Spaces:

Humanlearning
/

Cyber_analyst-round1

Sleeping

85.3 kB

Ctrl+K

1 contributor

feat: introduce GRPO GPU fallback support, enhance training script with warmstart tagging, and add learning rate parameter for improved training flexibility

1b6d30b 12 days ago

configs
feat: introduce GRPO GPU fallback support, enhance training script with warmstart tagging, and add learning rate parameter for improved training flexibility 12 days ago
__init__.py

63 Bytes
feat: update training configuration and documentation for Modal execution, including new model integration and enhanced tracking utilities 12 days ago
eval_before_after.py

2.01 kB
feat: integrate Trackio for experiment tracking and add Modal training infrastructure with environment and test utilities. 13 days ago
grpo_curriculum.py

10.2 kB
feat: enhance CyberSecurity_OWASP observation model with scenario prompt, improve GRPO batch configuration validation, and add scenario grouping for adaptive difficulty curriculum 12 days ago
reward_funcs.py

1.28 kB
feat: enhance scenario authoring and caching mechanisms, update action submission terminology, and improve reward configuration for CyberSecurity_OWASP environment 12 days ago
rollout.py

5.07 kB
feat: enhance scenario authoring and caching mechanisms, update action submission terminology, and improve reward configuration for CyberSecurity_OWASP environment 12 days ago
trackio_utils.py

52.1 kB
feat: introduce reward ablation configurations for enhanced training flexibility, implement YAML loading with extends support, and add reward variant tracking in training scripts 12 days ago
train_grpo.py

3.17 kB
feat: enhance scenario authoring and caching mechanisms, update action submission terminology, and improve reward configuration for CyberSecurity_OWASP environment 12 days ago