diff --git a/.gitattributes b/.gitattributes
index 1e1387fe800b4a5b5fbd1cb39e73b9b3165c6f9c..7510f0d61defe3b9b601860acba0c7942cd12dbb 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -34404,3 +34404,31 @@ train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_32878-tokenized-chun
 train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_50344-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
 train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_74093-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
 train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_92367-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_92367-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_19440-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_74093-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_6798-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_61869-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49158-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_42577-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_22122-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_14762-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_6375-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_3947-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
+train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_83523-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds filter=lfs diff=lfs merge=lfs -text
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_14762-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_14762-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..88b04c0a66da856f4fb38106d5891d366622da44
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_14762-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56f80737269c21ce9e851a61847d10afe8e1c273602c8209487f2579a501462b
+size 15046331
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_22122-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_22122-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..56c13140055dba6c7ecd32baf0db2018b2c78ab7
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_22122-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97d928a461491a75128c273c5be9d250c93269b4bac55cb8d5d7dcf47f982015
+size 21268630
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_42577-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_42577-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..8d5a5d075f6daf5167c70cf1feaddf1d258ec301
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_42577-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ea0dd12a37b2fe1a7bc18ba0b3824330f3dfb52d8fd9ed56a662ce604e4a726
+size 67108399
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49158-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49158-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..8694481f08fc1b5a13d541945eb040eb37e5626e
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49158-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e87779f97bfafb223eb46774f4e16626fd2d831535f420e2c698e5c18f735c3
+size 67108079
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..a471e6ad224e7f28cc06e1a13c2f28b9c0e7222f
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab3579059f5884224dffb18b133a0a0e81a481a9c14bf590e275f26e73c0d2cf
+size 67108801
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..af2834a45035875c44cfaa26c6ea416554d40ab2
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_49850-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87d40ae9e8ede118a039988c9cc37f4de1c4773cdb22015ed517783c40a0a8e2
+size 18064566
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_6375-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_6375-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..2658af33a0aa1248d625e5af7c2e6418f93d91dd
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v2/split_6375-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db9b2ee7b6ec4e72b5d83b2ea2535a828b27be2ae295c1a442e61d5e3e60f1ea
+size 67108650
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_13877-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_13877-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..aeded3978774e04b5fe5762b2e90bb983ed3a9c0
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_13877-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107968, "hashes": {}}, "samples": 44684, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47850797, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 7814519, "hashes": {}}, "samples": 5172, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 5573484, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_13877-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_13877-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..e03810738befb365043c048fd746b5360526a9e8
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_13877-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 35918368,
+  "num_truncated_tokens": 35891889
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_16304-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_16304-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..317efa63ec0d08a1e644a4aca268a6a3fab3f969
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_16304-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108286, "hashes": {}}, "samples": 42620, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47441180, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 20789992, "hashes": {}}, "samples": 13307, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 14736433, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_16304-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_16304-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..509ba00feb92220a37849fc8e08895d5a588bc31
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_16304-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 42217883,
+  "num_truncated_tokens": 42181414
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_19929-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_19929-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..6cbc259c47627812f3ba9c8669ff6369fcf3c840
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_19929-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67106966, "hashes": {}}, "samples": 44423, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47925361, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 10189009, "hashes": {}}, "samples": 6760, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 7274195, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_19929-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_19929-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..2e1ca8dd0a1f85531a2a67de627907c50823c98a
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_19929-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37064078,
+  "num_truncated_tokens": 37035240
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_1999-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_1999-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..1870157a165b260424a44732fcb1bde8492a30f9
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_1999-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107895, "hashes": {}}, "samples": 44393, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47759464, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 10623585, "hashes": {}}, "samples": 7063, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 7595769, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_1999-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_1999-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..659d92561861aeb564d4ef491f9da50e6db5a2fe
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_1999-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37273346,
+  "num_truncated_tokens": 37243948
