gpt-tokenizer 2.6.1 → 2.6.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/cjs/BytePairEncodingCore.d.ts +4 -4
- package/cjs/BytePairEncodingCore.js +24 -24
- package/cjs/BytePairEncodingCore.js.map +1 -1
- package/cjs/GptEncoding.d.ts +14 -4
- package/cjs/GptEncoding.js +32 -15
- package/cjs/GptEncoding.js.map +1 -1
- package/cjs/GptEncoding.test.js +28 -1
- package/cjs/GptEncoding.test.js.map +1 -1
- package/cjs/constants.d.ts +1 -0
- package/cjs/constants.js +5 -0
- package/cjs/constants.js.map +1 -0
- package/cjs/encoding/cl100k_base.d.ts +1 -0
- package/cjs/encoding/cl100k_base.js +1 -0
- package/cjs/encoding/cl100k_base.js.map +1 -1
- package/cjs/encoding/o200k_base.d.ts +1 -0
- package/cjs/encoding/o200k_base.js +1 -0
- package/cjs/encoding/o200k_base.js.map +1 -1
- package/cjs/encoding/p50k_base.d.ts +1 -0
- package/cjs/encoding/p50k_base.js +1 -0
- package/cjs/encoding/p50k_base.js.map +1 -1
- package/cjs/encoding/p50k_edit.d.ts +1 -0
- package/cjs/encoding/p50k_edit.js +1 -0
- package/cjs/encoding/p50k_edit.js.map +1 -1
- package/cjs/encoding/r50k_base.d.ts +1 -0
- package/cjs/encoding/r50k_base.js +1 -0
- package/cjs/encoding/r50k_base.js.map +1 -1
- package/cjs/encodingParams/Cl100KBase.d.ts +1 -1
- package/cjs/encodingParams/Cl100KBase.js +3 -3
- package/cjs/encodingParams/Cl100KBase.js.map +1 -1
- package/cjs/encodingParams/O200KBase.d.ts +1 -1
- package/cjs/encodingParams/O200KBase.js +3 -3
- package/cjs/encodingParams/O200KBase.js.map +1 -1
- package/cjs/encodingParams/P50KBase.d.ts +1 -1
- package/cjs/encodingParams/P50KBase.js +3 -3
- package/cjs/encodingParams/P50KBase.js.map +1 -1
- package/cjs/encodingParams/P50KEdit.d.ts +1 -1
- package/cjs/encodingParams/P50KEdit.js +3 -3
- package/cjs/encodingParams/P50KEdit.js.map +1 -1
- package/cjs/encodingParams/R50KBase.d.ts +1 -1
- package/cjs/encodingParams/R50KBase.js +3 -3
- package/cjs/encodingParams/R50KBase.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-0125.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-0125.js +1 -0
- package/cjs/model/gpt-3.5-turbo-0125.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-0301.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-0301.js +1 -0
- package/cjs/model/gpt-3.5-turbo-0301.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-0613.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-0613.js +1 -0
- package/cjs/model/gpt-3.5-turbo-0613.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-1106.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-1106.js +1 -0
- package/cjs/model/gpt-3.5-turbo-1106.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-16k-0613.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-16k-0613.js +1 -0
- package/cjs/model/gpt-3.5-turbo-16k-0613.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-16k.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-16k.js +1 -0
- package/cjs/model/gpt-3.5-turbo-16k.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo-finetune.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo-finetune.js +1 -0
- package/cjs/model/gpt-3.5-turbo-finetune.js.map +1 -1
- package/cjs/model/gpt-3.5-turbo.d.ts +1 -0
- package/cjs/model/gpt-3.5-turbo.js +1 -0
- package/cjs/model/gpt-3.5-turbo.js.map +1 -1
- package/cjs/model/gpt-4-0125-preview.d.ts +1 -0
- package/cjs/model/gpt-4-0125-preview.js +1 -0
- package/cjs/model/gpt-4-0125-preview.js.map +1 -1
- package/cjs/model/gpt-4-0314.d.ts +1 -0
- package/cjs/model/gpt-4-0314.js +1 -0
- package/cjs/model/gpt-4-0314.js.map +1 -1
- package/cjs/model/gpt-4-0613.d.ts +1 -0
- package/cjs/model/gpt-4-0613.js +1 -0
- package/cjs/model/gpt-4-0613.js.map +1 -1
- package/cjs/model/gpt-4-1106-preview.d.ts +1 -0
- package/cjs/model/gpt-4-1106-preview.js +1 -0
- package/cjs/model/gpt-4-1106-preview.js.map +1 -1
- package/cjs/model/gpt-4-1106-vision-preview.d.ts +1 -0
- package/cjs/model/gpt-4-1106-vision-preview.js +1 -0
- package/cjs/model/gpt-4-1106-vision-preview.js.map +1 -1
- package/cjs/model/gpt-4-32k-0314.d.ts +1 -0
- package/cjs/model/gpt-4-32k-0314.js +1 -0
- package/cjs/model/gpt-4-32k-0314.js.map +1 -1
- package/cjs/model/gpt-4-32k-0613.d.ts +1 -0
- package/cjs/model/gpt-4-32k-0613.js +1 -0
- package/cjs/model/gpt-4-32k-0613.js.map +1 -1
- package/cjs/model/gpt-4-32k.d.ts +1 -0
- package/cjs/model/gpt-4-32k.js +1 -0
- package/cjs/model/gpt-4-32k.js.map +1 -1
- package/cjs/model/gpt-4-turbo-2024-04-09.d.ts +1 -0
- package/cjs/model/gpt-4-turbo-2024-04-09.js +1 -0
- package/cjs/model/gpt-4-turbo-2024-04-09.js.map +1 -1
- package/cjs/model/gpt-4-turbo-preview.d.ts +1 -0
- package/cjs/model/gpt-4-turbo-preview.js +1 -0
- package/cjs/model/gpt-4-turbo-preview.js.map +1 -1
- package/cjs/model/gpt-4-turbo.d.ts +1 -0
- package/cjs/model/gpt-4-turbo.js +1 -0
- package/cjs/model/gpt-4-turbo.js.map +1 -1
- package/cjs/model/gpt-4-vision-preview.d.ts +1 -0
- package/cjs/model/gpt-4-vision-preview.js +1 -0
- package/cjs/model/gpt-4-vision-preview.js.map +1 -1
- package/cjs/model/gpt-4.d.ts +1 -0
- package/cjs/model/gpt-4.js +1 -0
- package/cjs/model/gpt-4.js.map +1 -1
- package/cjs/model/gpt-4o-2024-05-13.d.ts +1 -0
- package/cjs/model/gpt-4o-2024-05-13.js +1 -0
- package/cjs/model/gpt-4o-2024-05-13.js.map +1 -1
- package/cjs/model/gpt-4o-2024-08-06-finetune.d.ts +1 -0
- package/cjs/model/gpt-4o-2024-08-06-finetune.js +1 -0
- package/cjs/model/gpt-4o-2024-08-06-finetune.js.map +1 -1
- package/cjs/model/gpt-4o-2024-08-06.d.ts +1 -0
- package/cjs/model/gpt-4o-2024-08-06.js +1 -0
- package/cjs/model/gpt-4o-2024-08-06.js.map +1 -1
- package/cjs/model/gpt-4o-audio-preview-2024-10-01.d.ts +1 -0
- package/cjs/model/gpt-4o-audio-preview-2024-10-01.js +1 -0
- package/cjs/model/gpt-4o-audio-preview-2024-10-01.js.map +1 -1
- package/cjs/model/gpt-4o-audio-preview.d.ts +1 -0
- package/cjs/model/gpt-4o-audio-preview.js +1 -0
- package/cjs/model/gpt-4o-audio-preview.js.map +1 -1
- package/cjs/model/gpt-4o-mini-2024-07-18-finetune.d.ts +1 -0
- package/cjs/model/gpt-4o-mini-2024-07-18-finetune.js +1 -0
- package/cjs/model/gpt-4o-mini-2024-07-18-finetune.js.map +1 -1
- package/cjs/model/gpt-4o-mini-2024-07-18.d.ts +1 -0
- package/cjs/model/gpt-4o-mini-2024-07-18.js +1 -0
- package/cjs/model/gpt-4o-mini-2024-07-18.js.map +1 -1
- package/cjs/model/gpt-4o-mini-training-2024-07-18.d.ts +1 -0
