diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..becc091a5fc82031039734f03f3377672be4b0fa --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b795496c9c302e4d189d3d026bf1e414700039878502561378ddf58362a7ab83 +size 7837872 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..b59fb25a630e78cdaca99936940baa155b04ec05 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7447721, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_00.json": { + "num_bytes": 8441128, + "checksum": null + } + }, + "download_size": 8441128, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15888849, + "splits": { + "train": { + "name": "train", + "num_bytes": 7447721, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..7cfc4cf53abc2ceb0e262203f840eb5508fbc794 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "fcdb0d51c472f4c5", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..ca7595473cb83bbe64ed69936075c1e50fd30cf5 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_00/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_00/train", + "rows": 5681, + "lengths_sample_mean_old": 3414.91, + "lengths_sample_mean_new": 4886.91, + "timestamp": "2026-05-11T15:38:42" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..bc19c76b8079166a1939e91188fde2023c1dfcbf --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a0631e215c50f00f42c5c05c315bca64a097eb1ba41b7eb9ec83ddd1ca1a9409 +size 7760272 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..9753431187f753de6088f363a3c32dc44a55eac6 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7368826, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_01.json": { + "num_bytes": 8374839, + "checksum": null + } + }, + "download_size": 8374839, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15743665, + "splits": { + "train": { + "name": "train", + "num_bytes": 7368826, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..a40927967141fc6180f6556fa4d9c3690745e0c6 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "1fa6968a695d7ed9", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..4a21a73a25cc41e2eeb4dbb3dd1cb2b71e2b4a1d --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_01/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_01/train", + "rows": 5681, + "lengths_sample_mean_old": 3423.39, + "lengths_sample_mean_new": 4895.39, + "timestamp": "2026-05-11T15:38:54" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..30b817d35905c65b51333119cc2387ddf0d3a9e8 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e2543b8b3429a0b696c74b11b53393d337f8682179423d3b2013708d59a0ab0d +size 7853432 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..6f954de72e7e74d0f41e01d288d7974880fca499 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7463924, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_02.json": { + "num_bytes": 8451092, + "checksum": null + } + }, + "download_size": 8451092, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15915016, + "splits": { + "train": { + "name": "train", + "num_bytes": 7463924, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..b33febbb73c9baf1a06fa5abd4124a111eadf40f --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "67922778fe789d97", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..b65b23d323b80a73e4988ac997c7ea4ca6521d8e --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_02/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_02/train", + "rows": 5681, + "lengths_sample_mean_old": 3379.45, + "lengths_sample_mean_new": 4851.45, + "timestamp": "2026-05-11T15:38:55" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..2324cfeedd5f019e1b3c1e47d31883f380670b2d --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:492dfdbc3d3fa27eab18aac4ba67a2deb3bf1d3b911d27885de11b3bb6bccafa +size 7934096 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..7857d783c3661746d540b0aefc631a9bf2e0598b --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7544302, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_03.json": { + "num_bytes": 8531728, + "checksum": null + } + }, + "download_size": 8531728, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16076030, + "splits": { + "train": { + "name": "train", + "num_bytes": 7544302, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..bb5c66de272cfaf073f54d537a6086e316a27c49 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "034aaf22051fa0ae", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..6d85706329f20f210ebc4343a97e68ebb7f1e9b1 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_03/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_03/train", + "rows": 5681, + "lengths_sample_mean_old": 3382.52, + "lengths_sample_mean_new": 4854.52, + "timestamp": "2026-05-11T15:38:56" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..21bdbd000d33473a02ff7ad5a67cb6ea69c82cd6 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c22a22113c5e2e046241aff63714ddcd44cb8c8140e553aa9df7fc3f66128210 +size 7676528 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..da60ed10e3f50ccfa09894c94607d3c1f63dd2f0 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7285950, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_04.json": { + "num_bytes": 