| license: apache-2.0 | |
| base_model: HuggingFaceTB/SmolLM2-135M-Instruct | |
| tags: | |
| - code-tape | |
| - subtitle | |
| - merged | |
| - text-generation | |
| # code-tape Subtitle Postprocessor Merged v12 | |
| Merged full model for the code-tape subtitle postprocessor. This model combines SmolLM2-135M-Instruct with the v12 LoRA adapter. | |
| ## Task | |
| Given subtitle text plus code-tape context, output strict JSON for: | |
| - sparse subtitle corrections for frontend/code terminology | |
| - playback chapter jump points | |
| The prompt separates `inputSegments` (`id`, `text`) from `timeline` (`id`, `startMs`, `endMs`) to reduce schema-copy failures such as segment outputs containing timing fields. | |
| ## Validation Snapshot | |
| - 12-sample PyTorch generation probe: 12/12 valid JSON | |
| - Unknown segment references: 0 | |
| - Extra timing fields inside `segments`: 0 | |
| - Average generation time in local PyTorch probe: 2.58s | |