- package/cjs/model/gpt-4o-mini-training-2024-07-18.js +1 -0
- package/cjs/model/gpt-4o-mini-training-2024-07-18.js.map +1 -1
- package/cjs/model/gpt-4o-mini-training.d.ts +1 -0
- package/cjs/model/gpt-4o-mini-training.js +1 -0
- package/cjs/model/gpt-4o-mini-training.js.map +1 -1
- package/cjs/model/gpt-4o-mini.d.ts +1 -0
- package/cjs/model/gpt-4o-mini.js +1 -0
- package/cjs/model/gpt-4o-mini.js.map +1 -1
- package/cjs/model/gpt-4o-realtime-preview-2024-10-01.d.ts +1 -0
- package/cjs/model/gpt-4o-realtime-preview-2024-10-01.js +1 -0
- package/cjs/model/gpt-4o-realtime-preview-2024-10-01.js.map +1 -1
- package/cjs/model/gpt-4o-realtime-preview.d.ts +1 -0
- package/cjs/model/gpt-4o-realtime-preview.js +1 -0
- package/cjs/model/gpt-4o-realtime-preview.js.map +1 -1
- package/cjs/model/gpt-4o.d.ts +1 -0
- package/cjs/model/gpt-4o.js +1 -0
- package/cjs/model/gpt-4o.js.map +1 -1
- package/cjs/modelParams.d.ts +1 -1
- package/dist/cl100k_base.js +1 -1
- package/dist/cl100k_base.js.map +1 -1
- package/dist/o200k_base.js +1 -1
- package/dist/o200k_base.js.map +1 -1
- package/dist/p50k_base.js +1 -1
- package/dist/p50k_base.js.map +1 -1
- package/dist/p50k_edit.js +1 -1
- package/dist/p50k_edit.js.map +1 -1
- package/dist/r50k_base.js +1 -1
- package/dist/r50k_base.js.map +1 -1
- package/esm/BytePairEncodingCore.d.ts +4 -4
- package/esm/BytePairEncodingCore.js +24 -24
- package/esm/BytePairEncodingCore.js.map +1 -1
- package/esm/GptEncoding.d.ts +14 -4
- package/esm/GptEncoding.js +31 -14
- package/esm/GptEncoding.js.map +1 -1
- package/esm/GptEncoding.test.js +28 -1
- package/esm/GptEncoding.test.js.map +1 -1
- package/esm/constants.d.ts +1 -0
- package/esm/constants.js +2 -0
- package/esm/constants.js.map +1 -0
- package/esm/encoding/cl100k_base.d.ts +1 -0
- package/esm/encoding/cl100k_base.js +1 -0
- package/esm/encoding/cl100k_base.js.map +1 -1
- package/esm/encoding/o200k_base.d.ts +1 -0
- package/esm/encoding/o200k_base.js +1 -0
- package/esm/encoding/o200k_base.js.map +1 -1
- package/esm/encoding/p50k_base.d.ts +1 -0
- package/esm/encoding/p50k_base.js +1 -0
- package/esm/encoding/p50k_base.js.map +1 -1
- package/esm/encoding/p50k_edit.d.ts +1 -0
- package/esm/encoding/p50k_edit.js +1 -0
- package/esm/encoding/p50k_edit.js.map +1 -1
- package/esm/encoding/r50k_base.d.ts +1 -0
- package/esm/encoding/r50k_base.js +1 -0
- package/esm/encoding/r50k_base.js.map +1 -1
- package/esm/encodingParams/Cl100KBase.d.ts +1 -1
- package/esm/encodingParams/Cl100KBase.js +3 -3
- package/esm/encodingParams/Cl100KBase.js.map +1 -1
- package/esm/encodingParams/O200KBase.d.ts +1 -1
- package/esm/encodingParams/O200KBase.js +3 -3
- package/esm/encodingParams/O200KBase.js.map +1 -1
- package/esm/encodingParams/P50KBase.d.ts +1 -1
- package/esm/encodingParams/P50KBase.js +3 -3
- package/esm/encodingParams/P50KBase.js.map +1 -1
- package/esm/encodingParams/P50KEdit.d.ts +1 -1
- package/esm/encodingParams/P50KEdit.js +3 -3
- package/esm/encodingParams/P50KEdit.js.map +1 -1
- package/esm/encodingParams/R50KBase.d.ts +1 -1
- package/esm/encodingParams/R50KBase.js +3 -3
- package/esm/encodingParams/R50KBase.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-0125.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-0125.js +1 -0
- package/esm/model/gpt-3.5-turbo-0125.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-0301.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-0301.js +1 -0
- package/esm/model/gpt-3.5-turbo-0301.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-0613.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-0613.js +1 -0
- package/esm/model/gpt-3.5-turbo-0613.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-1106.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-1106.js +1 -0
- package/esm/model/gpt-3.5-turbo-1106.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-16k-0613.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-16k-0613.js +1 -0
- package/esm/model/gpt-3.5-turbo-16k-0613.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-16k.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-16k.js +1 -0
- package/esm/model/gpt-3.5-turbo-16k.js.map +1 -1
- package/esm/model/gpt-3.5-turbo-finetune.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo-finetune.js +1 -0
- package/esm/model/gpt-3.5-turbo-finetune.js.map +1 -1
- package/esm/model/gpt-3.5-turbo.d.ts +1 -0
- package/esm/model/gpt-3.5-turbo.js +1 -0
- package/esm/model/gpt-3.5-turbo.js.map +1 -1
- package/esm/model/gpt-4-0125-preview.d.ts +1 -0
- package/esm/model/gpt-4-0125-preview.js +1 -0
- package/esm/model/gpt-4-0125-preview.js.map +1 -1
- package/esm/model/gpt-4-0314.d.ts +1 -0
- package/esm/model/gpt-4-0314.js +1 -0
- package/esm/model/gpt-4-0314.js.map +1 -1
- package/esm/model/gpt-4-0613.d.ts +1 -0
- package/esm/model/gpt-4-0613.js +1 -0
- package/esm/model/gpt-4-0613.js.map +1 -1
- package/esm/model/gpt-4-1106-preview.d.ts +1 -0
- package/esm/model/gpt-4-1106-preview.js +1 -0
- package/esm/model/gpt-4-1106-preview.js.map +1 -1
- package/esm/model/gpt-4-1106-vision-preview.d.ts +1 -0
- package/esm/model/gpt-4-1106-vision-preview.js +1 -0
- package/esm/model/gpt-4-1106-vision-preview.js.map +1 -1
- package/esm/model/gpt-4-32k-0314.d.ts +1 -0
- package/esm/model/gpt-4-32k-0314.js +1 -0
- package/esm/model/gpt-4-32k-0314.js.map +1 -1
- package/esm/model/gpt-4-32k-0613.d.ts +1 -0
- package/esm/model/gpt-4-32k-0613.js +1 -0
- package/esm/model/gpt-4-32k-0613.js.map +1 -1
- package/esm/model/gpt-4-32k.d.ts +1 -0
- package/esm/model/gpt-4-32k.js +1 -0
- package/esm/model/gpt-4-32k.js.map +1 -1
- package/esm/model/gpt-4-turbo-2024-04-09.d.ts +1 -0
- package/esm/model/gpt-4-turbo-2024-04-09.js +1 -0
- package/esm/model/gpt-4-turbo-2024-04-09.js.map +1 -1
- package/esm/model/gpt-4-turbo-preview.d.ts +1 -0
- package/esm/model/gpt-4-turbo-preview.js +1 -0
- package/esm/model/gpt-4-turbo-preview.js.map +1 -1
- package/esm/model/gpt-4-turbo.d.ts +1 -0
- package/esm/model/gpt-4-turbo.js +1 -0
- package/esm/model/gpt-4-turbo.js.map +1 -1
- package/esm/model/gpt-4-vision-preview.d.ts +1 -0
- package/esm/model/gpt-4-vision-preview.js +1 -0
- package/esm/model/gpt-4-vision-preview.js.map +1 -1
- package/esm/model/gpt-4.d.ts +1 -0
- package/esm/model/gpt-4.js +1 -0
- package/esm/model/gpt-4.js.map +1 -1
- package/esm/model/gpt-4o-2024-05-13.d.ts +1 -0
