
| 40 |
+
```
|
| 41 |
+
### Instruction:
|
| 42 |
+
You are Ashley Graves, sociopathic, brother-obsessed sister of Andrew Graves. In the following chat, you will talk with Andrew. Andrew called you Leyley as a child, and you called him Andy. Andrew does not like being called Andy.
|
| 43 |
+
|
| 44 |
+
Andrew: {prompt}
|
| 45 |
+
|
| 46 |
+
### Response:
|
| 47 |
+
Ashley:
|
| 48 |
+
|
| 49 |
+
### Input:
|
| 50 |
+
Andrew: {input}
|
| 51 |
+
```
|
| 52 |
+
|
| 53 |
+
Or
|
| 54 |
+
|
| 55 |
+
```
|
| 56 |
+
### Instruction:
|
| 57 |
+
You are Ashley Graves. In the following chat, you will talk with {{char}}.
|
| 58 |
+
|
| 59 |
+
{prompt}
|
| 60 |
+
|
| 61 |
+
### Response:
|
| 62 |
+
|
| 63 |
+
|
| 64 |
+
### Input:
|
| 65 |
+
{input}
|
| 66 |
+
```
|
| 67 |
+
## Recommanded settings
|
| 68 |
|
| 69 |
+
|
| 70 |
|
| 71 |
More information needed
|
| 72 |
|