Fetching metadata from the HF Docker repository... Ankit
Updated Loss Reward
623b83c - __pycache__ training results
- agents reward curve
- artifacts added
- configs fix: GRPO zero-gradient trap, DSL prompt seed, best_reward, teacher oracle
- scripts fix: GRPO zero-gradient trap, DSL prompt seed, best_reward, teacher oracle
- server training results
- tests imp DSL, reward structure, paradigm shift
- 53 Bytes V1
- 324 Bytes Track jsonl with LFS
- 79 Bytes discovery plot
- 28.7 kB ipynb
- 4.85 kB Updated Loss Reward
- 2.64 kB dockerfile for hf dep
- 9.63 kB Add LFS tracking
- 1.81 kB Add LFS tracking
- 1.06 kB submodule access,timeout, gro checkpoints, mdl penalty
- 5.9 kB discovery plot
- 2.98 kB V1
- 3.68 kB imp DSL, reward structure, paradigm shift
- 2.58 kB fix: GRPO zero-gradient trap, DSL prompt seed, best_reward, teacher oracle
- 3.35 kB V1
- 2.68 kB submodule access,timeout, gro checkpoints, mdl penalty
- 606 Bytes imp DSL, reward structure, paradigm shift
- 1.89 kB reward curve
- 17.5 kB imp DSL, reward structure, paradigm shift
- 13.8 kB submodule access,timeout, gro checkpoints, mdl penalty
- 6.87 kB imp DSL, reward structure, paradigm shift
- 9.75 kB imp DSL, reward structure, paradigm shift
- 18.3 kB discovery plot
- 13.8 kB fix: GRPO zero-gradient trap, DSL prompt seed, best_reward, teacher oracle
- 14.3 kB imp DSL, reward structure, paradigm shift
- 1.08 MB reward curve
