Olmo-3-7B-Instruct config.json
repo
Fields
id
sources/model-fr-hf-config-olmo-3-7bPublisher
Hugging Face (Ai2)
Kind
repo
Published
2025-11-19
Accessed
2026-10-03
notes
hidden 4096, 32 layers, 32 q heads, 32 kv heads (full MHA, no GQA), intermediate 11008, sliding_window 4096, layer_types with 8 full_attention layers out of 32 (3 sliding : 1 full), rope_scaling yarn factor 8 from an 8192 base, max_position_embeddings 65536, vocab 100278. | SOURCE AUDIT 2026-10-03: kind was 'spec-sheet' but the URL is a raw Hugging Face config.json served as text/plain, not a vendor spec sheet; set to 'repo'. The schema field it fills (layer/window/head counts) is machine-readable model metadata and should be read as such.
Notes
hidden 4096, 32 layers, 32 q heads, 32 kv heads (full MHA, no GQA), intermediate 11008, sliding_window 4096, layer_types with 8 full_attention layers out of 32 (3 sliding : 1 full), rope_scaling yarn factor 8 from an 8192 base, max_position_embeddings 65536, vocab 100278. | SOURCE AUDIT 2026-10-03: kind was 'spec-sheet' but the URL is a raw Hugging Face config.json served as text/plain, not a vendor spec sheet; set to 'repo'. The schema field it fills (layer/window/head counts) is machine-readable model metadata and should be read as such.
Cross-references
This record references no other records.