Update README.md
Browse files
README.md
CHANGED
|
@@ -61,4 +61,9 @@ Rather than training one model, we trained multiple models, each on a different
|
|
| 61 |
- DType: `float16`
|
| 62 |
- Torch.Compile: `false`
|
| 63 |
- DataLoader Workers: `4`
|
| 64 |
-
- Seed: `311`
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 61 |
- DType: `float16`
|
| 62 |
- Torch.Compile: `false`
|
| 63 |
- DataLoader Workers: `4`
|
| 64 |
+
- Seed: `311`
|
| 65 |
+
|
| 66 |
+
# Results
|
| 67 |
+
|
| 68 |
+
| Dataset | Tokens | Final Val Loss | ARC Easy | ARC Challenge | HellaSwag | PiQA | SWAG | BLiMP |
|
| 69 |
+
|:--------|-------:|---------------:|---------:|--------------:|----------:|-----:|-----:|------:|
|