Spaces:

rishabh16196
/

prompt_golf_env

Sleeping

App Files Files Community

prompt_golf_env / training

121 kB

Ctrl+K

Ctrl+K

3 contributors

History: 39 commits

Don Rishabh

training/TRAINING.md: add "Quick start — just run the .sh" subsection

96d773b 12 days ago

TRAINING.md

16.2 kB
training/TRAINING.md: add "Quick start — just run the .sh" subsection 12 days ago
build_before_after_csv.py

14.1 kB
build_before_after_csv: --min-verbose-accuracy flag 12 days ago
eval_before_after.py

9.73 kB
tasks_policy: long-context policy-compression tasks 13 days ago
hf_job_eval.sh

3.11 kB
v3: multi-turn env, thinking tokens, cross-family Qwen->Llama, multi-step GRPO 13 days ago
hf_job_profile.sh

2.23 kB
v3: multi-turn env, thinking tokens, cross-family Qwen->Llama, multi-step GRPO 13 days ago
hf_job_train.sh

4.62 kB
hf_job_train: add ENABLE_THINKING env var (default true) 13 days ago
hf_job_train_multistep.sh

4.08 kB
multistep: gradient checkpointing + tighter memory defaults 12 days ago
make_plots.py

4.04 kB
Initial commit: Prompt Golf environment for OpenEnv 15 days ago
profile_baseline.py

8.03 kB
tasks_policy: long-context policy-compression tasks 13 days ago
replay_to_trackio.py

7.12 kB
trackio: post-hoc replay of train_metrics.jsonl into a HF Space dashboard 12 days ago
requirements.txt

427 Bytes
Align HF Jobs deps with spaces_pipeline_env Colab stack 14 days ago
train_grpo.py

21.8 kB
tasks_policy: long-context policy-compression tasks 13 days ago
train_grpo_multistep.py

25.5 kB
multistep: gradient checkpointing + tighter memory defaults 12 days ago