Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
Spaces:
Draken1606
/
undertrial-ai
like
0
Running
App
Files
Files
Community
Fetching metadata from the HF Docker repository...
main
undertrial-ai
/
training
124 kB
Ctrl+K
Ctrl+K
2 contributors
History:
34 commits
Draken1606
Fix all stale 1.5B refs to 7B + LR/beta/completion fixes
7fa6a21
about 14 hours ago
UndertriAI_GRPO_Training.ipynb
Safe
6.89 kB
Fix notebook: remove hardcoded 384, use defaults (640, lr=5e-5, beta=0.04)
about 14 hours ago
__init__.py
Safe
1 Bytes
training script
12 days ago
parse_job_log.py
Safe
14.3 kB
Add training evidence: curriculum results, plots (LFS), parse_job_log helper
11 days ago
reward_unit_test.py
Safe
2.42 kB
training script
12 days ago
run_hf_job.py
Safe
6.98 kB
hf job
12 days ago
train_grpo.py
Safe
93.4 kB
Fix all stale 1.5B refs to 7B + LR/beta/completion fixes
about 14 hours ago
