Model · bloom-560m
bloom-560m
Developer: BigScience
Availability: available · checked 2026-09-24 · source
Raw record: /data/models/bloom-560m.json
Fields
- id
- bloom-560m
- identifiers
- huggingface
- bigscience/bloom-560m
- developer
- BigScience
- release_date
- 2022-05-19partial · sourcenote: HF repo creation date; may predate or postdate public release.
- weights_status
- open
- availability
- available · checked 2026-09-24 · source
- license
- bigscience-bloom-rail-1.0partial · sourcenote: From HF card metadata (uploader-declared); confirm against the license text.
- architecture
- family
- decoder_only
- note
- family read from config 'architectures': ['BloomForCausalLM'].
- n_layers
- 24recorded · source
- hidden_size
- nullnot_recorded
- n_heads
- 16recorded · source
- vocab_size
- 250880recorded · source
- context_length
- nullnot_recorded
- positional_encoding
- ALiBi (attention bias by distance; no position embeddings)recorded · source
- training_data
- ROOTS corpus: 46 natural and 13 programming languages, ~1.5TB / 350B unique tokens. Same recipe as the 7.1B model.recorded · source
- techniques
- primary_sources
- record_history
- date:2026-09-24 · change:ingested as candidate from HF (bigscience/bloom-560m@ac2ae5fab2ce3f9f40dc79b5ca9f637430d24971) · by:ingest_hf.py ·date:2026-09-24 · change:preparer: developer, positional encoding, trained_on set (matches bloom-7b1) · by:claude (preparer, Sonnet 5) ·date:2026-09-24 · change:preparer cleanup: removed 2 stale flag(s), filled training_data · by:claude (preparer, Sonnet 5) ·date:2026-09-24 · change:reviewed and promoted from staging (1 edge(s) accepted) · by:Wilson Pruitt ·
Parents
Training data
- trained_on → ROOTS corpus declared source Paper describes one training recipe (ROOTS corpus) for the whole BLOOM suite, incl. the 560M model. Same as bloom-7b1.
Children
No edges recorded.
Read in
No station on the reading path has touched this record yet.