a-t-c-g

3 ms·

1. 1.▲
   by a-t-c-g 6mo ago

   there are plenty of OSS finetuned models + base models around. If you're looking for doing these on your own dataset, worth getting in touch with cartesien.io or wire up https://github.com/SalesforceAIResearch/Pret
2. 2.▲
   by a-t-c-g 6mo ago

   natural language may provide part of the scaffolding for reasoning, but the capability itself seems to depend more on learned transformations over internal representations than on language alone refs: https://arxiv.org/abs&#
3. 3.▲
   by a-t-c-g 6mo ago

   Yes - some degree of reasoning appears to be latent in the structure of language itself. But models trained explicitly on reasoning-focused data still perform better than models trained only on general corpora.* *At least up to 300B paramet
4. 4.▲
   by a-t-c-g 6mo ago

   Benchmarks, we have internal ones testing reasoning fine-tuned v/s frontier + prompts For some use cases it can be parity performance at 1/20th the cost up to exceeds at 1/10th the cost. Trade-off is ofc narrow applicability
5. 5.▲
   by a-t-c-g 6mo ago

   The quality of custom models trained with proper reasoning datasets[0] even with small parameters (3-7B is sweet spot) is incredible now [0]: cartesien.io or Salesforce's WebscaleRL