Laguna-XS-2.1

View as Markdown

Laguna-XS-2.1 has a checked-in NeMo AutoModel recipe for text generation. The Hugging Face configuration declares the LagunaForCausalLM architecture.

Fine-Tune Laguna-XS-2.1

Follow the installation instructions, then run the recipe from the repository root:

automodel examples/llm_finetune/laguna/laguna_xs_2p1_hellaswag_ep8.yaml --nproc-per-node 8

Choose a Workflow

GoalStart Here
Run the primary recipe for this modelUse the recipe configuration.
Prepare your environmentFollow the installation instructions.

Configuration

SettingConfiguration
Hardware-
StrategyFSDP2; tp_size=1, pp_size=1, cp_size=1, ep_size=8
Nodes1
FeaturesActivation checkpointing (activation_checkpointing=true), Packed sequences (packed_sequence_size=1024)
Advancedbfloat16, attention=sdpa, linear=torch, experts=torch_mm, dispatcher=torch

Model Reference

Model Architecture

PropertyValue
TaskText generation
ArchitectureLagunaForCausalLM

Available Models