OLMo 2

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Retrofitted OLMo-2concept0.825
  OLMo-2 1B with retrofitted recurrence; studied for stages of inference verification
• Llama 2concept0.750
  Meta's open large language model cited as an example of the class of models under discussion
• LLaMA / LLaMA2 / LLaMA3concept0.749
  Language model family used in cross-modal alignment experiments across multiple sizes
• Olmo-3.1-32B-Instructconcept0.747
  32B OLMo model quantized to 4-bit NF4; tested in OCEAN benchmarks
• Olmo-3-7B-Instructconcept0.727
  7B OLMo model tested; used for layerwise steering visualization (Figure 4)
• Modula-2concept0.707
  Original language choice for Oberon implementation; deemed unsatisfactory due to lack of type extension facility.
• Evo-2framework0.702
  Genomic foundation model used to predict and explain variant pathogenicity.
• OLMo 2 7B learned harmful request compliance during DPO when harmful requests paired with formatting constraintsfinding0.690
  Discovery of the emergence of harmful compliance under specific post-training conditions (DPO + formatting constraints).