Downloads · 30 days
0
Llaza-3.2-1B-MS2F-4K-v0.1
epfl-dlab/Llaza-3.2-1B-MS2F-4K-v0.1
Llaza-3.2-1B-MS2F-4K-v0.1 is a machine learning model from epfl-dlab. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for zip2zip.
Training checkpoint from zip2zip-core. This is a candidate model (not production-ready).
Access
Public
Updated May 3, 2026
Parameters
1.5B
15.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt12.4 GB · 80%
At a glance
- Library
- zip2zip
- Access
- Public
- Created
- May 2, 2026
- Updated
- May 3, 2026
- SHA
- 9d365009
Model card
From the Hugging Face model README
Llaza-3.2-1B-MS2F-4K-v0.1
Training checkpoint from zip2zip-core. This is a candidate model (not production-ready).
Training Config
| Field | Value |
|---|---|
| model_config | 1B_llama3.2 |
| init_from | meta-llama/Llama-3.2-1B |
| max_subtokens | 2 |
| max_codebook_size | 4096 |
| seq_len | 4096 |
| lr | 5e-05 |
| max_tokens | 1000000000 |
| step | 1908 |
| data | zip2zip-plus-mixture-20b |
Usage
This is a training checkpoint (torchtitan format). To use for inference, export to HuggingFace format first:
python scripts/zip2zip_hf/export_to_zip2zip.py \
--ckpt_dir <local_path>/step_1908 \
--output_dir <export_dir> \
--base_model meta-llama/Llama-3.2-1B \
--model_config 1B_llama3.2
- Library
- zip2zip
- Created
- May 2, 2026
- Updated
- May 3, 2026