SmolLM CPT LoRA
Collection
22 items • Updated • 1
This model 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.
LoRA setup:
- Rank: 256
- Alpha: 512
- Target modules: ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"]]
Training:
- 1 Epoch
- Learning rate: 8e-4
- LR scheduler: Cosine
- Warmup ratio: 0.05
- Batch size: 1
- 4 A100 (40GB) GPUs
- Gradient accumulation steps: 64
- Effective batch size: 256
- Max. context length: 8192 tokens
- Downloads last month
- 1
Model tree for jekunz/smollm-135m-lora-all-fineweb-icelandic
Base model
HuggingFaceTB/SmolLM2-135M Quantized
HuggingFaceTB/SmolLM2-135M-Instruct
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("HuggingFaceTB/SmolLM2-135M-Instruct") model = PeftModel.from_pretrained(base_model, "jekunz/smollm-135m-lora-all-fineweb-icelandic")