Upload queue_status.json with huggingface_hub

Delete stale LR=2e-4 model weights (7 models)

Hard-delete model: _test/hard_delete_test

[test] Register dummy model for hard_delete test

Hard-delete model: qwen2.5-0.5b/owt20M

Hard-delete model: math/qwen2.5-0.5b/math_1pct_seed5

Hard-delete model: math/qwen2.5-0.5b/math_1pct_seed4

Hard-delete model: math/qwen2.5-0.5b/math_1pct_seed3

Hard-delete model: math/qwen2.5-0.5b/math_1pct_seed2

Hard-delete model: math/qwen2.5-0.5b/math_1pct_seed1

Hard-delete model: math/qwen2.5-0.5b/math_1pct_seed0

Re-register base model (qwen2.5-0.5b/base)

Hard-delete model entry: math/qwen2.5-0.5b/math_1pct_seed3

Hard-delete model entry: math/qwen2.5-0.5b/math_1pct_seed4

Hard-delete model entry: math/qwen2.5-0.5b/math_1pct_seed5

Hard-delete model entry: math/qwen2.5-0.5b/math_1pct_seed0

Hard-delete model entry: qwen2.5-0.5b/owt20M

Hard-delete model entry: math/qwen2.5-0.5b/math_1pct_seed1

Hard-delete model entry: math/qwen2.5-0.5b/math_1pct_seed2

Hard-delete model entry: qwen2.5-0.5b/base

Upload queue_status.json with huggingface_hub

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed3

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed0

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed5

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed1

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed2

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed4

Upload queue_status.json with huggingface_hub

Update eval metrics: qwen2.5-0.5b/base

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Upload queue_status.json with huggingface_hub

Update eval metrics: qwen2.5-0.5b/owt20M

Update eval_results_path to hash-keyed path

Update base model eval accuracy from eval_results.jsonl

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed3

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed4

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed1

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed0

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed2

Upload queue_status.json with huggingface_hub

Update eval metrics: math/qwen2.5-0.5b/math_1pct_seed5

Update eval metrics: qwen2.5-0.5b/owt20M

Upload queue_status.json with huggingface_hub