something-else
/

rocm-rwkv

Model card Files Files and versions

something-else commited on Mar 26, 2024

Commit

17aceae

·

verified ·

1 Parent(s): b085b1a

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -29,7 +29,7 @@ tags:
 7B rocm-rwkv pth record: I called this model Tlanuwa since I added an extra training focusing on cherokee after each run.
 9B rocm-rwkv pth record:  40 layers embd=4096 ctx= 16384 I am calling this model Quetzal. I called this model Quetzal since I added an extra training focusing on Spanish and the dataset Axolotl-Spanish-Nahuatl after each run.
-- rwkv-9Q-stp101-N8.pth: 9B rocm-rwkv model trained with Slim pajama chunk1-10 for the first epoch and an aditional training with chunk1-2 and a mix of multi-language and code after that I am using the N8 dataset. I am currendly with the N8 dataset 4.222 GTokes. This pth has a loss of 1.904.

 7B rocm-rwkv pth record: I called this model Tlanuwa since I added an extra training focusing on cherokee after each run.
 9B rocm-rwkv pth record:  40 layers embd=4096 ctx= 16384 I am calling this model Quetzal. I called this model Quetzal since I added an extra training focusing on Spanish and the dataset Axolotl-Spanish-Nahuatl after each run.
+- rwkv-9Q-stp101-N8.pth: 9B rocm-rwkv model trained with Slim pajama chunk1-10 for the first epoch and an aditional training with chunk1-2 and a mix of multi-language and code after that I am using the N8 dataset. I am currendly with the N8 dataset 4.222 GTokes. This pth has a loss of 1.904 regarding the N8 dataset.