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_22353-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_22353-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..3bc58507823e7e9c4cf4ac19e8ba872eab92ec66
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_22353-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67106927, "hashes": {}}, "samples": 44265, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47738561, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 9387838, "hashes": {}}, "samples": 6285, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 6690246, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_22353-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_22353-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..f4c6a19cdb1ecb44048332159c5b10518e381bb6
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_22353-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 36682916,
+  "num_truncated_tokens": 36655313
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_23808-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_23808-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..146dbbcbb56ddc3d134a406041ad2522882019d2
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_23808-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107158, "hashes": {}}, "samples": 44188, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47724656, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 12236697, "hashes": {}}, "samples": 7935, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8708952, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_23808-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_23808-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..0ba9e4aa6af7e5c8268874f7bb0d14630cc232ba
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_23808-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38058822,
+  "num_truncated_tokens": 38028779
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_24986-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_24986-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..9163dbfebd8035d43fe86f4da57ba6f340672fd9
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_24986-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108207, "hashes": {}}, "samples": 43671, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47671085, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 16584762, "hashes": {}}, "samples": 10578, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 11776110, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_24986-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_24986-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..efeadf19664e1f7a5de61dfc9bbd302263707a02
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_24986-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 40167541,
+  "num_truncated_tokens": 40133682
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_30156-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_30156-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..788ad4f99df0054caf442c6267802a75c2710a39
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_30156-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108084, "hashes": {}}, "samples": 43952, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47775697, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 14142813, "hashes": {}}, "samples": 8940, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 10016314, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_30156-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_30156-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..dbedbf18b89293ae502f2addfaf88c5953810b74
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_30156-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38988468,
+  "num_truncated_tokens": 38957313
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_32568-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_32568-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..8c33cc35f2ef0de1ec63b410c40f6c0ea60e806f
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_32568-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107992, "hashes": {}}, "samples": 43556, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47585966, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 16536180, "hashes": {}}, "samples": 10652, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 11744863, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_32568-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_32568-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..a46e8c8814eb903a55d41581847c64b41096ad86
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_32568-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 40144399,
+  "num_truncated_tokens": 40110984
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_33781-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_33781-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..2b1408a5f619a43b34c6a620e13c34d083ca0243
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_33781-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108125, "hashes": {}}, "samples": 43832, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47701635, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 11779513, "hashes": {}}, "samples": 7721, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8396947, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_33781-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_33781-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..d25fe2099712377170ce44fe81b0d17ab2ab6dbd
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_33781-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37848226,
+  "num_truncated_tokens": 37818912
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_34356-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_34356-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..e1dbbed73385084c7c722f3fe45c7c1f535f10ae
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_34356-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108267, "hashes": {}}, "samples": 43355, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47741092, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 17564486, "hashes": {}}, "samples": 11156, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 12420382, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_34356-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_34356-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..7a2c78e3e700368c78fe61e50c57ab35961d92b3
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_34356-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 40649133,
+  "num_truncated_tokens": 40615196
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_36628-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_36628-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..921e019ae86f32a709932a05356e8ea8723c3632
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_36628-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108261, "hashes": {}}, "samples": 44409, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47809231, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 12445295, "hashes": {}}, "samples": 7916, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8783937, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_36628-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_36628-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..2f806bd8c69e49aa9ff24baa8dc583f087ba60f5
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_36628-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38157449,
+  "num_truncated_tokens": 38126663
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_40648-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_40648-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..8cc93178a83c35175d1003d758d5a49cd07f2875
