Released in February 2023, Meta’s Llama builds on the general transformer decoder framework with some key additions such as pre-normalization, SwiGLU activations, and Rotary Positional Embeddings (RoPE). More information is available in the companion paper “LLaMA: Open and Efficient Foundation Language Models”. With 7B, 13B, 33B, and 65B offerings - Llama has options for every inference budget.
- Data Preparation
- Training with Predefined Configurations
- Checkpoint Conversion
- Model Evaluation
- Parameter Efficient Fine-Tuning (PEFT)
- Model Export to TensorRT-LLM
- Model Deployment
- Llama-2 Results
Feature |
Status |
---|---|
Data parallelism | ✓ |
Tensor parallelism | ✓ |
Pipeline parallelism | ✓ |
Interleaved Pipeline Parallelism Sched | N/A |
Sequence parallelism | ✓ |
Selective activation checkpointing | ✓ |
Gradient checkpointing | ✓ |
Partial gradient checkpointing | ✓ |
FP32/TF32 | ✓ |
AMP/FP16 | ✗ |
BF16 | ✓ |
TransformerEngine/FP8 | ✗ |
Multi-GPU | ✓ |
Multi-Node | ✓ |
Inference | N/A |
Slurm | ✓ |
Base Command Manager | ✓ |
Base Command Platform | ✓ |
Distributed data preprcessing | ✓ |
NVfuser | ✗ |
P-Tuning and Prompt Tuning | ✓ |
IA3 and Adapter learning | ✓ |
Distributed Optimizer | ✓ |
Distributed Checkpoint | ✓ |
Fully Shared Data Parallel | ✓ |