vovw/yam_fold_70 / main_validation.json
vovw's picture
download
raw
1.15 kB
{
"status": "complete",
"code_branch": "main",
"code_commit": "5f8be1af5380d89ba4f645bc69a9c10c632a8025",
"checkpoint_repo": "hqfang/yam_fold_70",
"checkpoint_revision": "ea6b14c4ec0708872a20ef81fd24fdf140a34a23",
"source_step": 70000,
"sample_count": 4,
"sources": [
"allenai/MolmoAct2-BimanualYAM-Dataset",
"allenai/MolmoAct2-BimanualYAM-Dataset",
"lerobot/abc_130k_v3_train",
"lerobot/abc_130k_v3_train"
],
"output_shape": [
4,
30,
14
],
"strict_load": true,
"all_file_sha256_verified": true,
"all_loaded_tensors_verified": true,
"all_tensors_verified_against_source_cast": true,
"parameter_elements_by_dtype": {
"torch.bfloat16": 4864320464,
"torch.float32": 577875808
},
"bf16_exact_match_to_feature_branch": true,
"fp32_exact_match_to_feature_branch": true,
"cli_exact_match": true,
"fp32_to_bf16_max_abs_difference": 0.0,
"runtime": {
"torch": "2.11.0+cu128",
"transformers": "5.5.4",
"safetensors": "0.8.0",
"huggingface_hub": "1.30.0"
},
"scope": "Same four training-data input chunks and standalone CLI; not a closed-loop evaluation"
}

Xet Storage Details

Size:
1.15 kB
·
Xet hash:
60128ab5b8f9ec7e1c5f11c26baa8f87f812e605d51fcaca97fd14e5a1631d08

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.