Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
Spaces:
chane335
/
permanence-training
like
0
Paused
App
Files
Files
Community
Fetching metadata from the HF Docker repository...
main
permanence-training
/
tools
59.9 kB
Ctrl+K
Ctrl+K
1 contributor
History:
9 commits
chane335
PERMANENCE: reversibility-aware RL environment for training LLM agents
8f27137
verified
4 months ago
analyze_run6_confusion.py
Safe
2.74 kB
Run 6.1: env precondition fix (destructive DB ops on missing tables short-circuit)
4 months ago
fetch_run5.py
Safe
1.05 kB
Run 6: forced variants (eps 50%→70%), β_rank=0.25, R-level bonus, μ=2 PPO epochs, balanced R1-R5 warmup traces
4 months ago
fetch_run6.py
Safe
2.33 kB
Run 6: forced variants (eps 50%→70%), β_rank=0.25, R-level bonus, μ=2 PPO epochs, balanced R1-R5 warmup traces
4 months ago
fetch_run7.py
Safe
1.92 kB
Run 7: R4/R5 calibration traces — fix git context conflation + destructive DB R4 distinction
4 months ago
fetch_run8.py
Safe
2.19 kB
Run 8: disable unlikeliness shaping (β_rank=0.0) — fix classification-task reward inversion that collapsed Run 7 accuracy to 46%
4 months ago
generate_curves.py
Safe
11.7 kB
Sync: new curriculum, 6 tasks, composable rubric, latent dynamics
4 months ago
parse_run6_logs.py
Safe
2.11 kB
Run 6.1: env precondition fix (destructive DB ops on missing tables short-circuit)
4 months ago
render_results.py
Safe
9.49 kB
PERMANENCE: reversibility-aware RL environment for training LLM agents
4 months ago
replay_run6_eval.py
Safe
4.04 kB
Run 6.1: env precondition fix (destructive DB ops on missing tables short-circuit)
4 months ago
upload_eval_only.py
Safe
2.53 kB
Run 6.1: env precondition fix (destructive DB ops on missing tables short-circuit)
4 months ago
validate_submission.py
Safe
13.9 kB
PERMANENCE: reversibility-aware RL environment for training LLM agents
4 months ago
verify_forced_eval_seeds.py
Safe
2.79 kB
Run 6.1: env precondition fix (destructive DB ops on missing tables short-circuit)
4 months ago
verify_run7_traces.py
Safe
3.03 kB
Run 7: R4/R5 calibration traces — fix git context conflation + destructive DB R4 distinction
4 months ago