Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- HuggingFaceFW/fineweb-2
|
| 5 |
+
language:
|
| 6 |
+
- is
|
| 7 |
+
pipeline_tag: text-generation
|
| 8 |
+
library_name: transformers
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
This is a model with the same specifications as SmolLM2-135M trained from scratch on the Icelandic portion of Fineweb-2. It is intended as a baseline for my research is probably rather bad for most purposes :)
|
| 12 |
+
|
| 13 |
+
Training:
|
| 14 |
+
- 1 Epoch
|
| 15 |
+
- Learning rate: 5e-4
|
| 16 |
+
- LR scheduler: Cosine
|
| 17 |
+
- Warmup ratio: 0.05
|
| 18 |
+
- Batch size: 1
|
| 19 |
+
- 4 A100 (40GB) GPUs
|
| 20 |
+
- Gradient accumulation steps: 64
|
| 21 |
+
- Effective batch size: 256
|
| 22 |
+
- Max. context length: 8192 tokens
|