8279218, + "checksum": null + } + }, + "download_size": 8279218, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15565168, + "splits": { + "train": { + "name": "train", + "num_bytes": 7285950, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..950aca7fea932c8ab6eedf3d6f0edbe47dd01291 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "5fa0e5d6216678a0", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..32682f8175642962174a055ce604847f24edb4b6 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_04/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_04/train", + "rows": 5681, + "lengths_sample_mean_old": 3430.91, + "lengths_sample_mean_new": 4902.91, + "timestamp": "2026-05-11T15:38:56" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..7168b1d901d4d451e84ffff83bd02c237132a87e --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:762b517a0a1e6a9b326029cfd36d2887cccf8d0fe5ccd8764e902a33bcc898d4 +size 7741288 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..f4a6592669cf04d687175727a1242a3157f1f9fc --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7351520, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_05.json": { + "num_bytes": 8342040, + "checksum": null + } + }, + "download_size": 8342040, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15693560, + "splits": { + "train": { + "name": "train", + "num_bytes": 7351520, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..c531636bf9a4412fc56ff9cac4656ec7bfe5a1e4 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "b220e4881c83491d", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..3e3b8aee12bcd46c69aa1cd88a0989fd3e368b7c --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_05/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_05/train", + "rows": 5681, + "lengths_sample_mean_old": 3404.79, + "lengths_sample_mean_new": 4876.79, + "timestamp": "2026-05-11T15:38:57" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..466d9607bbede51a86083b6f2a883eb9dee26283 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47774d9c292ae6dd9207825f1449ac86267dafbcac1ad6534c05321c67af6f59 +size 7977632 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..e4d829e34a1aabe81e5ecc49b355c8a6a3840114 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7588693, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_06.json": { + "num_bytes": 8571022, + "checksum": null + } + }, + "download_size": 8571022, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16159715, + "splits": { + "train": { + "name": "train", + "num_bytes": 7588693, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..9827cad7d9a0348a39e676050684ee392396a078 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "699bcf19aee2da88", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..f74c359c0869de81444dc54930569bf8e0628663 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_06/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_06/train", + "rows": 5681, + "lengths_sample_mean_old": 3432.08, + "lengths_sample_mean_new": 4904.08, + "timestamp": "2026-05-11T15:38:58" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..046126a688a6bc517f927b9b05bed4dc81f54853 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab4612f3e5fb5d040685446c5a7b01a3bc0793307086fc6b7aff195914e3ce16 +size 7979912 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..07000a9a735529ddee3ce91d219dae59261506ee --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7590275, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_07.json": { + "num_bytes": 8581233, + "checksum": null + } + }, + "download_size": 8581233, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16171508, + "splits": { + "train": { + "name": "train", + "num_bytes": 7590275, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..97e9fff85351c36bb6addb181c78b11f04519671 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "f15df0b8a41351d5", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..a2d407aa91e00434ad5c5c071676cf3093904a5f --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_07/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_07/train", + "rows": 5681, + "lengths_sample_mean_old": 3467.97, + "lengths_sample_mean_new": 4939.97, + "timestamp": "2026-05-11T15:38:58" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..c6d9df8456cb65d377accb82d0e03e80cc120ebd --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50747840a95d741f81227cca68f4e2e3ea23d2b3753f9d2a0b0da046a5f9de02 +size 7878016 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..85586e975ae9411d5ea80ea7035380bbc3b37566 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7487777, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_08.json": { + "num_bytes": 8478861, + "checksum": null + } + }, + "download_size": 8478861, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15966638, + "splits": { + "train": { + "name": "train", + "num_bytes": 7487777, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..756a505fd813b204ee074b52d422db2b6d82dfdb --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "efb20606a7897f10", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..3f6c91a42ea3ece67f89496e2a964486bb9a0d30 