SmolLM CPT
Collection
Continued Pre-Training of SmolLM models on the Fineweb-2 portions of Scandinavian languages.
•
9 items
•
Updated
This is a SmolLM2-135M-Instruct model fine-tuned on the Icelandic portion of Fineweb-2. It is intended for my research and has not been evaluated more broadly yet.
Training:
- 1 Epoch
- Learning rate: 5e-4
- LR scheduler: Cosine
- Warmup ratio: 0.05
- Batch size: 1
- 8 A100 (80GB) GPUs
- Gradient accumulation steps: 32
- Effective batch size: 256
- Max. context length: 8192 tokens
- Downloads last month
- 8
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for jekunz/smollm-360m-cpt-fineweb-icelandic
Base model
HuggingFaceTB/SmolLM2-360M-Instruct