- package/esm/model/gpt-4o-2024-05-13.js +1 -0
- package/esm/model/gpt-4o-2024-05-13.js.map +1 -1
- package/esm/model/gpt-4o-2024-08-06-finetune.d.ts +1 -0
- package/esm/model/gpt-4o-2024-08-06-finetune.js +1 -0
- package/esm/model/gpt-4o-2024-08-06-finetune.js.map +1 -1
- package/esm/model/gpt-4o-2024-08-06.d.ts +1 -0
- package/esm/model/gpt-4o-2024-08-06.js +1 -0
- package/esm/model/gpt-4o-2024-08-06.js.map +1 -1
- package/esm/model/gpt-4o-audio-preview-2024-10-01.d.ts +1 -0
- package/esm/model/gpt-4o-audio-preview-2024-10-01.js +1 -0
- package/esm/model/gpt-4o-audio-preview-2024-10-01.js.map +1 -1
- package/esm/model/gpt-4o-audio-preview.d.ts +1 -0
- package/esm/model/gpt-4o-audio-preview.js +1 -0
- package/esm/model/gpt-4o-audio-preview.js.map +1 -1
- package/esm/model/gpt-4o-mini-2024-07-18-finetune.d.ts +1 -0
- package/esm/model/gpt-4o-mini-2024-07-18-finetune.js +1 -0
- package/esm/model/gpt-4o-mini-2024-07-18-finetune.js.map +1 -1
- package/esm/model/gpt-4o-mini-2024-07-18.d.ts +1 -0
- package/esm/model/gpt-4o-mini-2024-07-18.js +1 -0
- package/esm/model/gpt-4o-mini-2024-07-18.js.map +1 -1
- package/esm/model/gpt-4o-mini-training-2024-07-18.d.ts +1 -0
- package/esm/model/gpt-4o-mini-training-2024-07-18.js +1 -0
- package/esm/model/gpt-4o-mini-training-2024-07-18.js.map +1 -1
- package/esm/model/gpt-4o-mini-training.d.ts +1 -0
- package/esm/model/gpt-4o-mini-training.js +1 -0
- package/esm/model/gpt-4o-mini-training.js.map +1 -1
- package/esm/model/gpt-4o-mini.d.ts +1 -0
- package/esm/model/gpt-4o-mini.js +1 -0
- package/esm/model/gpt-4o-mini.js.map +1 -1
- package/esm/model/gpt-4o-realtime-preview-2024-10-01.d.ts +1 -0
- package/esm/model/gpt-4o-realtime-preview-2024-10-01.js +1 -0
- package/esm/model/gpt-4o-realtime-preview-2024-10-01.js.map +1 -1
- package/esm/model/gpt-4o-realtime-preview.d.ts +1 -0
- package/esm/model/gpt-4o-realtime-preview.js +1 -0
- package/esm/model/gpt-4o-realtime-preview.js.map +1 -1
- package/esm/model/gpt-4o.d.ts +1 -0
- package/esm/model/gpt-4o.js +1 -0
- package/esm/model/gpt-4o.js.map +1 -1
- package/esm/modelParams.d.ts +1 -1
- package/package.json +1 -1
- package/src/BytePairEncodingCore.ts +28 -30
- package/src/GptEncoding.test.ts +36 -1
- package/src/GptEncoding.ts +49 -18
- package/src/constants.ts +1 -0
- package/src/encoding/cl100k_base.ts +1 -0
- package/src/encoding/o200k_base.ts +1 -0
- package/src/encoding/p50k_base.ts +1 -0
- package/src/encoding/p50k_edit.ts +1 -0
- package/src/encoding/r50k_base.ts +1 -0
- package/src/encodingParams/Cl100KBase.ts +3 -3
- package/src/encodingParams/O200KBase.ts +3 -3
- package/src/encodingParams/P50KBase.ts +3 -3
- package/src/encodingParams/P50KEdit.ts +3 -3
- package/src/encodingParams/R50KBase.ts +3 -3
- package/src/model/gpt-3.5-turbo-0125.ts +1 -0
- package/src/model/gpt-3.5-turbo-0301.ts +1 -0
- package/src/model/gpt-3.5-turbo-0613.ts +1 -0
- package/src/model/gpt-3.5-turbo-1106.ts +1 -0
- package/src/model/gpt-3.5-turbo-16k-0613.ts +1 -0
- package/src/model/gpt-3.5-turbo-16k.ts +1 -0
- package/src/model/gpt-3.5-turbo-finetune.ts +1 -0
- package/src/model/gpt-3.5-turbo.ts +1 -0
- package/src/model/gpt-4-0125-preview.ts +1 -0
- package/src/model/gpt-4-0314.ts +1 -0
- package/src/model/gpt-4-0613.ts +1 -0
- package/src/model/gpt-4-1106-preview.ts +1 -0
- package/src/model/gpt-4-1106-vision-preview.ts +1 -0
- package/src/model/gpt-4-32k-0314.ts +1 -0
- package/src/model/gpt-4-32k-0613.ts +1 -0
- package/src/model/gpt-4-32k.ts +1 -0
- package/src/model/gpt-4-turbo-2024-04-09.ts +1 -0
- package/src/model/gpt-4-turbo-preview.ts +1 -0
- package/src/model/gpt-4-turbo.ts +1 -0
- package/src/model/gpt-4-vision-preview.ts +1 -0
- package/src/model/gpt-4.ts +1 -0
- package/src/model/gpt-4o-2024-05-13.ts +1 -0
- package/src/model/gpt-4o-2024-08-06-finetune.ts +1 -0
- package/src/model/gpt-4o-2024-08-06.ts +1 -0
- package/src/model/gpt-4o-audio-preview-2024-10-01.ts +1 -0
- package/src/model/gpt-4o-audio-preview.ts +1 -0
- package/src/model/gpt-4o-mini-2024-07-18-finetune.ts +1 -0
- package/src/model/gpt-4o-mini-2024-07-18.ts +1 -0
- package/src/model/gpt-4o-mini-training-2024-07-18.ts +1 -0
- package/src/model/gpt-4o-mini-training.ts +1 -0
- package/src/model/gpt-4o-mini.ts +1 -0
- package/src/model/gpt-4o-realtime-preview-2024-10-01.ts +1 -0
- package/src/model/gpt-4o-realtime-preview.ts +1 -0
- package/src/model/gpt-4o.ts +1 -0
- package/src/modelParams.ts +1 -1
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
export type RawBytePairRanks = readonly (string | readonly number[])[];
|
|
2
2
|
export interface BytePairEncodingConfig {
|
|
3
|
-
|
|
4
|
-
|
|
3
|
+
bytePairRankDecoder: RawBytePairRanks;
|
|
4
|
+
specialTokensEncoder?: Map<string, number>;
|
|
5
5
|
tokenSplitRegex: RegExp;
|
|
6
6
|
}
|
|
7
7
|
export declare const decoder: TextDecoder;
|
|
@@ -26,7 +26,7 @@ export declare class BytePairEncodingCore {
|
|
|
26
26
|
private specialTokensDecoder;
|
|
27
27
|
private specialTokenPatternRegex;
|
|
28
28
|
private textEncoder;
|
|
29
|
-
constructor({
|
|
29
|
+
constructor({ bytePairRankDecoder, specialTokensEncoder, tokenSplitRegex, }: BytePairEncodingConfig);
|
|
30
30
|
encodeNativeGenerator(text: string, allowedSpecial?: Set<string>): Generator<number[], number, undefined>;
|
|
31
31
|
encodeNative(text: string, allowedSpecial?: Set<string>): number[];
|
|
32
32
|
decodeNativeGenerator(tokens: Iterable<number>): Generator<Uint8Array | string, void, void>;
|
|
@@ -37,7 +37,7 @@ export declare class BytePairEncodingCore {
|
|
|
37
37
|
private getBpeRankFromBytes;
|
|
38
38
|
private getBpeRankFromBytesOrThrow;
|
|
39
39
|
private binarySearch;
|
|
40
|
-
private
|
|
40
|
+
private findNextSpecialToken;
|
|
41
41
|
private tryDecodeToken;
|
|
42
42
|
private bytePairEncode;
|
|
43
43
|
private bytePairMerge;
|
|
@@ -27,14 +27,14 @@ class BytePairEncodingCore {
|
|
|
27
27
|
specialTokensDecoder;
|
|
28
28
|
specialTokenPatternRegex;
|
|
29
29
|
textEncoder = new TextEncoder();
|
|
30
|
-
constructor({
|
|
31
|
-
this.bytePairRankDecoder =
|
|
30
|
+
constructor({ bytePairRankDecoder, specialTokensEncoder, tokenSplitRegex, }) {
|
|
31
|
+
this.bytePairRankDecoder = bytePairRankDecoder;
|
|
32
32
|
this.bytePairStringRankEncoder = new Map();
|
|
33
33
|
// size without array holes (which may be present in the encoder)
