This PR adds the `pipeline_tag: text-generation` to the model card metadata. This will improve the model's discoverability in the Hub's model filters and ensure it is correctly categorized for its intended use case (language modeling / text generation).
- Add model card (88323b13e2e4f9d026666abfa834bfe6b6bf47ad)
- Card: point to the paper and configs for training settings; amplification is not part of COUNT (8a081fa7c976ff343ab5493b7c3e8a527a697c83)
- Card: just point to the full configs (e95cd169d505ce53eb878d564b92ef32bac306ea)
- Card: resuming with the released code (--ignore-data-fingerprints; sparse optimizer state refused) (3640e86785d2afc714d26c574e41dcbdcc689b6d)