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_08/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_08/train", + "rows": 5681, + "lengths_sample_mean_old": 3423.25, + "lengths_sample_mean_new": 4895.25, + "timestamp": "2026-05-11T15:38:59" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..560701c12dfeb7671e58d112b232cef63e175e03 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:087cd98bc9fe8f0a0ec829c69a6cb3858308f95539634c3962e87d837a468ef3 +size 7931944 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..5e87277e15cf2bf55a0f54ba7bff9f233d97cd61 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7541773, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_09.json": { + "num_bytes": 8536644, + "checksum": null + } + }, + "download_size": 8536644, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16078417, + "splits": { + "train": { + "name": "train", + "num_bytes": 7541773, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..485058e4fabfe8fd7345ed50bf3f25060de43217 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "80f39657123e1abf", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..849309adac86a6211e6c3f385819a7d1165c020f --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_09/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_09/train", + "rows": 5681, + "lengths_sample_mean_old": 3423.52, + "lengths_sample_mean_new": 4895.52, + "timestamp": "2026-05-11T15:39:00" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..08120efa372acfbaf6501f8a76654daf90b5b1a4 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:84196cb29e4f2b038df58d8b36196f09e2479b30da98901093e277dfbc766906 +size 8073904 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..782f0e4335635fefec72255602260ddc4dce4908 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7684339, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_10.json": { + "num_bytes": 8671454, + "checksum": null + } + }, + "download_size": 8671454, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16355793, + "splits": { + "train": { + "name": "train", + "num_bytes": 7684339, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..6d689805f77eb9ab275366d1fbdddff505fd6885 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "40d0c61caab0704d", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..cdc722b32c85bee0fcf768e21046fd7d6a355925 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_10/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_10/train", + "rows": 5681, + "lengths_sample_mean_old": 3450.13, + "lengths_sample_mean_new": 4922.13, + "timestamp": "2026-05-11T15:39:00" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..a08a3d27ba5544ab9eb93495c11047e26af4d731 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8c9d335089cc30d89c1fadc34fa7ef9ec4892d7995bde9cb3e0186b9449b14e +size 7786696 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..565f5f20807c6a25464f4ec2d6c6ac2423d47d83 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7397014, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_11.json": { + "num_bytes": 8385641, + "checksum": null + } + }, + "download_size": 8385641, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15782655, + "splits": { + "train": { + "name": "train", + "num_bytes": 7397014, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..4c49f0e92255a0a719ed258a3f8c6101d5f9a83d --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "f24435657501f2f2", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..970b7fb13806c594ab3364b95fc538544a8bd402 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_11/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_11/train", + "rows": 5681, + "lengths_sample_mean_old": 3410.52, + "lengths_sample_mean_new": 4882.52, + "timestamp": "2026-05-11T15:39:01" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..529d1f95ebff21021040673a2f99cb95655bd83c --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f4b94e087f25942e59e0fe5bb89a165dc4895d3f0ba80c8c8c4988828e35b7c +size 7788360 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..0cb47f8271e399d38764948ce1b1265fa42f0408 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7397632, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_12.json": { + "num_bytes": 8394845, + "checksum": null + } + }, + "download_size": 8394845, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15792477, + "splits": { + "train": { + "name": "train", + "num_bytes": 7397632, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..56b72aa4f3a3a69d652bd10725fed38cfee7de28 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "6365462d25ec8cc5", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..a458010fcdc5f91a26a149f352b8df59da7d0738 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_12/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_12/train", + "rows": 5681, + "lengths_sample_mean_old": 3359.01, + "lengths_sample_mean_new": 4831.01, + "timestamp": "2026-05-11T15:39:02" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..cfc624dd138ca5cb5b8e50a22d565efa257d4217 