Downloads · 30 days
13
11% of all-time downloads
armor-dp-babylm-2026-d768-currdyn
liaocccccc/armor-dp-babylm-2026-d768-currdyn
armor-dp-babylm-2026-d768-currdyn is a text generation model from liaocccccc. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This is the submission-facing ARMOR-DP checkpoint for the BabyLM 2026 multilingual track. It uses UTF-8 byte tokenization, language adapters, and a curriculum-dynamic exposure controller over English, Dutch, and Chinese.
All-time downloads
119
Public
Parameters
96.5M
386 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors386 MB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- apache-2.0
- Model type
- babylm-tinylm
- Access
- Public
- Created
- Jul 20, 2026
- Updated
- Jul 20, 2026
- SHA
- 22108457
Try a prompt
Model card
From the Hugging Face model README
ARMOR-DP BabyLM 2026 d768 Curriculum-Dynamic
This is the submission-facing ARMOR-DP checkpoint for the BabyLM 2026 multilingual track. It uses UTF-8 byte tokenization, language adapters, and a curriculum-dynamic exposure controller over English, Dutch, and Chinese.
- Repository:
liaocccccc/armor-dp-babylm-2026-d768-currdyn - Source checkpoint:
final_100m10_l512_b4_d768l12_curriculum_dynamic - Target budget:
100000000 - Predicted non-padding tokens:
86991960 - Context length:
512 - Architecture: d=768, layers=12, heads=12
- Schedule:
curriculum_dynamic - Mechanism:
language_adapters - Max-epoch cap:
10 - Consumed epochs: EN=6.6105, NL=6.5892, ZH=6.3699
Use with trust_remote_code=True:
from transformers import AutoTokenizer, AutoModelForCausalLM
tok = AutoTokenizer.from_pretrained("liaocccccc/armor-dp-babylm-2026-d768-currdyn", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("liaocccccc/armor-dp-babylm-2026-d768-currdyn", trust_remote_code=True)
- Task
- Text Generation
- Library
- transformers
- Type
- babylm-tinylm
- License
- apache-2.0
- Created
- Jul 20, 2026
- Updated
- Jul 20, 2026