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_40648-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 39964463,
+  "num_truncated_tokens": 39931396
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_44631-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_44631-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..338220cb86a8d3eaacdde35eaa30e7711c7d469a
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_44631-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107013, "hashes": {}}, "samples": 44316, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47868453, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 10846596, "hashes": {}}, "samples": 7112, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 7697089, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_44631-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_44631-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..657b6d1266a274a286d1dbb31ce2774c75f22efe
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_44631-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37385307,
+  "num_truncated_tokens": 37356030
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45066-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45066-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..fc4f3fb69d3d1632cc48503c535964aa3b2dcbad
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45066-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107799, "hashes": {}}, "samples": 43241, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 48069941, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 16609133, "hashes": {}}, "samples": 10744, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 11937218, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45066-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45066-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..28ba8663a1553cc244e8e7e10b14b86b35e2ed88
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45066-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 40187404,
+  "num_truncated_tokens": 40153955
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45645-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45645-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..6d9048400dc8d57b77342cc8d997b56f12a29e0f
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45645-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107844, "hashes": {}}, "samples": 44062, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47660630, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 12254810, "hashes": {}}, "samples": 8257, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8750404, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45645-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45645-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..badd8ccc2b38da477fd14b25cd7c068d72c7580e
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_45645-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38062275,
+  "num_truncated_tokens": 38031519
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_46238-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_46238-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..fc63b4345771386aaf9f37f987728f8f6c990fdc
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_46238-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108439, "hashes": {}}, "samples": 42602, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47882635, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 21901005, "hashes": {}}, "samples": 13840, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 15437280, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_46238-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_46238-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..4583ccce0ecf8c2bcf6b6bf0e1de394ba8739805
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_46238-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 42757472,
+  "num_truncated_tokens": 42720113
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_49927-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_49927-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..0e428bef22ae48e216222e701659d027c69d1193
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_49927-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108611, "hashes": {}}, "samples": 43014, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47817122, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 19611344, "hashes": {}}, "samples": 12391, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 13961961, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_49927-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_49927-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..e4d6f9709f74509420eab9526a2c2bdefe58f54c
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_49927-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 41644942,
+  "num_truncated_tokens": 41609075
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_55848-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_55848-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..7d0803d0ace2c45eac5c3759b288718cd882ae2d
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_55848-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107460, "hashes": {}}, "samples": 44636, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47897268, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 7947673, "hashes": {}}, "samples": 5308, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 5688402, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_55848-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_55848-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..e48ea3ac983b679f5e27d77b3ea52bf6ad0425c2
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_55848-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 35981972,
+  "num_truncated_tokens": 35955189
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_58354-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_58354-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..4000154893ea991e93e0b4feed1173c05d1dd7d7
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_58354-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108270, "hashes": {}}, "samples": 42327, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47609315, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 21401123, "hashes": {}}, "samples": 13830, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 15281390, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_58354-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_58354-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..76b222d57def0db2c382d8129ff65a52f35981a7
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_58354-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 42516340,
+  "num_truncated_tokens": 42479398
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_74868-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_74868-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..c8edbbe2b2bf2564e8b4d31e9446ef609f5111ae
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_74868-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107379, "hashes": {}}, "samples": 44028, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47738354, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 12500253, "hashes": {}}, "samples": 8111, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8775184, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_74868-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_74868-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..7e218bb300651ef085bc2fff7248569ab973233f
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_74868-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38190173,