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:970c2e6e5faf45897b621a8e944522e087c5c45adb3aa4f7f7758291981b38b0 +size 7796856 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..3f78b91818c9ddac7f26fb418556b151338fe132 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7405908, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_13.json": { + "num_bytes": 8404811, + "checksum": null + } + }, + "download_size": 8404811, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15810719, + "splits": { + "train": { + "name": "train", + "num_bytes": 7405908, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..509133c1c2cef596c10f2e21401dca589273d4c5 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "b04b7839b157007f", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..9f7128940abfb1a5c91e25a207be92f656e08c0e --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_13/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_13/train", + "rows": 5681, + "lengths_sample_mean_old": 3429.66, + "lengths_sample_mean_new": 4901.66, + "timestamp": "2026-05-11T15:39:03" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..9980727cc7335962320bae74c223f3fc3df0647c --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b8ed357d1c95691d43205cfe2d39ddb7211c787b4b9337b25f338a1327b195de +size 7916400 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..1f5f31f242c6c70523b2b48d035593617db53e0d --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7526045, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_14.json": { + "num_bytes": 8519986, + "checksum": null + } + }, + "download_size": 8519986, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16046031, + "splits": { + "train": { + "name": "train", + "num_bytes": 7526045, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..466a76ddf445b40ce2ab66cc232a842441fbabf8 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "d18724de3192c0f7", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..4857888ce4b46fead88ec08d7873674045c8f0d3 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_14/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_14/train", + "rows": 5681, + "lengths_sample_mean_old": 3396.02, + "lengths_sample_mean_new": 4868.02, + "timestamp": "2026-05-11T15:39:03" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..7c5dc1c394705413239e08ca1c4fe76773caa18c --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e7294c8ea9e50259f283327b010214a5ec00ba41fbfb0e4a0edc3d17de5872b +size 7844816 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..675c725ffcab63a63c7f6040add40f629c6302f4 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7455328, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_15.json": { + "num_bytes": 8440301, + "checksum": null + } + }, + "download_size": 8440301, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15895629, + "splits": { + "train": { + "name": "train", + "num_bytes": 7455328, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..5f7656db6f9be4ca4f69bc3841a864cbb1a175d8 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "46105b9ef5806df6", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..63ee985b9f882ba2310c5bceb8b04e34dfbc6ef4 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_15/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_15/train", + "rows": 5681, + "lengths_sample_mean_old": 3444.22, + "lengths_sample_mean_new": 4916.22, + "timestamp": "2026-05-11T15:39:04" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..2e86602049750ed6390875d627a249badea5f1f1 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78dbcf45290f6db4c3afa52c6c3703a80ae15287e34d09e65028b81f3d0460b0 +size 7902976 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..f4b0f5212e00804091a41e20f5b4a1681e679ecc --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7514211, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_16.json": { + "num_bytes": 8497254, + "checksum": null + } + }, + "download_size": 8497254, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16011465, + "splits": { + "train": { + "name": "train", + "num_bytes": 7514211, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..2c519a0676ecb1cf00b3b75bdfc63a2c2d4fce04 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "4042a93cddbe74a8", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..b36c72c4b79fa50c9db286aafe693504516faaa8 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_16/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_16/train", + "rows": 5681, + "lengths_sample_mean_old": 3417.89, + "lengths_sample_mean_new": 4889.89, + "timestamp": "2026-05-11T15:39:05" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..eb5828de7b511c25dc8e7a6de5de8402dd6dc2ab --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16f6d8f0857d5a6d8434bd216846285a217f20d85ab0b3f864fcf710ae16bc9b +size 7733216 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..e5a8862d6bc73fa2e77d2f630e5d6a883d312222 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7342615, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_17.json": { + "num_bytes": 8337621, + "checksum": null + } + }, + "download_size": 