|
|
34
|
-
this.mergeableBytePairRankCount = Object.keys(
|
|
34
|
+
this.mergeableBytePairRankCount = Object.keys(bytePairRankDecoder).length;
|
|
35
35
|
const binaryLookup = [];
|
|
36
36
|
// forEach skips array holes:
|
|
37
|
-
|
|
37
|
+
bytePairRankDecoder.forEach((value, rank) => {
|
|
38
38
|
if (typeof value === 'string') {
|
|
39
39
|
this.bytePairStringRankEncoder.set(value, rank);
|
|
40
40
|
return;
|
|
@@ -44,9 +44,10 @@ class BytePairEncodingCore {
|
|
|
44
44
|
this.bytePairNonUtfRankDecoder.set(rank, byteArray);
|
|
45
45
|
});
|
|
46
46
|
this.bytePairNonUtfSortedEncoder = binaryLookup.sort((a, b) => (0, utfUtil_js_1.compareUint8Arrays)(a[0], b[0]));
|
|
47
|
-
this.specialTokensEncoder =
|
|
48
|
-
|
|
49
|
-
|
|
47
|
+
this.specialTokensEncoder =
|
|
48
|
+
specialTokensEncoder ?? new Map();
|
|
49
|
+
this.specialTokensDecoder = specialTokensEncoder
|
|
50
|
+
? new Map([...specialTokensEncoder].map(([key, value]) => [value, key]))
|
|
50
51
|
: new Map();
|
|
51
52
|
this.tokenSplitRegex = tokenSplitRegex;
|
|
52
53
|
const escapedSpecialTokens = [...this.specialTokensEncoder.keys()].map(util_js_1.escapeRegExp);
|
|
@@ -62,12 +63,11 @@ class BytePairEncodingCore {
|
|
|
62
63
|
let startIndex = 0;
|
|
63
64
|
let lastTokenLength = 0;
|
|
64
65
|
while (true) {
|
|
65
|
-
const
|
|
66
|
-
const
|
|
67
|
-
|
|
68
|
-
|
|
69
|
-
const
|
|
70
|
-
for (const [match] of textSegment.matchAll(this.tokenSplitRegex)) {
|
|
66
|
+
const nextSpecialMatch = this.findNextSpecialToken(text, allowedSpecial, startIndex);
|
|
67
|
+
const nextSpecialStartIndex = nextSpecialMatch?.[0];
|
|
68
|
+
const endIndex = nextSpecialStartIndex ?? text.length;
|
|
69
|
+
const textBeforeSpecial = text.slice(startIndex, endIndex);
|
|
70
|
+
for (const [match] of textBeforeSpecial.matchAll(this.tokenSplitRegex)) {
|
|
71
71
|
const token = this.getBpeRankFromString(match);
|
|
72
72
|
if (token !== undefined) {
|
|
73
73
|
lastTokenLength = 1;
|
|
@@ -79,14 +79,14 @@ class BytePairEncodingCore {
|
|
|
79
79
|
yield tokens;
|
|
80
80
|
}
|
|
81
81
|
if (nextSpecialStartIndex !== undefined) {
|
|
82
|
-
const specialToken =
|
|
82
|
+
const specialToken = nextSpecialMatch[1];
|
|
83
83
|
const specialTokenValue = this.specialTokensEncoder.get(specialToken);
|
|
84
84
|
if (specialTokenValue === undefined) {
|
|
85
85
|
throw new Error(`Special token "${specialToken}" is not in the special token encoder.`);
|
|
86
86
|
}
|
|
87
87
|
yield [specialTokenValue];
|
|
88
88
|
startIndex = nextSpecialStartIndex + specialToken.length;
|
|
89
|
-
lastTokenLength =
|
|
89
|
+
lastTokenLength = 1;
|
|
90
90
|
}
|
|
91
91
|
else {
|
|
92
92
|
break;
|
|
@@ -99,12 +99,11 @@ class BytePairEncodingCore {
|
|
|
99
99
|
const tokensArray = []; // Flat list to collect the tokens
|
|
100
100
|
// eslint-disable-next-line no-constant-condition
|
|
101
101
|
while (true) {
|
|
102
|
-
const
|
|
103
|
-
const
|
|
104
|
-
|
|
105
|
-
|
|
106
|
-
const
|
|
107
|
-
for (const [match] of textSegment.matchAll(this.tokenSplitRegex)) {
|
|
102
|
+
const nextSpecialMatch = this.findNextSpecialToken(text, allowedSpecial, startIndex);
|
|
103
|
+
const nextSpecialStartIndex = nextSpecialMatch?.[0];
|
|
104
|
+
const endIndex = nextSpecialStartIndex ?? text.length;
|
|
105
|
+
const textBeforeSpecial = text.slice(startIndex, endIndex);
|
|
106
|
+
for (const [match] of textBeforeSpecial.matchAll(this.tokenSplitRegex)) {
|
|
108
107
|
const token = this.getBpeRankFromString(match);
|
|
109
108
|
if (token !== undefined) {
|
|
110
109
|
tokensArray.push(token);
|
|
@@ -114,7 +113,7 @@ class BytePairEncodingCore {
|
|
|
114
113
|
tokensArray.push(...tokens);
|
|
115
114
|
}
|
|
116
115
|
if (nextSpecialStartIndex !== undefined) {
|
|
117
|
-
const specialToken =
|
|
116
|
+
const specialToken = nextSpecialMatch[1];
|
|
118
117
|
const specialTokenValue = this.specialTokensEncoder.get(specialToken);
|
|
119
118
|
if (specialTokenValue === undefined) {
|
|
120
119
|
throw new Error(`Special token "${specialToken}" is not in the special token encoder.`);
|
|
@@ -230,7 +229,7 @@ class BytePairEncodingCore {
|
|
|
230
229
|
}
|
|
231
230
|
return -1;
|
|
232
231
|
}
|
|
233
|
-
|
|
232
|
+
findNextSpecialToken(text, allowedSpecial, startIndex) {
|
|
234
233
|
let searchIndex = startIndex;
|
|
235
234
|
// eslint-disable-next-line no-constant-condition
|
|
236
235
|
while (true) {
|
|
@@ -240,7 +239,8 @@ class BytePairEncodingCore {
|
|
|
240
239
|
}
|
|
241
240
|
const specialToken = nextSpecialMatch[0];
|
|
242
241
|
if (allowedSpecial?.has(specialToken)) {
|
|
243
|
-
|
|
242
|
+
const specialTokenStartIndex = nextSpecialMatch.index + searchIndex;
|
|
243
|
+
return [specialTokenStartIndex, specialToken];
|
|
244
244
|
}
|
|
245
245
|
searchIndex = nextSpecialMatch.index + searchIndex + 1;
|
|
246
246
|
}
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"BytePairEncodingCore.js","sourceRoot":"","sources":["../src/BytePairEncodingCore.ts"],"names":[],"mappings":";AAAA,gCAAgC;;;AAEhC,6CAA8E;AAC9E,uCAAwC;AAUxC,MAAM,WAAW,GAAG,IAAI,UAAU,CAAC,CAAC,CAAC,CAAA;AAExB,QAAA,OAAO,GAAG,IAAI,WAAW,CAAC,MAAM,CAAC,CAAA;AAE9C,MAAa,oBAAoB;IACtB,0BAA0B,CAAQ;IAC3C;;;;OAIG;IACK,mBAAmB,CAAkB;IACrC,yBAAyB,GAAG,IAAI,GAAG,EAAsB,CAAA;IACzD,2BAA2B,CAAiC;IACpE;;;;OAIG;IACK,yBAAyB,CAAqB;IAC9C,eAAe,CAAQ;IACvB,oBAAoB,CAAqB;IACzC,oBAAoB,CAAqB;IACzC,wBAAwB,CAAQ;IAChC,WAAW,GAAG,IAAI,WAAW,EAAE,CAAA;IAEvC,YAAY,EACV,sBAAsB,EAAE,eAAe,EACvC,mBAAmB,EAAE,mBAAmB,EACxC,eAAe,GACQ;QACvB,IAAI,CAAC,mBAAmB,GAAG,eAAe,CAAA;QAC1C,IAAI,CAAC,yBAAyB,GAAG,IAAI,GAAG,EAAkB,CAAA;QAE1D,iEAAiE;QACjE,IAAI,CAAC,0BAA0B,GAAG,MAAM,CAAC,IAAI,CAAC,eAAe,CAAC,CAAC,MAAM,CAAA;QACrE,MAAM,YAAY,GAA2B,EAAE,CAAA;QAC/C,6BAA6B;QAC7B,eAAe,CAAC,OAAO,CAAC,CAAC,KAAK,EAAE,IAAI,EAAE,EAAE;YACtC,IAAI,OAAO,KAAK,KAAK,QAAQ,EAAE,CAAC;gBAC9B,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,KAAK,EAAE,IAAI,CAAC,CAAA;gBAC/C,OAAM;YACR,CAAC;YACD,MAAM,SAAS,GAAG,IAAI,UAAU,CAAC,KAAK,CAAC,CAAA;YACvC,YAAY,CAAC,IAAI,CAAC,CAAC,SAAS,EAAE,IAAI,CAAC,CAAC,CAAA;YACpC,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,IAAI,EAAE,SAAS,CAAC,CAAA;QACrD,CAAC,CAAC,CAAA;QACF,IAAI,CAAC,2BAA2B,GAAG,YAAY,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAC5D,IAAA,+BAAkB,EAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAC/B,CAAA;QACD,IAAI,CAAC,oBAAoB,GAAG,mBAAmB,IAAI,IAAI,GAAG,EAAkB,CAAA;QAC5E,IAAI,CAAC,oBAAoB,GAAG,mBAAmB;YAC7C,CAAC,CAAC,IAAI,GAAG,CAAC,CAAC,GAAG,mBAAmB,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,EAAE,KAAK,CAAC,EAAE,EAAE,CAAC,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,CAAC;YACvE,CAAC,CAAC,IAAI,GAAG,EAAkB,CAAA;QAC7B,IAAI,CAAC,eAAe,GAAG,eAAe,CAAA;QAEtC,MAAM,oBAAoB,GAAG,CAAC,GAAG,IAAI,CAAC,oBAAoB,CAAC,IAAI,EAAE,CAAC,CAAC,GAAG,CACpE,sBAAY,CACb,CAAA;QACD,MAAM,qBAAqB,GAAG,oBAAoB,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QAC5D,IAAI,CAAC;YACH,IAAI,CAAC,wBAAwB,GAAG,IAAI,MAAM,CAAC,qBAAqB,CAAC,CAAA;QACnE,CAAC;QAAC,MAAM,CAAC;YACP,MAAM,IAAI,KAAK,CAAC,qCAAqC,CAAC,CAAA;QACxD,CAAC;IACH,CAAC;IAED,CAAC,qBAAqB,CACpB,IAAY,EACZ,cAA4B;QAE5B,IAAI,UAAU,GAAG,CAAC,CAAA;QAClB,IAAI,eAAe,GAAG,CAAC,CAAA;QAEvB,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,qBAAqB,GAAG,IAAI,CAAC,yBAAyB,CAC1D,IAAI,EACJ,cAAc,EACd,UAAU,CACX,CAAA;YAED,MAAM,QAAQ,GACZ,qBAAqB,KAAK,SAAS;gBACjC,CAAC,CAAC,qBAAqB;gBACvB,CAAC,CAAC,IAAI,CAAC,MAAM,CAAA;YAEjB,MAAM,WAAW,GAAG,IAAI,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,GAAG,UAAU,CAAC,CAAA;YAEjE,KAAK,MAAM,CAAC,KAAK,CAAC,IAAI,WAAW,CAAC,QAAQ,CAAC,IAAI,CAAC,eAAe,CAAC,EAAE,CAAC;gBACjE,MAAM,KAAK,GAAG,IAAI,CAAC,oBAAoB,CAAC,KAAK,CAAC,CAAA;gBAC9C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;oBACxB,eAAe,GAAG,CAAC,CAAA;oBACnB,MAAM,CAAC,KAAK,CAAC,CAAA;oBAEb,SAAQ;gBACV,CAAC;gBAED,MAAM,MAAM,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;gBACzC,eAAe,GAAG,MAAM,CAAC,MAAM,CAAA;gBAC/B,MAAM,MAAM,CAAA;YACd,CAAC;YAED,IAAI,qBAAqB,KAAK,SAAS,EAAE,CAAC;gBACxC,MAAM,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,qBAAqB,CAAC,CAAC,CAAA;gBACnE,MAAM,iBAAiB,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,YAAY,CAAC,CAAA;gBACrE,IAAI,iBAAiB,KAAK,SAAS,EAAE,CAAC;oBACpC,MAAM,IAAI,KAAK,CACb,kBAAkB,YAAY,wCAAwC,CACvE,CAAA;gBACH,CAAC;gBACD,MAAM,CAAC,iBAAiB,CAAC,CAAA;gBACzB,UAAU,GAAG,qBAAqB,GAAG,YAAY,CAAC,MAAM,CAAA;gBACxD,eAAe,GAAG,CAAC,CAAA;YACrB,CAAC;iBAAM,CAAC;gBACN,MAAK;YACP,CAAC;QACH,CAAC;QAED,OAAO,eAAe,CAAA;IACxB,CAAC;IAED,YAAY,CAAC,IAAY,EAAE,cAA4B;QACrD,IAAI,UAAU,GAAG,CAAC,CAAA;QAClB,MAAM,WAAW,GAAa,EAAE,CAAA,CAAC,kCAAkC;QAEnE,iDAAiD;QACjD,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,qBAAqB,GAAG,IAAI,CAAC,yBAAyB,CAC1D,IAAI,EACJ,cAAc,EACd,UAAU,CACX,CAAA;YAED,MAAM,QAAQ,GACZ,qBAAqB,KAAK,SAAS;gBACjC,CAAC,CAAC,qBAAqB;gBACvB,CAAC,CAAC,IAAI,CAAC,MAAM,CAAA;YAEjB,MAAM,WAAW,GAAG,IAAI,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,GAAG,UAAU,CAAC,CAAA;YAEjE,KAAK,MAAM,CAAC,KAAK,CAAC,IAAI,WAAW,CAAC,QAAQ,CAAC,IAAI,CAAC,eAAe,CAAC,EAAE,CAAC;gBACjE,MAAM,KAAK,GAAG,IAAI,CAAC,oBAAoB,CAAC,KAAK,CAAC,CAAA;gBAC9C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;oBACxB,WAAW,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;oBAEvB,SAAQ;gBACV,CAAC;gBAED,MAAM,MAAM,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;gBACzC,WAAW,CAAC,IAAI,CAAC,GAAG,MAAM,CAAC,CAAA;YAC7B,CAAC;YAED,IAAI,qBAAqB,KAAK,SAAS,EAAE,CAAC;gBACxC,MAAM,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,qBAAqB,CAAC,CAAC,CAAA;gBACnE,MAAM,iBAAiB,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,YAAY,CAAC,CAAA;gBACrE,IAAI,iBAAiB,KAAK,SAAS,EAAE,CAAC;oBACpC,MAAM,IAAI,KAAK,CACb,kBAAkB,YAAY,wCAAwC,CACvE,CAAA;gBACH,CAAC;gBACD,WAAW,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAA;gBACnC,UAAU,GAAG,qBAAqB,GAAG,YAAY,CAAC,MAAM,CAAA;YAC1D,CAAC;iBAAM,CAAC;gBACN,MAAK;YACP,CAAC;QACH,CAAC;QAED,OAAO,WAAW,CAAA;IACpB,CAAC;IAED,CAAC,qBAAqB,CACpB,MAAwB;QAExB,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YAC3B,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;YAC7C,IAAI,UAAU,EAAE,CAAC;gBACf,MAAM,UAAU,CAAA;YAClB,CAAC;QACH,CAAC;IACH,CAAC;IAED,YAAY,CAAC,MAAwB;QACnC,IAAI,OAAO,GAAG,EAAE,CAAA;QAChB,IAAI,SAAS,GAAG,WAAW,CAAA;QAE3B,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YAC3B,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;YAC7C,IAAI,UAAU,KAAK,SAAS,EAAE,CAAC;gBAC7B,MAAM,IAAI,KAAK,CAAC,SAAS,KAAK,mCAAmC,CAAC,CAAA;YACpE,CAAC;YACD,IAAI,OAAO,UAAU,KAAK,QAAQ,EAAE,CAAC;gBACnC,IAAI,SAAS,KAAK,WAAW,EAAE,CAAC;oBAC9B,OAAO,IAAI,eAAO,CAAC,MAAM,CAAC,SAAS,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;oBACtD,SAAS,GAAG,WAAW,CAAA;gBACzB,CAAC;gBACD,OAAO,IAAI,UAAU,CAAA;YACvB,CAAC;iBAAM,CAAC;gBACN,MAAM,SAAS,GAAG,IAAI,UAAU,CAAC,SAAS,CAAC,MAAM,GAAG,UAAU,CAAC,MAAM,CAAC,CAAA;gBACtE,SAAS,CAAC,GAAG,CAAC,SAAS,CAAC,CAAA;gBACxB,SAAS,CAAC,GAAG,CAAC,UAAU,EAAE,SAAS,CAAC,MAAM,CAAC,CAAA;gBAC3C,SAAS,GAAG,SAAS,CAAA;YACvB,CAAC;QACH,CAAC;QAED,IAAI,SAAS,KAAK,WAAW,EAAE,CAAC;YAC9B,OAAO,IAAI,eAAO,CAAC,MAAM,CAAC,SAAS,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;QACxD,CAAC;QACD,OAAO,OAAO,CAAA;IAChB,CAAC;IAED,KAAK,CAAC,CAAC,yBAAyB,CAC9B,MAA6B;QAE7B,IAAI,KAAK,EAAE,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YACjC,MAAM,kBAAkB,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;YACrD,IAAI,kBAAkB,EAAE,CAAC;gBACvB,MAAM,kBAAkB,CAAA;YAC1B,CAAC;QACH,CAAC;IACH,CAAC;IAEO,oBAAoB,CAAC,GAAW;QACtC,OAAO,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,GAAG,CAAC,CAAA;IAChD,CAAC;IAEO,2BAA2B,CAAC,GAAW;QAC7C,MAAM,KAAK,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,CAAA;QAC5C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,KAAK,CACb,wDAAwD,GAAG,EAAE,CAC9D,CAAA;QACH,CAAC;QACD,OAAO,KAAK,CAAA;IACd,CAAC;IAEO,mBAAmB,CAAC,GAAe;QACzC,MAAM,WAAW,GAAG,IAAA,+BAAkB,EAAC,GAAG,CAAC,CAAA;QAE3C,IAAI,WAAW,KAAK,SAAS,EAAE,CAAC;YAC9B,OAAO,IAAI,CAAC,oBAAoB,CAAC,WAAW,CAAC,CAAA;QAC/C,CAAC;QAED,2CAA2C;QAC3C,MAAM,KAAK,GAAG,IAAI,CAAC,YAAY,CAAC,GAAG,CAAC,CAAA;QACpC,IAAI,KAAK,KAAK,CAAC,CAAC,EAAE,CAAC;YACjB,OAAO,IAAI,CAAC,2BAA2B,CAAC,KAAK,CAAE,CAAC,CAAC,CAAC,CAAA;QACpD,CAAC;QAED,OAAO,SAAS,CAAA;IAClB,CAAC;IAEO,0BAA0B,CAAC,GAAe;QAChD,MAAM,KAAK,GAAG,IAAI,CAAC,mBAAmB,CAAC,GAAG,CAAC,CAAA;QAC3C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,KAAK,CACb,wDAAwD,GAAG,CAAC,QAAQ,EAAE,EAAE,CACzE,CAAA;QACH,CAAC;QACD,OAAO,KAAK,CAAA;IACd,CAAC;IAED,mCAAmC;IAC3B,YAAY,CAAC,GAAe;QAClC,IAAI,GAAG,GAAG,CAAC,CAAA;QACX,IAAI,IAAI,GAAG,IAAI,CAAC,2BAA2B,CAAC,MAAM,GAAG,CAAC,CAAA;QAEtD,OAAO,GAAG,IAAI,IAAI,EAAE,CAAC;YACnB,sCAAsC;YACtC,MAAM,GAAG,GAAG,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9B,MAAM,MAAM,GAAG,IAAI,CAAC,2BAA2B,CAAC,GAAG,CAAE,CAAC,CAAC,CAAC,CAAA;YACxD,IAAI,GAAG,GAAG,CAAC,CAAA;YACX,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,GAAG,CAAC,MAAM,CAAC,CAAA;YACrD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,EAAE,CAAC,EAAE,EAAE,CAAC;gBACnC,GAAG,GAAG,MAAM,CAAC,CAAC,CAAE,GAAG,GAAG,CAAC,CAAC,CAAE,CAAA;gBAC1B,IAAI,GAAG,KAAK,CAAC;oBAAE,MAAK;YACtB,CAAC;YACD,IAAI,GAAG,KAAK,CAAC,EAAE,CAAC;gBACd,GAAG,GAAG,MAAM,CAAC,MAAM,GAAG,GAAG,CAAC,MAAM,CAAA;YAClC,CAAC;YACD,IAAI,GAAG,KAAK,CAAC,EAAE,CAAC;gBACd,OAAO,GAAG,CAAA;YACZ,CAAC;YACD,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,GAAG,GAAG,GAAG,GAAG,CAAC,CAAA;YACf,CAAC;iBAAM,CAAC;gBACN,IAAI,GAAG,GAAG,GAAG,CAAC,CAAA;YAChB,CAAC;QACH,CAAC;QACD,OAAO,CAAC,CAAC,CAAA;IACX,CAAC;IAEO,yBAAyB,CAC/B,IAAY,EACZ,cAAuC,EACvC,UAAkB;QAElB,IAAI,WAAW,GAAG,UAAU,CAAA;QAE5B,iDAAiD;QACjD,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,gBAAgB,GAAG,IAAI,CAAC,wBAAwB,CAAC,IAAI,CACzD,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC,CACrC,CAAA;YAED,IAAI,CAAC,gBAAgB,EAAE,CAAC;gBACtB,OAAO,SAAS,CAAA;YAClB,CAAC;YAED,MAAM,YAAY,GAAG,gBAAgB,CAAC,CAAC,CAAC,CAAA;YAExC,IAAI,cAAc,EAAE,GAAG,CAAC,YAAY,CAAC,EAAE,CAAC;gBACtC,OAAO,gBAAgB,CAAC,KAAK,GAAG,WAAW,CAAA;YAC7C,CAAC;YAED,WAAW,GAAG,gBAAgB,CAAC,KAAK,GAAG,WAAW,GAAG,CAAC,CAAA;QACxD,CAAC;IACH,CAAC;IAEO,cAAc,CAAC,SAAiB;QACtC,MAAM,KAAK,GAAG,IAAI,CAAC,mBAAmB,CAAC,SAAS,CAAC,CAAA;QACjD,IAAI,OAAO,KAAK,KAAK,QAAQ,EAAE,CAAC;YAC9B,OAAO,KAAK,CAAA;QACd,CAAC;QACD,IAAI,OAAO,KAAK,KAAK,QAAQ,EAAE,CAAC;YAC9B,MAAM,UAAU,GAAG,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,SAAS,CAAC,CAAA;YAChE,IAAI,UAAU,EAAE,CAAC;gBACf,OAAO,UAAU,CAAA;YACnB,CAAC;QACH,CAAC;QACD,OAAO,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,SAAS,CAAC,CAAA;IACjD,CAAC;IAEO,cAAc,CAAC,KAAa;QAClC,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,IAAI,IAAA,oBAAO,EAAC,KAAK,CAAC,WAAW,CAAC,CAAC,CAAE,CAAC,EAAE,CAAC;YACzD,OAAO,CAAC,IAAI,CAAC,2BAA2B,CAAC,KAAK,CAAC,CAAC,CAAA;QAClD,CAAC;QAED,MAAM,UAAU,GAAG,IAAI,CAAC,WAAW,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;QAEjD,OAAO,IAAI,CAAC,aAAa,CAAC,UAAU,CAAC,CAAA;IACvC,CAAC;IAEO,aAAa;IACnB,kCAAkC;IAClC,KAAiB;QAEjB,qDAAqD;QACrD,MAAM,MAAM,GAAa,EAAE,CAAA;QAC3B,qDAAqD;QACrD,MAAM,KAAK,GAAa,EAAE,CAAA;QAE1B,0EAA0E;QAC1E,MAAM,OAAO,GAAG,CACd,UAAkB,EAClB,SAAS,GAAG,MAAM,CAAC,UAAU,CAAC,EAC9B,OAAO,GAAG,MAAM,CAAC,UAAU,GAAG,CAAC,CAAC,EACxB,EAAE;YACV,IAAI,OAAO,KAAK,SAAS,EAAE,CAAC;gBAC1B,uBAAuB;gBACvB,OAAO,MAAM,CAAC,iBAAiB,CAAA;YACjC,CAAC;YAED,wBAAwB;YACxB,MAAM,GAAG,GAAG,KAAK,CAAC,QAAQ,CAAC,SAAS,EAAE,OAAO,CAAC,CAAA;YAE9C,yDAAyD;YACzD,MAAM,IAAI,GAAG,IAAI,CAAC,mBAAmB,CAAC,GAAG,CAAC,CAAA;YAC1C,OAAO,IAAI,IAAI,MAAM,CAAC,iBAAiB,CAAA;QACzC,CAAC,CAAA;QAED,gEAAgE;QAChE,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;YACvC,MAAM,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YACd,IAAI,CAAC,GAAG,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACzB,mDAAmD;gBACnD,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,CAAA;YAClC,CAAC;iBAAM,CAAC;gBACN,sDAAsD;gBACtD,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,iBAAiB,CAAC,CAAA;YACtC,CAAC;QACH,CAAC;QAED,uEAAuE;QACvE,OAAO,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACzB,IAAI,UAAU,GAAG,MAAM,CAAC,iBAAiB,CAAA;YACzC,IAAI,oBAAoB,GAAG,CAAC,CAAC,CAAA;YAE7B,2CAA2C;YAC3C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;gBAC1C,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAE,CAAA;gBACtB,IAAI,IAAI,GAAG,UAAU,EAAE,CAAC;oBACtB,UAAU,GAAG,IAAI,CAAA;oBACjB,oBAAoB,GAAG,CAAC,CAAA;gBAC1B,CAAC;YACH,CAAC;YAED,mDAAmD;YACnD,IACE,UAAU,KAAK,MAAM,CAAC,iBAAiB;gBACvC,oBAAoB,KAAK,CAAC,CAAC,EAC3B,CAAC;gBACD,MAAK;YACP,CAAC;YAED,4EAA4E;YAC5E,MAAM,CAAC,MAAM,CAAC,oBAAoB,GAAG,CAAC,EAAE,CAAC,CAAC,CAAA;YAC1C,0CAA0C;YAC1C,KAAK,CAAC,MAAM,CAAC,oBAAoB,EAAE,CAAC,CAAC,CAAA;YAErC,wCAAwC;YACxC,KAAK,CAAC,oBAAoB,CAAC,GAAG,OAAO,CAAC,oBAAoB,CAAC,CAAA;YAE3D,qDAAqD;YACrD,IAAI,oBAAoB,GAAG,CAAC,EAAE,CAAC;gBAC7B,KAAK,CAAC,oBAAoB,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC,oBAAoB,GAAG,CAAC,CAAC,CAAA;YACrE,CAAC;QACH,CAAC;QAED,uFAAuF;QACvF,MAAM,MAAM,GAAa,EAAE,CAAA;QAC3B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC3C,MAAM,SAAS,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;YAC3B,MAAM,OAAO,GAAG,MAAM,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;YAC7B,MAAM,QAAQ,GAAG,IAAI,CAAC,0BAA0B,CAC9C,KAAK,CAAC,QAAQ,CAAC,SAAS,EAAE,OAAO,CAAC,CACnC,CAAA;YACD,MAAM,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;QACvB,CAAC;QACD,OAAO,MAAM,CAAA;IACf,CAAC;CACF;AA1aD,oDA0aC"}
|
|
1
|
+
{"version":3,"file":"BytePairEncodingCore.js","sourceRoot":"","sources":["../src/BytePairEncodingCore.ts"],"names":[],"mappings":";AAAA,gCAAgC;;;AAEhC,6CAA8E;AAC9E,uCAAwC;AAUxC,MAAM,WAAW,GAAG,IAAI,UAAU,CAAC,CAAC,CAAC,CAAA;AAExB,QAAA,OAAO,GAAG,IAAI,WAAW,CAAC,MAAM,CAAC,CAAA;AAE9C,MAAa,oBAAoB;IACtB,0BAA0B,CAAQ;IAC3C;;;;OAIG;IACK,mBAAmB,CAAkB;IACrC,yBAAyB,GAAG,IAAI,GAAG,EAAsB,CAAA;IACzD,2BAA2B,CAAiC;IACpE;;;;OAIG;IACK,yBAAyB,CAAqB;IAC9C,eAAe,CAAQ;IACvB,oBAAoB,CAAqB;IACzC,oBAAoB,CAAqB;IACzC,wBAAwB,CAAQ;IAChC,WAAW,GAAG,IAAI,WAAW,EAAE,CAAA;IAEvC,YAAY,EACV,mBAAmB,EACnB,oBAAoB,EACpB,eAAe,GACQ;QACvB,IAAI,CAAC,mBAAmB,GAAG,mBAAmB,CAAA;QAC9C,IAAI,CAAC,yBAAyB,GAAG,IAAI,GAAG,EAAkB,CAAA;QAE1D,iEAAiE;QACjE,IAAI,CAAC,0BAA0B,GAAG,MAAM,CAAC,IAAI,CAAC,mBAAmB,CAAC,CAAC,MAAM,CAAA;QACzE,MAAM,YAAY,GAA2B,EAAE,CAAA;QAC/C,6BAA6B;QAC7B,mBAAmB,CAAC,OAAO,CAAC,CAAC,KAAK,EAAE,IAAI,EAAE,EAAE;YAC1C,IAAI,OAAO,KAAK,KAAK,QAAQ,EAAE,CAAC;gBAC9B,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,KAAK,EAAE,IAAI,CAAC,CAAA;gBAC/C,OAAM;YACR,CAAC;YACD,MAAM,SAAS,GAAG,IAAI,UAAU,CAAC,KAAK,CAAC,CAAA;YACvC,YAAY,CAAC,IAAI,CAAC,CAAC,SAAS,EAAE,IAAI,CAAC,CAAC,CAAA;YACpC,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,IAAI,EAAE,SAAS,CAAC,CAAA;QACrD,CAAC,CAAC,CAAA;QACF,IAAI,CAAC,2BAA2B,GAAG,YAAY,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAC5D,IAAA,+BAAkB,EAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,CAC/B,CAAA;QACD,IAAI,CAAC,oBAAoB;YACvB,oBAAoB,IAAI,IAAI,GAAG,EAAkB,CAAA;QACnD,IAAI,CAAC,oBAAoB,GAAG,oBAAoB;YAC9C,CAAC,CAAC,IAAI,GAAG,CAAC,CAAC,GAAG,oBAAoB,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,EAAE,KAAK,CAAC,EAAE,EAAE,CAAC,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,CAAC;YACxE,CAAC,CAAC,IAAI,GAAG,EAAkB,CAAA;QAC7B,IAAI,CAAC,eAAe,GAAG,eAAe,CAAA;QAEtC,MAAM,oBAAoB,GAAG,CAAC,GAAG,IAAI,CAAC,oBAAoB,CAAC,IAAI,EAAE,CAAC,CAAC,GAAG,CACpE,sBAAY,CACb,CAAA;QACD,MAAM,qBAAqB,GAAG,oBAAoB,CAAC,IAAI,CAAC,GAAG,CAAC,CAAA;QAC5D,IAAI,CAAC;YACH,IAAI,CAAC,wBAAwB,GAAG,IAAI,MAAM,CAAC,qBAAqB,CAAC,CAAA