+  "num_truncated_tokens": 38159799
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_75855-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_75855-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..72df7589044c782babf9b94ace5f3fdcd1e8b937
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_75855-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108309, "hashes": {}}, "samples": 43595, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47918781, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 14418856, "hashes": {}}, "samples": 9402, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 10284419, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_75855-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_75855-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..57efa09d38d0350c438c6689c49446c776c530ed
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_75855-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 39123249,
+  "num_truncated_tokens": 39091107
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78573-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78573-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..d71894127ecbb8a8a9087b58f0b83f8217dff123
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78573-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107900, "hashes": {}}, "samples": 43630, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47863273, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 14121022, "hashes": {}}, "samples": 9018, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 10132598, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78573-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78573-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..f1ee44602c13c494ebad7352bf237f87075b13ff
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78573-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38984825,
+  "num_truncated_tokens": 38953407
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78824-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78824-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..2f1dd18d7b9453f08078a181aecbd930eee41f03
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78824-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108350, "hashes": {}}, "samples": 42983, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47518270, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 19225843, "hashes": {}}, "samples": 12267, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 13592201, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78824-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78824-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..566fb2f88ac2dd8908f31cf2f13a84d6d4ca17d3
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_78824-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 41456878,
+  "num_truncated_tokens": 41422097
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_80564-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_80564-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..45d36e389a1da7bbe70394d19586fba8232a20f7
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_80564-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 40671936,
+  "num_truncated_tokens": 40638306
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_84900-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_84900-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..8a054973d644465e248f9416ff95a30e073bb23b
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_84900-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107332, "hashes": {}}, "samples": 42841, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47630557, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 19898461, "hashes": {}}, "samples": 12757, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 14086282, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_84900-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_84900-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..a573bc7b8ae57873f35e0db4ab597ab592556e93
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_84900-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 41781960,
+  "num_truncated_tokens": 41745861
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_86165-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_86165-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..a3c0c3c66aada532684f704765a2ebe331e070db
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_86165-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108593, "hashes": {}}, "samples": 42719, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47555267, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 21994316, "hashes": {}}, "samples": 13845, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 15631046, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_86165-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_86165-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..aad2a3ef20995befe72826f3e8f39c317964d394
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_86165-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 42800503,
+  "num_truncated_tokens": 42762924
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_88176-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_88176-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..e81f3941b42bcb65c92436b05dc8dfc822a3cffa
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_88176-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107243, "hashes": {}}, "samples": 44326, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47811906, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 8815521, "hashes": {}}, "samples": 5897, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 6321921, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_88176-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_88176-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..c21d97c973a285f1d7b5deae14959257be66e078
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v3/split_88176-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 36407052,
+  "num_truncated_tokens": 36380148
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_38717-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_38717-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..88fb43349f650ca2e4be73d52b8debd7e77eac16
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_38717-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108762, "hashes": {}}, "samples": 43269, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47979766, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 16218755, "hashes": {}}, "samples": 10510, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 11619172, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_38717-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_38717-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..3b51af113cf2d9f06425fe94dac3c488ea8c014d
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_38717-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 39999070,
+  "num_truncated_tokens": 39965853
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_52198-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_52198-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..8d196a104f818eefcd9499c489623a17034b4d09
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_52198-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37295904,
+  "num_truncated_tokens": 37267370
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_64813-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_64813-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..fd75cfd9e40a0a8c33a8ecc43a89c26ee52afa36
