Fischerboot commited on
Commit
40e2b15
·
verified ·
1 Parent(s): 47ef1ed

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -40
README.md CHANGED
@@ -3,43 +3,4 @@ license: apache-2.0
3
  language:
4
  - en
5
  ---
6
- <div align="center">
7
-
8
- # KobbleTinyV2-1.1B
9
- </div>
10
-
11
- This is a finetune of https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T trained on a small 50mb subset of the Kobble Dataset.
12
- Training was done in under 2 hours on a single Nvidia RTX 2060 Mobile GPU with qLora (LR 1.5e-4, rank 8, alpha 16, batch size 2, gradient acc. 4, 2048 ctx).
13
-
14
- You can obtain the GGUF quantization of this model here: https://huggingface.co/concedo/KobbleTinyV2-1.1B-GGUF
15
-
16
- Update: KobbleTiny has been upgraded to V2! The old V1 is [still available at this link](https://huggingface.co/concedo/KobbleTiny/tree/eb0c96864bfecfd6ac9ece1a42c4654b4997eb72).
17
-
18
- <video width="320" controls autoplay src="https://cdn-uploads.huggingface.co/production/uploads/63cd4b6d1c8a5d1d7d76a778/zjHfohCnEu2Y9CWSWgf0n.mp4"></video>
19
-
20
- Try it live now: https://concedo-koboldcpp-kobbletiny.hf.space/
21
-
22
- ## Dataset and Objectives
23
-
24
- The Kobble Dataset is a semi-private aggregated dataset made from multiple online sources and web scrapes.
25
- It contains content chosen and formatted specifically to work with KoboldAI software and Kobold Lite.
26
-
27
- #### Dataset Categories:
28
- - Instruct: Single turn instruct examples presented in the Alpaca format, with an emphasis on uncensored and unrestricted responses.
29
- - Chat: Two participant roleplay conversation logs in a multi-turn raw chat format that KoboldAI uses.
30
- - Story: Unstructured fiction excerpts, including literature containing various erotic and provocative content.
31
-
32
- <!-- prompt-template start -->
33
- ## Prompt template: Alpaca
34
-
35
- ```
36
- ### Instruction:
37
- {prompt}
38
-
39
- ### Response:
40
- ```
41
-
42
- <!-- prompt-template end -->
43
-
44
- **Note:** *No assurances will be provided about the **origins, safety, or copyright status** of this model, or of **any content** within the Kobble dataset.*
45
- *If you belong to a country or organization that has strict AI laws or restrictions against unlabelled or unrestricted content, you are advised not to use this model.*
 
3
  language:
4
  - en
5
  ---
6
+ \