Lambent commited on
Commit
8390a6a
·
verified ·
1 Parent(s): f0b5c39

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +22 -1
README.md CHANGED
@@ -1,5 +1,5 @@
1
  ---
2
- license: apache-2.0
3
  base_model: HuggingFaceTB/cosmo-1b
4
  tags:
5
  - generated_from_trainer
@@ -11,6 +11,27 @@ model-index:
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
  should probably proofread and complete it, then remove this comment. -->
13
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  [<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)
15
  <details><summary>See axolotl config</summary>
16
 
 
1
  ---
2
+ license: cc
3
  base_model: HuggingFaceTB/cosmo-1b
4
  tags:
5
  - generated_from_trainer
 
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
  should probably proofread and complete it, then remove this comment. -->
13
 
14
+ Trying out some LISA training.
15
+ This one used the same learning rate as the LORA training and only 4 layers each 10 steps.
16
+ Honestly these numbers are probably noise, with how close they are.
17
+
18
+ | Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
19
+ |-------------------------------------------------------------------------------|------:|------:|---------:|-------:|------:|
20
+ |[CosmoAlpacaLisa-0.2-1b](https://huggingface.co/Lambent/CosmoAlpacaLisa-0.2-1b)| 23.81| 51.75| 39.31| 29.04| 35.98|
21
+
22
+ | Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
23
+ |-----------------------------------------------------------------------|------:|------:|---------:|-------:|------:|
24
+ |[CosmoAlpacaLisa-1b](https://huggingface.co/Lambent/CosmoAlpacaLisa-1b)| 23.89| 51.93| 39.93| 28.68| 36.11|
25
+
26
+ | Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
27
+ |-------------------------------------------------------------------------|------:|------:|---------:|-------:|------:|
28
+ |[CosmoAlpacaLight-1b](https://huggingface.co/Lambent/CosmoAlpacaLight-1b)| 24.28| 51.31| 40.33| 29.47| 36.35|
29
+
30
+ | Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
31
+ |---------------------------------------------------------|------:|------:|---------:|-------:|------:|
32
+ |[cosmo-1b](https://huggingface.co/HuggingFaceTB/cosmo-1b)| 22.97| 52.01| 38.02| 28.73| 35.43|
33
+
34
+
35
  [<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)
36
  <details><summary>See axolotl config</summary>
37