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_64813-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108407, "hashes": {}}, "samples": 44314, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47781392, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 10807891, "hashes": {}}, "samples": 7140, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 7740563, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_64813-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_64813-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..134383f33b18a091384dfd7d2c8d4c94d586b684
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_64813-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37365828,
+  "num_truncated_tokens": 37336545
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_66330-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_66330-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..48a73f778cea2dd9a6f7948e2d15ec846a32258b
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_66330-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108465, "hashes": {}}, "samples": 44078, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 48071435, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 11418737, "hashes": {}}, "samples": 7489, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8156559, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_66330-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_66330-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..c21b1c78d5e03f72ac91a678a646bb0f7c14aeee
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_66330-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37667586,
+  "num_truncated_tokens": 37637884
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_75686-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_75686-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..ee37d6f3589bf9c65d6ed07c26a8157605ae79dd
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_75686-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107133, "hashes": {}}, "samples": 43986, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47634166, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 11071329, "hashes": {}}, "samples": 7283, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 7846243, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_75686-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_75686-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..3a3fb8aa9c447cb5c59d6ec88bc46cf79f1dea9c
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_75686-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37502412,
+  "num_truncated_tokens": 37473725
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_76853-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_76853-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..cb4a4e55ad7503aab74a93537928f18965746f51
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_76853-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108192, "hashes": {}}, "samples": 42928, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47488083, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 21498062, "hashes": {}}, "samples": 13428, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 15171382, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_76853-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_76853-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..245103f454617bec83d0e197f782146c5e1eee62
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_76853-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 42558686,
+  "num_truncated_tokens": 42521729
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_77504-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_77504-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..4cb527a3509de8d3f6256eb7ffddbd01b882f9a3
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_77504-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107443, "hashes": {}}, "samples": 44093, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47960059, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 11636561, "hashes": {}}, "samples": 7696, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8307793, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_77504-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_77504-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..634bdb9d400cf73539a9e6e16d250c0fef3585e6
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_77504-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37769213,
+  "num_truncated_tokens": 37739308
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_80467-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_80467-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..d1d7d61466648e0cd59bd5fd5440b2e29c31d98e
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_80467-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107895, "hashes": {}}, "samples": 44367, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47619948, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 10046862, "hashes": {}}, "samples": 6584, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 7111429, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_80467-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_80467-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..ced44fdad5a6b2bab1ccd8701ce8cdccc77bc3c2
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_80467-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 37000540,
+  "num_truncated_tokens": 36972396
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_82492-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_82492-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..b1b753589326f342dc26f4555d5e1ade7f2002ef
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_82492-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108373, "hashes": {}}, "samples": 43667, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47670011, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 15103661, "hashes": {}}, "samples": 9851, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 10745042, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_82492-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_82492-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..150c7f5d972f06387583283bfb2939c88d0ccbd7
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_82492-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 39449720,
+  "num_truncated_tokens": 39416980
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_84949-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_84949-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..97b544ef9a01f55d6361d84d475df99559a93379
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_84949-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67107581, "hashes": {}}, "samples": 44139, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47773526, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 12483693, "hashes": {}}, "samples": 8104, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 8862920, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_84949-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_84949-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..3648771fc2eb7d6f7268985e1b15e30aac3fb877
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_84949-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 38178859,
+  "num_truncated_tokens": 38147960
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_89229-tokenized-chunked-1024-512-128-backfill-nodups/index.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_89229-tokenized-chunked-1024-512-128-backfill-nodups/index.json
new file mode 100644
index 0000000000000000000000000000000000000000..a54678d7afa6c6ad9b09453b2b98387ff3d3990a
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_89229-tokenized-chunked-1024-512-128-backfill-nodups/index.json
@@ -0,0 +1 @@