8337621, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15680236, + "splits": { + "train": { + "name": "train", + "num_bytes": 7342615, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..1094604c484e52a4d3d76f46444ac95652fff379 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "f96e38d76f48af8c", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..dc1718e020c984f261ebe3c33d621bfbf356b5fc --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_17/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_17/train", + "rows": 5681, + "lengths_sample_mean_old": 3408.96, + "lengths_sample_mean_new": 4880.96, + "timestamp": "2026-05-11T15:39:05" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..8d14d3f45c5b0f5e62d2d421ec89bc63eed4dd6d --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd67c96d5816dca61016612495ee8b6810b47ccd13afb295a07db89cdf1437c5 +size 7949272 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..ddb216f3b809932dbb778e54c17010ffdd06444e --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7559827, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_18.json": { + "num_bytes": 8548766, + "checksum": null + } + }, + "download_size": 8548766, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 16108593, + "splits": { + "train": { + "name": "train", + "num_bytes": 7559827, + "num_examples": 5681, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..351781f04e8232e3a960909a8262116b132889f8 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "34f2ad302a987515", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..04f64fe21ee40369e0eda8f7464719b582e8ae9f --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_18/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_18/train", + "rows": 5681, + "lengths_sample_mean_old": 3370.67, + "lengths_sample_mean_new": 4842.67, + "timestamp": "2026-05-11T15:39:06" +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/data-00000-of-00001.arrow b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/data-00000-of-00001.arrow new file mode 100644 index 0000000000000000000000000000000000000000..ea32c36dc437b345a6759230bbb0f90b187c2d0e --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/data-00000-of-00001.arrow @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6dcbf64278ed2036ebef6b5fe8fcabef18b25655206e27e471495c4d1436d48e +size 7830288 diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/dataset_info.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/dataset_info.json new file mode 100644 index 0000000000000000000000000000000000000000..51324591261cbedeaab4bf3cae73c0469c97e026 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/dataset_info.json @@ -0,0 +1,60 @@ +{ + "builder_name": "json", + "citation": "", + "config_name": "default", + "dataset_name": "json", + "dataset_size": 7441014, + "description": "", + "download_checksums": { + "/share/m2v_intern_v3/wangjunjie09/model_cache/VisionEncoder/ms-swift-data/video_sft_chunks_local/chunk_19.json": { + "num_bytes": 8427096, + "checksum": null + } + }, + "download_size": 8427096, + "features": { + "messages": [ + { + "content": { + "dtype": "string", + "_type": "Value" + }, + "role": { + "dtype": "string", + "_type": "Value" + } + } + ], + "videos": { + "feature": { + "dtype": "string", + "_type": "Value" + }, + "_type": "Sequence" + }, + "lengths": { + "feature": { + "dtype": "int64", + "_type": "Value" + }, + "_type": "Sequence" + } + }, + "homepage": "", + "license": "", + "size_in_bytes": 15868110, + "splits": { + "train": { + "name": "train", + "num_bytes": 7441014, + "num_examples": 5676, + "dataset_name": "json" + } + }, + "version": { + "version_str": "0.0.0", + "major": 0, + "minor": 0, + "patch": 0 + } +} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/state.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/state.json new file mode 100644 index 0000000000000000000000000000000000000000..1f62fa1f9e617c9d3a80f4e625e766e746898a18 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/state.json @@ -0,0 +1 @@ +{"_data_files": [{"filename": "data-00000-of-00001.arrow"}], "_fingerprint": "3802f4b9491a82c2", "_format_columns": null, "_format_kwargs": {}, "_format_type": null, "_output_all_columns": false, "_split": null} \ No newline at end of file diff --git a/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/vjepa21_patch_meta.json b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/vjepa21_patch_meta.json new file mode 100644 index 0000000000000000000000000000000000000000..5e4094adc569c7850e3ec31f43d8a0c365afd123 --- /dev/null +++ b/data/vmllm_cached/vjepa21/video_10pct_chunks/chunk_19/train/vjepa21_patch_meta.json @@ -0,0 +1,12 @@ +{ + "patcher": "siglip2_to_vjepa21", + "mode": "video", + "from": "siglip2 (SigLIP2 base 16f \u00d7 196 + 1 = 3137 video tokens)", + "to": "vjepa21 (V-JEPA 2.1 16f \u00d7 288 + 1 = 4609 video tokens)", + "delta_formula": "lengths[0] += 1472 * len(videos)", + "input": "/share/m2v_intern_v3/wangjunjie09/VisionEncoder/data/vmllm_cached/siglip2/video_10pct_chunks/chunk_19/train", + "rows": 5676, + "lengths_sample_mean_old": 3396.02, + "lengths_sample_mean_new": 4868.02, + "timestamp": "2026-05-11T15:39:07" +} \ No newline at end of file