OLMo-7B is a large language model released by Allen Institute for AI, released under the Apache-2.0 license, with a claimed knowledge cutoff of 2024-02-01. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.
Description
AI2's first OLMo release, notable for being genuinely fully open -- open weights, open training data (Dolma), open training code, and open training logs, not just a weights drop -- explicitly positioned as a research artifact for studying how LLMs are actually trained, not just a capability release.
Metadata
Hugging Face stats
2,322 downloads on Hugging Face · 655 likes · pipeline: text-generation · library: transformers
Weight Hashes
Weights file
SHA-256b1d7671818c11442286bb3edfd40f727de3d56f8b1677fc07bba61c8a3aebf63
SHA3-256713e89b580c291b6153f8ef41ea91b52711fb96136ff73f120bd4b0d2ea622dd
BLAKE3d44fdc52ab63a8e68135bfbc346764ddb3b20490df90bf5ce2c73259106f42b0
Config file
SHA-25635e3cc79de6a46a9d61133f4a18e675f37de4cee9bddcafefe726516904f5ed8
SHA3-2562aaff3b99e8e91aa7160264be10c59e375292897d41cb1a7970e3de0d0150cdc
BLAKE34aef8dc68785ed9db41ae79f9e7ce98ff4f823eaffa20e01e6b743dcd2fea291
Tokenizer file
SHA-2569ad33b4b39a9f83973c3f8c42a01948dd5b877a28ac9a5356956c4ff4ed0b714
SHA3-256db09e410e5e76706d486b705fa6693136ae64b70f9412fb2fd81bdcffaf205c5
BLAKE3668cb72c3636ccad80f2687c1a6c356255a9505481d3acbde896ea990deae48c
Verify it yourself
curl -L -o "model.safetensors" "https://huggingface.co/allenai/OLMo-7B/resolve/0f605aa2bf7f9585bcedfa9b419326f15b453ddd/model.safetensors" sha256sum "model.safetensors"
Expected SHA-256: b1d7671818c11442286bb3edfd40f727de3d56f8b1677fc07bba61c8a3aebf63
Anomalies Flagged
- other weight-like files present but not hashed: ['pytorch_model.bin']