Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,30 @@ datasets:
|
|
| 18 |
|
| 19 |
## Model Details
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
### Model Description
|
| 22 |
|
| 23 |
<!-- Provide a longer summary of what this model is. -->
|
|
|
|
| 18 |
|
| 19 |
## Model Details
|
| 20 |
|
| 21 |
+
Training run on 240606:
|
| 22 |
+
|
| 23 |
+
Step Training Loss Validation Loss
|
| 24 |
+
25 1.506400 1.164538
|
| 25 |
+
50 1.128500 1.059316
|
| 26 |
+
75 1.095100 1.040511
|
| 27 |
+
100 1.068700 1.031033
|
| 28 |
+
125 1.061300 1.024377
|
| 29 |
+
150 1.035700 1.017490
|
| 30 |
+
175 1.061200 1.012095
|
| 31 |
+
200 1.031600 1.007867
|
| 32 |
+
225 1.031900 1.002652
|
| 33 |
+
250 0.958300 1.003817
|
| 34 |
+
275 0.967900 1.000483
|
| 35 |
+
300 0.950000 0.998807
|
| 36 |
+
325 0.974300 0.996894
|
| 37 |
+
350 0.960700 0.994098
|
| 38 |
+
375 0.956000 0.991491
|
| 39 |
+
400 0.940500 0.988697
|
| 40 |
+
425 0.949100 0.987253
|
| 41 |
+
450 0.940600 0.986425 <-- Picked checkpoint
|
| 42 |
+
475 0.888300 0.994204
|
| 43 |
+
500 0.881700 0.994897
|
| 44 |
+
|
| 45 |
### Model Description
|
| 46 |
|
| 47 |
<!-- Provide a longer summary of what this model is. -->
|