SmolLM3-SWA-pre-linear-p30

Checkpoint from the len-gen-regularization length-generalization experiments. Loads with stock transformers:

from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("vapavlo/SmolLM3-SWA-pre-linear-p30")
model = AutoModelForCausalLM.from_pretrained("vapavlo/SmolLM3-SWA-pre-linear-p30")

Downloads last month: 4

Safetensors

Model size

3B params

Tensor type

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support