Dataset Viewer
Auto-converted to Parquet Duplicate
format
string
key
string
tokenizer
string
vocab_size
int64
pad_id
int64
seq_len
int64
sequences
int64
seed
int64
shuffle_merged
bool
shard_sequences
int64
shards
int64
datasets
list
datamill/packed-v1
1ac0653d7eb59f7b4d205f7e
tiktoken:tokenizer/tiktoken.model
32,768
32,767
2,048
9,569,531
2
true
16,384
585
[ { "name": "SYNTH", "cache": "7a7fa483328fa9ec4065c28b", "n_docs": 2601774, "n_tokens": 2500000805, "weight": 1 }, { "name": "web-hq", "cache": "0dc40c039d610b2b9f2a9f71", "n_docs": 2700282, "n_tokens": 2500000651, "weight": 1 }, { "name": "web-2025-47", "cache...

No dataset card yet

Downloads last month
-