| { |
| "status": "ok", |
| "started_at": 1777163399.0780032, |
| "finished_at": 1777164656.2574434, |
| "commands": [ |
| { |
| "args": [ |
| "python", |
| "scripts/bootstrap_data.py" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 0.504 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/build_training_corpus.py", |
| "--profile", |
| "massive", |
| "--with-local", |
| "--with-synthetic", |
| "--with-hf" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 4.013 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/train_sft_trl.py", |
| "--model-id", |
| "Qwen/Qwen2.5-0.5B-Instruct", |
| "--dataset-path", |
| "data/processed/training_corpus_sft.json", |
| "--output-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-0-5b-instruct", |
| "--report-path", |
| "outputs/reports/sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json", |
| "--epochs", |
| "2", |
| "--max-steps", |
| "0", |
| "--batch-size", |
| "2", |
| "--max-seq-len", |
| "512", |
| "--learning-rate", |
| "2e-05", |
| "--use-unsloth" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 251.4 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/merge_adapters_safe.py", |
| "--adapter-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-0-5b-instruct/sft_adapter", |
| "--output-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-0-5b-instruct/merged" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 7.16 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/test_inference_postsave.py", |
| "--samples", |
| "5", |
| "--base-model", |
| "Qwen/Qwen2.5-0.5B-Instruct", |
| "--merged-model", |
| "checkpoints/sweeps/qwen-qwen2-5-0-5b-instruct/merged", |
| "--adapter-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-0-5b-instruct/sft_adapter", |
| "--output", |
| "outputs/reports/sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 15.213 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/train_sft_trl.py", |
| "--model-id", |
| "Qwen/Qwen2.5-1.5B-Instruct", |
| "--dataset-path", |
| "data/processed/training_corpus_sft.json", |
| "--output-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-1-5b-instruct", |
| "--report-path", |
| "outputs/reports/sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json", |
| "--epochs", |
| "2", |
| "--max-steps", |
| "0", |
| "--batch-size", |
| "1", |
| "--max-seq-len", |
| "512", |
| "--learning-rate", |
| "2e-05", |
| "--use-unsloth" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 504.997 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/merge_adapters_safe.py", |
| "--adapter-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-1-5b-instruct/sft_adapter", |
| "--output-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-1-5b-instruct/merged" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 10.634 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/test_inference_postsave.py", |
| "--samples", |
| "5", |
| "--base-model", |
| "Qwen/Qwen2.5-1.5B-Instruct", |
| "--merged-model", |
| "checkpoints/sweeps/qwen-qwen2-5-1-5b-instruct/merged", |
| "--adapter-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-1-5b-instruct/sft_adapter", |
| "--output", |
| "outputs/reports/sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 17.029 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/train_sft_trl.py", |
| "--model-id", |
| "Qwen/Qwen2.5-3B-Instruct", |
| "--dataset-path", |
| "data/processed/training_corpus_sft.json", |
| "--output-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-3b-instruct", |
| "--report-path", |
| "outputs/reports/sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json", |
| "--epochs", |
| "1", |
| "--max-steps", |
| "0", |
| "--batch-size", |
| "1", |
| "--max-seq-len", |
| "512", |
| "--learning-rate", |
| "2e-05", |
| "--use-unsloth" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 394.356 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/merge_adapters_safe.py", |
| "--adapter-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-3b-instruct/sft_adapter", |
| "--output-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-3b-instruct/merged" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 15.472 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/test_inference_postsave.py", |
| "--samples", |
| "5", |
| "--base-model", |
| "Qwen/Qwen2.5-3B-Instruct", |
| "--merged-model", |
| "checkpoints/sweeps/qwen-qwen2-5-3b-instruct/merged", |
| "--adapter-dir", |
| "checkpoints/sweeps/qwen-qwen2-5-3b-instruct/sft_adapter", |
| "--output", |
| "outputs/reports/sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 20.373 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/evaluate_baselines.py" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 4.112 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/evaluate_all.py" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 3.787 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/evaluate_compare_runs.py", |
| "--baseline", |
| "outputs/reports/baselines.json", |
| "--candidate", |
| "outputs/reports/benchmark_report.json", |
| "--output", |
| "outputs/reports/improvement_report.json" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 0.033 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/benchmark_inference.py" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 2.376 |
| }, |
| { |
| "args": [ |
| "python", |
| "scripts/generate_hf_training_report.py", |
| "--mode", |
| "sft-baseline" |
| ], |
| "returncode": 0, |
| "elapsed_seconds": 1.791 |
| } |
| ], |
| "artifact_repo_id": "TheJackBright/polyguard-openenv-sft-baseline-artifacts", |
| "training_mode": "sft-baseline", |
| "model_sweep": [ |
| "Qwen/Qwen2.5-0.5B-Instruct", |
| "Qwen/Qwen2.5-1.5B-Instruct", |
| "Qwen/Qwen2.5-3B-Instruct" |
| ], |
| "improved": true, |
| "anti_hacking_passed": true, |
| "completed_run_ids": [ |
| "qwen-qwen2-5-0-5b-instruct", |
| "qwen-qwen2-5-1-5b-instruct", |
| "qwen-qwen2-5-3b-instruct" |
| ] |
| } |
