Update README.md
Browse files
README.md
CHANGED
@@ -75,8 +75,8 @@ curl http://127.0.0.1:9000/loadmodel -X POST -H "Content-Type: application/json"
|
|
75 |
- Training Duration: 4 days(Stage1,2)
|
76 |
- Stage1 180MToken (LR1e-4)
|
77 |
- Stage2 160MToken (Temperature 1.0KD LR5e-6)
|
78 |
-
- Stage2.5
|
79 |
-
- Stage3
|
80 |
|
81 |
## Acknowledgements
|
82 |
|
@@ -95,7 +95,7 @@ This work was made possible through the contributions of:
|
|
95 |
|
96 |
## Limitations
|
97 |
|
98 |
-
This is trained
|
99 |
This model is currently in a testing phase and does not guarantee any specific level of performance. Users should consider it experimental technology.
|
100 |
|
101 |
## MyStories(Generated by PRWKV)
|
|
|
75 |
- Training Duration: 4 days(Stage1,2)
|
76 |
- Stage1 180MToken (LR1e-4)
|
77 |
- Stage2 160MToken (Temperature 1.0KD LR5e-6)
|
78 |
+
- Stage2.5 120MToken (Temperature 2.0KD LR3e-5)
|
79 |
+
- Stage3 100MToken
|
80 |
|
81 |
## Acknowledgements
|
82 |
|
|
|
95 |
|
96 |
## Limitations
|
97 |
|
98 |
+
This is trained Stage3 early epoch.
|
99 |
This model is currently in a testing phase and does not guarantee any specific level of performance. Users should consider it experimental technology.
|
100 |
|
101 |
## MyStories(Generated by PRWKV)
|