;QACnE,CAAC;QAAC,MAAM,CAAC;YACP,MAAM,IAAI,KAAK,CAAC,qCAAqC,CAAC,CAAA;QACxD,CAAC;IACH,CAAC;IAED,CAAC,qBAAqB,CACpB,IAAY,EACZ,cAA4B;QAE5B,IAAI,UAAU,GAAG,CAAC,CAAA;QAClB,IAAI,eAAe,GAAG,CAAC,CAAA;QAEvB,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,gBAAgB,GAAG,IAAI,CAAC,oBAAoB,CAChD,IAAI,EACJ,cAAc,EACd,UAAU,CACX,CAAA;YACD,MAAM,qBAAqB,GAAG,gBAAgB,EAAE,CAAC,CAAC,CAAC,CAAA;YAEnD,MAAM,QAAQ,GAAG,qBAAqB,IAAI,IAAI,CAAC,MAAM,CAAA;YAErD,MAAM,iBAAiB,GAAG,IAAI,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAA;YAE1D,KAAK,MAAM,CAAC,KAAK,CAAC,IAAI,iBAAiB,CAAC,QAAQ,CAAC,IAAI,CAAC,eAAe,CAAC,EAAE,CAAC;gBACvE,MAAM,KAAK,GAAG,IAAI,CAAC,oBAAoB,CAAC,KAAK,CAAC,CAAA;gBAC9C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;oBACxB,eAAe,GAAG,CAAC,CAAA;oBACnB,MAAM,CAAC,KAAK,CAAC,CAAA;oBAEb,SAAQ;gBACV,CAAC;gBAED,MAAM,MAAM,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;gBACzC,eAAe,GAAG,MAAM,CAAC,MAAM,CAAA;gBAC/B,MAAM,MAAM,CAAA;YACd,CAAC;YAED,IAAI,qBAAqB,KAAK,SAAS,EAAE,CAAC;gBACxC,MAAM,YAAY,GAAG,gBAAiB,CAAC,CAAC,CAAC,CAAA;gBACzC,MAAM,iBAAiB,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,YAAY,CAAC,CAAA;gBACrE,IAAI,iBAAiB,KAAK,SAAS,EAAE,CAAC;oBACpC,MAAM,IAAI,KAAK,CACb,kBAAkB,YAAY,wCAAwC,CACvE,CAAA;gBACH,CAAC;gBACD,MAAM,CAAC,iBAAiB,CAAC,CAAA;gBACzB,UAAU,GAAG,qBAAqB,GAAG,YAAY,CAAC,MAAM,CAAA;gBACxD,eAAe,GAAG,CAAC,CAAA;YACrB,CAAC;iBAAM,CAAC;gBACN,MAAK;YACP,CAAC;QACH,CAAC;QAED,OAAO,eAAe,CAAA;IACxB,CAAC;IAED,YAAY,CAAC,IAAY,EAAE,cAA4B;QACrD,IAAI,UAAU,GAAG,CAAC,CAAA;QAClB,MAAM,WAAW,GAAa,EAAE,CAAA,CAAC,kCAAkC;QAEnE,iDAAiD;QACjD,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,gBAAgB,GAAG,IAAI,CAAC,oBAAoB,CAChD,IAAI,EACJ,cAAc,EACd,UAAU,CACX,CAAA;YACD,MAAM,qBAAqB,GAAG,gBAAgB,EAAE,CAAC,CAAC,CAAC,CAAA;YAEnD,MAAM,QAAQ,GAAG,qBAAqB,IAAI,IAAI,CAAC,MAAM,CAAA;YAErD,MAAM,iBAAiB,GAAG,IAAI,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAA;YAE1D,KAAK,MAAM,CAAC,KAAK,CAAC,IAAI,iBAAiB,CAAC,QAAQ,CAAC,IAAI,CAAC,eAAe,CAAC,EAAE,CAAC;gBACvE,MAAM,KAAK,GAAG,IAAI,CAAC,oBAAoB,CAAC,KAAK,CAAC,CAAA;gBAC9C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;oBACxB,WAAW,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;oBAEvB,SAAQ;gBACV,CAAC;gBAED,MAAM,MAAM,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;gBACzC,WAAW,CAAC,IAAI,CAAC,GAAG,MAAM,CAAC,CAAA;YAC7B,CAAC;YAED,IAAI,qBAAqB,KAAK,SAAS,EAAE,CAAC;gBACxC,MAAM,YAAY,GAAG,gBAAiB,CAAC,CAAC,CAAC,CAAA;gBACzC,MAAM,iBAAiB,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,YAAY,CAAC,CAAA;gBACrE,IAAI,iBAAiB,KAAK,SAAS,EAAE,CAAC;oBACpC,MAAM,IAAI,KAAK,CACb,kBAAkB,YAAY,wCAAwC,CACvE,CAAA;gBACH,CAAC;gBACD,WAAW,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAA;gBACnC,UAAU,GAAG,qBAAqB,GAAG,YAAY,CAAC,MAAM,CAAA;YAC1D,CAAC;iBAAM,CAAC;gBACN,MAAK;YACP,CAAC;QACH,CAAC;QAED,OAAO,WAAW,CAAA;IACpB,CAAC;IAED,CAAC,qBAAqB,CACpB,MAAwB;QAExB,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YAC3B,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;YAC7C,IAAI,UAAU,EAAE,CAAC;gBACf,MAAM,UAAU,CAAA;YAClB,CAAC;QACH,CAAC;IACH,CAAC;IAED,YAAY,CAAC,MAAwB;QACnC,IAAI,OAAO,GAAG,EAAE,CAAA;QAChB,IAAI,SAAS,GAAG,WAAW,CAAA;QAE3B,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YAC3B,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;YAC7C,IAAI,UAAU,KAAK,SAAS,EAAE,CAAC;gBAC7B,MAAM,IAAI,KAAK,CAAC,SAAS,KAAK,mCAAmC,CAAC,CAAA;YACpE,CAAC;YACD,IAAI,OAAO,UAAU,KAAK,QAAQ,EAAE,CAAC;gBACnC,IAAI,SAAS,KAAK,WAAW,EAAE,CAAC;oBAC9B,OAAO,IAAI,eAAO,CAAC,MAAM,CAAC,SAAS,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;oBACtD,SAAS,GAAG,WAAW,CAAA;gBACzB,CAAC;gBACD,OAAO,IAAI,UAAU,CAAA;YACvB,CAAC;iBAAM,CAAC;gBACN,MAAM,SAAS,GAAG,IAAI,UAAU,CAAC,SAAS,CAAC,MAAM,GAAG,UAAU,CAAC,MAAM,CAAC,CAAA;gBACtE,SAAS,CAAC,GAAG,CAAC,SAAS,CAAC,CAAA;gBACxB,SAAS,CAAC,GAAG,CAAC,UAAU,EAAE,SAAS,CAAC,MAAM,CAAC,CAAA;gBAC3C,SAAS,GAAG,SAAS,CAAA;YACvB,CAAC;QACH,CAAC;QAED,IAAI,SAAS,KAAK,WAAW,EAAE,CAAC;YAC9B,OAAO,IAAI,eAAO,CAAC,MAAM,CAAC,SAAS,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;QACxD,CAAC;QACD,OAAO,OAAO,CAAA;IAChB,CAAC;IAED,KAAK,CAAC,CAAC,yBAAyB,CAC9B,MAA6B;QAE7B,IAAI,KAAK,EAAE,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YACjC,MAAM,kBAAkB,GAAG,IAAI,CAAC,cAAc,CAAC,KAAK,CAAC,CAAA;YACrD,IAAI,kBAAkB,EAAE,CAAC;gBACvB,MAAM,kBAAkB,CAAA;YAC1B,CAAC;QACH,CAAC;IACH,CAAC;IAEO,oBAAoB,CAAC,GAAW;QACtC,OAAO,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,GAAG,CAAC,CAAA;IAChD,CAAC;IAEO,2BAA2B,CAAC,GAAW;QAC7C,MAAM,KAAK,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,CAAA;QAC5C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,KAAK,CACb,wDAAwD,GAAG,EAAE,CAC9D,CAAA;QACH,CAAC;QACD,OAAO,KAAK,CAAA;IACd,CAAC;IAEO,mBAAmB,CAAC,GAAe;QACzC,MAAM,WAAW,GAAG,IAAA,+BAAkB,EAAC,GAAG,CAAC,CAAA;QAE3C,IAAI,WAAW,KAAK,SAAS,EAAE,CAAC;YAC9B,OAAO,IAAI,CAAC,oBAAoB,CAAC,WAAW,CAAC,CAAA;QAC/C,CAAC;QAED,2CAA2C;QAC3C,MAAM,KAAK,GAAG,IAAI,CAAC,YAAY,CAAC,GAAG,CAAC,CAAA;QACpC,IAAI,KAAK,KAAK,CAAC,CAAC,EAAE,CAAC;YACjB,OAAO,IAAI,CAAC,2BAA2B,CAAC,KAAK,CAAE,CAAC,CAAC,CAAC,CAAA;QACpD,CAAC;QAED,OAAO,SAAS,CAAA;IAClB,CAAC;IAEO,0BAA0B,CAAC,GAAe;QAChD,MAAM,KAAK,GAAG,IAAI,CAAC,mBAAmB,CAAC,GAAG,CAAC,CAAA;QAC3C,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;YACxB,MAAM,IAAI,KAAK,CACb,wDAAwD,GAAG,CAAC,QAAQ,EAAE,EAAE,CACzE,CAAA;QACH,CAAC;QACD,OAAO,KAAK,CAAA;IACd,CAAC;IAED,mCAAmC;IAC3B,YAAY,CAAC,GAAe;QAClC,IAAI,GAAG,GAAG,CAAC,CAAA;QACX,IAAI,IAAI,GAAG,IAAI,CAAC,2BAA2B,CAAC,MAAM,GAAG,CAAC,CAAA;QAEtD,OAAO,GAAG,IAAI,IAAI,EAAE,CAAC;YACnB,sCAAsC;YACtC,MAAM,GAAG,GAAG,CAAC,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9B,MAAM,MAAM,GAAG,IAAI,CAAC,2BAA2B,CAAC,GAAG,CAAE,CAAC,CAAC,CAAC,CAAA;YACxD,IAAI,GAAG,GAAG,CAAC,CAAA;YACX,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,CAAC,MAAM,EAAE,GAAG,CAAC,MAAM,CAAC,CAAA;YACrD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,EAAE,CAAC,EAAE,EAAE,CAAC;gBACnC,GAAG,GAAG,MAAM,CAAC,CAAC,CAAE,GAAG,GAAG,CAAC,CAAC,CAAE,CAAA;gBAC1B,IAAI,GAAG,KAAK,CAAC;oBAAE,MAAK;YACtB,CAAC;YACD,IAAI,GAAG,KAAK,CAAC,EAAE,CAAC;gBACd,GAAG,GAAG,MAAM,CAAC,MAAM,GAAG,GAAG,CAAC,MAAM,CAAA;YAClC,CAAC;YACD,IAAI,GAAG,KAAK,CAAC,EAAE,CAAC;gBACd,OAAO,GAAG,CAAA;YACZ,CAAC;YACD,IAAI,GAAG,GAAG,CAAC,EAAE,CAAC;gBACZ,GAAG,GAAG,GAAG,GAAG,CAAC,CAAA;YACf,CAAC;iBAAM,CAAC;gBACN,IAAI,GAAG,GAAG,GAAG,CAAC,CAAA;YAChB,CAAC;QACH,CAAC;QACD,OAAO,CAAC,CAAC,CAAA;IACX,CAAC;IAEO,oBAAoB,CAC1B,IAAY,EACZ,cAAuC,EACvC,UAAkB;QAElB,IAAI,WAAW,GAAG,UAAU,CAAA;QAE5B,iDAAiD;QACjD,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