+{"shards": [{"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00000.mds", "bytes": 67108105, "hashes": {}}, "samples": 42506, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00000.mds.zstd", "bytes": 47352783, "hashes": {}}}, {"column_encodings": ["str", "ndarray:uint16"], "column_names": ["id", "input_ids"], "column_sizes": [null, null], "compression": "zstd", "format": "mds", "hashes": [], "raw_data": {"basename": "shard.00001.mds", "bytes": 20350437, "hashes": {}}, "samples": 13081, "size_limit": 67108864, "version": 2, "zip_data": {"basename": "shard.00001.mds.zstd", "bytes": 14465201, "hashes": {}}}], "version": 2}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_89229-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_89229-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
new file mode 100644
index 0000000000000000000000000000000000000000..1de5a0316d0163250e0e510587692afc91c53952
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled-v4/split_89229-tokenized-chunked-1024-512-128-backfill-nodups/num_tokens.json
@@ -0,0 +1,4 @@
+{
+  "num_tokens": 42008498,
+  "num_truncated_tokens": 41972839
+}
\ No newline at end of file
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_19440-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_19440-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..cb3f34a39a77d61f913fabc6b856ab11599895f7
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_19440-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:919423f3159ccc7a54074e7fb7bf139b122ff693a1674959b6f4e15c84e9e258
+size 24383997
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..b99c087ca111724d3d031ddd91f777fdea2ff570
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e01b172993721fe9d8d888b82fb2a3ccda48889d2bb535d095f26fa7d09e67e
+size 67108044
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..0bf8c3d0d851b92e085030d72d2d9c9d19367d36
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_23535-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b883acc14bec5fbc1f59cfea94dcdce1e997a525fde4e140063f546334d3f8d
+size 12687946
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..1b71fa4c091e75d3aa804a6e9c4b09672202a42a
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95fdd6f8182de3fbd419b3978f47a471d566d85cdd20553128dabc746c3a95d3
+size 67108640
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..cff300715eb627c1e0e150b568d05702ad26609d
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_33139-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f9d6542ee28864c055b8f84ce27e8b44997b802fa946a5e8d4fe108eee74e2a
+size 12488603
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_3947-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_3947-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..e626485c2a6952414c9ded85407010659fab7f0c
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_3947-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b51757aef9bd3ac166e088e40bce1ceb542c86d13b826533b2fc6544928d1b42
+size 67108159
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..bbd9ca23d2d4e0c32a2a2ad150fb6d08ce34d65e
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:420340a11c436822b3835c296ba10d2d9b24b45f3155ea66176aa6903b87c3a3
+size 67107372
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..c5ebeb2167a57710627f2ee95e052bd4d138b3c1
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_46311-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ec7469cb2ff60126fa0f473f89b1ea81a21ada4235886ca0b9c948eff3a0ab3
+size 12317833
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_61869-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_61869-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..244cbddc036075bfa7670e1f734d69f58d2fe745
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_61869-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:896a9dee92eca29cec34893fdc8e5dc16a1650ee220e94cda774be28362f2fb5
+size 16803915
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..52549a994c0d9194d57eabb479bad1052d374301
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8f7c8710b324ea1b71eadb9f532ddf4ad98baaee68cfc8ea13e4725e56d42b1
+size 67107229
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..51a4e236ed1ec3c7dad4140d3d52eebf9af4e6f4
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_62232-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48773d8f2ceb577d469dc730edc8e2cb67c1f0ece298f1005eb99e9c5973271c
+size 15886994
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..f06c622349a2406b5162b54bf3e2d18347298771
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9323e34671756e6e7031c2668a9d20fd50050ec5c2f524c0858f77e105f7f254
+size 67108225
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..d58ec50ca3de3c0c0a18473db3bdfb50f82051bc
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_63030-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91ed666574f476dd4dbd393e42b3a7950535a9b297fc29d3eb52a343dfe8a0bc
+size 11515631
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_6798-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_6798-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..dff74de40e2446cb6d311ac3d2b09115b12ea85b
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_6798-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa26c4940ac5535e8f03732fe34ed651403cb0add581ab54bbf8f38e2565f7f0
+size 10669352
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..ae08f7f5408ac763873621e30a4808cd65c472e9
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6afeb25ac07c2097f6483fe4ec624d3947d9fd14c032e1c91af1f6ed56dc4b86
+size 67108178
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..d431fffd36c40eb4e5fe0af7c4765be25291a12c
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_71561-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d0a519e0ed51c3097e0d9044621d6ec5fb727edf8adc5f550a776e49c69b23e
+size 16558160
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_74093-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_74093-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..92dd3b84c7e18d19883833de4e5bc1b193904774
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_74093-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5412a535312f8646838d608374105a28aba50c7e9f9cf01c5b0c9fae15f105e
+size 67108704
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_83523-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_83523-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..c12d7a1c65fb7cf16729042c85a451d68c0c40d0
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_83523-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:370c8e5de8576dcab74a6da99a3d7af829badfa9bd512abeff9c04f0a53dc43c
+size 10904981
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_92367-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_92367-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..7f271e9030c973a40a604f7a6f0bcbb84c80fbf5
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_92367-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16a5ef84806442f07da3a2e3d91dfcee96160784bbe96269d268f5b56f8c804f
+size 10842613
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
new file mode 100644
index 0000000000000000000000000000000000000000..05a3030e7a07d07dfab929ef878f5516a6b0fe25
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00000.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:554b32ae992b91332d9f98b7711cfd9b70ec9de12d4edf2880224bc18f8fbcc8
+size 67108469
diff --git a/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
new file mode 100644
index 0000000000000000000000000000000000000000..14c1a4c094bb1512fcbad653896833fdfa01fa05
--- /dev/null
+++ b/train/mlfoundations-dclm-baseline-1.0-parquet-sampled/split_96033-tokenized-chunked-1024-512-128-backfill-nodups/shard.00001.mds
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ce5066d2720c63fb0d3a98678d9f26674de30428e747b5ee077da976b5492b0
+size 14045539