Update README.md
Browse files
README.md
CHANGED
@@ -1,33 +1,41 @@
|
|
1 |
-
|
2 |
-
base_model: []
|
3 |
-
library_name: transformers
|
4 |
-
tags:
|
5 |
-
- mergekit
|
6 |
-
- merge
|
7 |
|
8 |
-
|
9 |
-
# 24b
|
10 |
|
11 |
-
|
12 |
|
13 |
-
|
14 |
-
### Merge Method
|
15 |
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
-
The following YAML configuration was used to produce this model:
|
26 |
|
27 |
-
|
28 |
-
base_model: MS3.2-Austral-24B-SFT+/mergekit/24B-KTO-austral/checkpoint-424
|
29 |
-
dtype: bfloat16
|
30 |
-
merge_method: passthrough
|
31 |
-
models:
|
32 |
-
- model: MS3.2-Austral-24B-SFT+/mergekit/24B-KTO-austral/checkpoint-424
|
33 |
-
```
|
|
|
1 |
+
# What is this
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
+
This the KTO checkpoint of my MS3.2 Austral winton train.
|
|
|
4 |
|
5 |
+
wandb: https://wandb.ai/new-eden/austral/runs/2iaj6moy?nw=nwuserdeltavector
|
6 |
|
7 |
+
Datasets:
|
|
|
8 |
|
9 |
+
```
|
10 |
+
datasets:
|
11 |
+
- path: Delta-Vector/Tauri-IFeval-Dans-Tulu-KTO
|
12 |
+
split: train
|
13 |
+
type: chatml.argilla
|
14 |
+
- path: Delta-Vector/Tauri-Opus-accepted-hermes-rejected-shuffled
|
15 |
+
split: train
|
16 |
+
type: chatml.argilla
|
17 |
+
- path: Delta-Vector/Tauri-Opus-Accepted-GPT-Rejected-Opus-Writing-Prompts
|
18 |
+
split: train
|
19 |
+
type: chatml.argilla
|
20 |
+
- path: Delta-Vector/Tauri-Helpsteer3-Edit
|
21 |
+
split: train
|
22 |
+
type: chatml.argilla
|
23 |
+
- path: Delta-Vector/Tauri-Helpsteer-3-Preference-KTO
|
24 |
+
split: train
|
25 |
+
type: chatml.argilla
|
26 |
+
- path: NewEden/Purpura-Arkhaios-CC-KTO
|
27 |
+
split: train
|
28 |
+
type: chatml.argilla
|
29 |
+
- path: Delta-Vector/Tauri-KTO-Instruct-Mix
|
30 |
+
split: train
|
31 |
+
type: chatml.argilla
|
32 |
+
- path: Delta-Vector/Tauri-LIT-RL-KTO
|
33 |
+
split: train
|
34 |
+
type: chatml.argilla
|
35 |
+
- path: Delta-Vector/Tauri-Synth-1-KTO-R1-No-Think
|
36 |
+
split: train
|
37 |
+
type: chatml.argilla
|
38 |
+
```
|
39 |
|
|
|
40 |
|
41 |
+
Trained on 8xA100s using Axolotl. Ty to my work & Auri <3
|
|
|
|
|
|
|
|
|
|
|
|