Update README.md
Browse files
README.md
CHANGED
|
@@ -5,12 +5,12 @@ base_model:
|
|
| 5 |
|
| 6 |
Fine tuned with my relatively small (about 3000 samples) sample of roleplay conversations. this datasets RP conversations have 3-15+ turns. this is my $60 attempt to force qwen3 30b to be able to handle RP stuff.
|
| 7 |
|
| 8 |
-
3 epochs, 4e-4 learning rate, because screw you qwen3
|
| 9 |
more info after I can test it.
|
| 10 |
Q4_K_S gguf incoming soon.
|
| 11 |
|
| 12 |
-
Reasoning -
|
| 13 |
|
|
|
|
| 14 |
here is a cool ST setting that as processed good results.
|
| 15 |
|
| 16 |
|
|
|
|
| 5 |
|
| 6 |
Fine tuned with my relatively small (about 3000 samples) sample of roleplay conversations. this datasets RP conversations have 3-15+ turns. this is my $60 attempt to force qwen3 30b to be able to handle RP stuff.
|
| 7 |
|
| 8 |
+
3 epochs, 4e-4 learning rate, because screw you qwen3...
|
| 9 |
more info after I can test it.
|
| 10 |
Q4_K_S gguf incoming soon.
|
| 11 |
|
|
|
|
| 12 |
|
| 13 |
+
Reasoning -
|
| 14 |
here is a cool ST setting that as processed good results.
|
| 15 |
|
| 16 |
|