Downloads · 30 days
6
22% of all-time downloads
TEMA-Qwen2.5-Omni-7B-TemporalInit
bluesky7/TEMA-Qwen2.5-Omni-7B-TemporalInit
TEMA-Qwen2.5-Omni-7B-TemporalInit is a machine learning model from bluesky7. 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 peft. The card lists the license as apache-2.0.
All-time downloads
27
Public
Repo size
652 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors652 MB · 100%
At a glance
- Library
- peft
- License
- apache-2.0
- Access
- Public
- Created
- Sep 23, 2026
- Updated
- Sep 25, 2026
- SHA
- ebf317d4
Base models
Model card
From the Hugging Face model README
TEMA Temporal Initialization
Temporal-grounding LoRA adapter for Qwen2.5-Omni-7B, used to initialize TEMA dialogue SFT.
| Setting | Value |
|---|---|
| Objective | Single-audio temporal localization |
| Training steps | 1,867 |
| LoRA rank / alpha | 64 / 128 |
| Learning rate | 1e-5 |
Task definitions
For the downstream TEMA-Dialog and TEMA-Bench tasks, see the 18-subtask guide, which covers definitions, examples, evidence requirements, and internal task labels.
Use
hf download bluesky7/TEMA-Qwen2.5-Omni-7B-TemporalInit --local-dir models/temporal-init
scripts/train_sft.sh --model models/qwen-original --init-adapter models/temporal-init \
--data-root data_local --output runs/sft --gpus 0,1,2,3
- Library
- peft
- License
- apache-2.0
- Languages
- en
- Created
- Sep 23, 2026
- Updated
- Sep 25, 2026