,gBAAgB,GAAG,IAAI,CAAC,wBAAwB,CAAC,IAAI,CACzD,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC,CACrC,CAAA;YAED,IAAI,CAAC,gBAAgB,EAAE,CAAC;gBACtB,OAAO,SAAS,CAAA;YAClB,CAAC;YAED,MAAM,YAAY,GAAG,gBAAgB,CAAC,CAAC,CAAC,CAAA;YAExC,IAAI,cAAc,EAAE,GAAG,CAAC,YAAY,CAAC,EAAE,CAAC;gBACtC,MAAM,sBAAsB,GAAG,gBAAgB,CAAC,KAAK,GAAG,WAAW,CAAA;gBACnE,OAAO,CAAC,sBAAsB,EAAE,YAAY,CAAC,CAAA;YAC/C,CAAC;YAED,WAAW,GAAG,gBAAgB,CAAC,KAAK,GAAG,WAAW,GAAG,CAAC,CAAA;QACxD,CAAC;IACH,CAAC;IAEO,cAAc,CAAC,SAAiB;QACtC,MAAM,KAAK,GAAG,IAAI,CAAC,mBAAmB,CAAC,SAAS,CAAC,CAAA;QACjD,IAAI,OAAO,KAAK,KAAK,QAAQ,EAAE,CAAC;YAC9B,OAAO,KAAK,CAAA;QACd,CAAC;QACD,IAAI,OAAO,KAAK,KAAK,QAAQ,EAAE,CAAC;YAC9B,MAAM,UAAU,GAAG,IAAI,CAAC,yBAAyB,CAAC,GAAG,CAAC,SAAS,CAAC,CAAA;YAChE,IAAI,UAAU,EAAE,CAAC;gBACf,OAAO,UAAU,CAAA;YACnB,CAAC;QACH,CAAC;QACD,OAAO,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,SAAS,CAAC,CAAA;IACjD,CAAC;IAEO,cAAc,CAAC,KAAa;QAClC,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,IAAI,IAAA,oBAAO,EAAC,KAAK,CAAC,WAAW,CAAC,CAAC,CAAE,CAAC,EAAE,CAAC;YACzD,OAAO,CAAC,IAAI,CAAC,2BAA2B,CAAC,KAAK,CAAC,CAAC,CAAA;QAClD,CAAC;QAED,MAAM,UAAU,GAAG,IAAI,CAAC,WAAW,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;QAEjD,OAAO,IAAI,CAAC,aAAa,CAAC,UAAU,CAAC,CAAA;IACvC,CAAC;IAEO,aAAa;IACnB,kCAAkC;IAClC,KAAiB;QAEjB,qDAAqD;QACrD,MAAM,MAAM,GAAa,EAAE,CAAA;QAC3B,qDAAqD;QACrD,MAAM,KAAK,GAAa,EAAE,CAAA;QAE1B,0EAA0E;QAC1E,MAAM,OAAO,GAAG,CACd,UAAkB,EAClB,SAAS,GAAG,MAAM,CAAC,UAAU,CAAC,EAC9B,OAAO,GAAG,MAAM,CAAC,UAAU,GAAG,CAAC,CAAC,EACxB,EAAE;YACV,IAAI,OAAO,KAAK,SAAS,EAAE,CAAC;gBAC1B,uBAAuB;gBACvB,OAAO,MAAM,CAAC,iBAAiB,CAAA;YACjC,CAAC;YAED,wBAAwB;YACxB,MAAM,GAAG,GAAG,KAAK,CAAC,QAAQ,CAAC,SAAS,EAAE,OAAO,CAAC,CAAA;YAE9C,yDAAyD;YACzD,MAAM,IAAI,GAAG,IAAI,CAAC,mBAAmB,CAAC,GAAG,CAAC,CAAA;YAC1C,OAAO,IAAI,IAAI,MAAM,CAAC,iBAAiB,CAAA;QACzC,CAAC,CAAA;QAED,gEAAgE;QAChE,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;YACvC,MAAM,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YACd,IAAI,CAAC,GAAG,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACzB,mDAAmD;gBACnD,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,CAAA;YAClC,CAAC;iBAAM,CAAC;gBACN,sDAAsD;gBACtD,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,iBAAiB,CAAC,CAAA;YACtC,CAAC;QACH,CAAC;QAED,uEAAuE;QACvE,OAAO,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACzB,IAAI,UAAU,GAAG,MAAM,CAAC,iBAAiB,CAAA;YACzC,IAAI,oBAAoB,GAAG,CAAC,CAAC,CAAA;YAE7B,2CAA2C;YAC3C,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;gBAC1C,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAE,CAAA;gBACtB,IAAI,IAAI,GAAG,UAAU,EAAE,CAAC;oBACtB,UAAU,GAAG,IAAI,CAAA;oBACjB,oBAAoB,GAAG,CAAC,CAAA;gBAC1B,CAAC;YACH,CAAC;YAED,mDAAmD;YACnD,IACE,UAAU,KAAK,MAAM,CAAC,iBAAiB;gBACvC,oBAAoB,KAAK,CAAC,CAAC,EAC3B,CAAC;gBACD,MAAK;YACP,CAAC;YAED,4EAA4E;YAC5E,MAAM,CAAC,MAAM,CAAC,oBAAoB,GAAG,CAAC,EAAE,CAAC,CAAC,CAAA;YAC1C,0CAA0C;YAC1C,KAAK,CAAC,MAAM,CAAC,oBAAoB,EAAE,CAAC,CAAC,CAAA;YAErC,wCAAwC;YACxC,KAAK,CAAC,oBAAoB,CAAC,GAAG,OAAO,CAAC,oBAAoB,CAAC,CAAA;YAE3D,qDAAqD;YACrD,IAAI,oBAAoB,GAAG,CAAC,EAAE,CAAC;gBAC7B,KAAK,CAAC,oBAAoB,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC,oBAAoB,GAAG,CAAC,CAAC,CAAA;YACrE,CAAC;QACH,CAAC;QAED,uFAAuF;QACvF,MAAM,MAAM,GAAa,EAAE,CAAA;QAC3B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;YAC3C,MAAM,SAAS,GAAG,MAAM,CAAC,CAAC,CAAC,CAAA;YAC3B,MAAM,OAAO,GAAG,MAAM,CAAC,CAAC,GAAG,CAAC,CAAC,CAAA;YAC7B,MAAM,QAAQ,GAAG,IAAI,CAAC,0BAA0B,CAC9C,KAAK,CAAC,QAAQ,CAAC,SAAS,EAAE,OAAO,CAAC,CACnC,CAAA;YACD,MAAM,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAA;QACvB,CAAC;QACD,OAAO,MAAM,CAAA;IACf,CAAC;CACF;AAxaD,oDAwaC"}
|
package/cjs/GptEncoding.d.ts
CHANGED
|
@@ -1,9 +1,19 @@
|
|
|
1
|
+
import { ALL_SPECIAL_TOKENS } from './constants.js';
|
|
1
2
|
import { type EncodingName, type ModelName } from './mapping.js';
|
|
2
3
|
import { type GetMergeableRanksAsyncFn, type GetMergeableRanksFn } from './modelParams.js';
|
|
3
|
-
export declare const ALL_SPECIAL_TOKENS = "all";
|
|
4
4
|
export interface EncodeOptions {
|
|
5
|
-
|
|
6
|
-
|
|
5
|
+
/**
|
|
6
|
+
* A list of special tokens that are allowed in the input.
|
|
7
|
+
* If set to 'all', all special tokens are allowed except those in disallowedSpecial.
|
|
8
|
+
* @default undefined
|
|
9
|
+
*/
|
|
10
|
+
allowedSpecial?: Set<string> | typeof ALL_SPECIAL_TOKENS;
|
|
11
|
+
/**
|
|
12
|
+
* A list of special tokens that are disallowed in the input.
|
|
13
|
+
* If set to 'all', all special tokens are disallowed except those in allowedSpecial.
|
|
14
|
+
* @default 'all'
|
|
15
|
+
*/
|
|
16
|
+
disallowedSpecial?: Set<string> | typeof ALL_SPECIAL_TOKENS;
|
|
7
17
|
}
|
|
8
18
|
export interface ChatMessage {
|
|
9
19
|
role?: 'system' | 'user' | 'assistant';
|
|
@@ -21,7 +31,7 @@ export declare class GptEncoding {
|
|
|
21
31
|
static FimSuffix: string;
|
|
22
32
|
modelName?: ModelName;
|
|
23
33
|
private bytePairEncodingCoreProcessor;
|
|
24
|
-
private
|
|
34
|
+
private specialTokensEncoder;
|
|
25
35
|
private specialTokensSet;
|
|
26
36
|
private allSpecialTokenRegex;
|
|
27
37
|
private defaultSpecialTokenConfig;
|
package/cjs/GptEncoding.js
CHANGED
|
@@ -1,15 +1,15 @@
|
|
|
1
1
|
"use strict";
|
|
2
2
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
3
|
-
exports.GptEncoding =
|
|
3
|
+
exports.GptEncoding = void 0;
|
|
4
4
|
/* eslint-disable @typescript-eslint/member-ordering */
|
|
5
5
|
/* eslint-disable no-param-reassign */
|
|
6
6
|
const BytePairEncodingCore_js_1 = require("./BytePairEncodingCore.js");
|
|
7
|
+
const constants_js_1 = require("./constants.js");
|
|
7
8
|
const mapping_js_1 = require("./mapping.js");
|
|
8
9
|
const modelParams_js_1 = require("./modelParams.js");
|
|
9
10
|
const specialTokens_js_1 = require("./specialTokens.js");
|
|
10
11
|
const utfUtil_js_1 = require("./utfUtil.js");
|
|
11
12
|
const util_js_1 = require("./util.js");
|
|
12
|
-
exports.ALL_SPECIAL_TOKENS = 'all';
|
|
13
13
|
class GptEncoding {
|
|
14
14
|
static EndOfPrompt = specialTokens_js_1.EndOfPrompt;
|
|
15
15
|
static EndOfText = specialTokens_js_1.EndOfText;
|
|
@@ -18,25 +18,25 @@ class GptEncoding {
|
|
|
18
18
|
static FimSuffix = specialTokens_js_1.FimSuffix;
|
|
19
19
|
modelName;
|
|
20
20
|
bytePairEncodingCoreProcessor;
|
|
21
|
-
|
|
21
|
+
specialTokensEncoder;
|
|
22
22
|
specialTokensSet;
|
|
23
23
|
allSpecialTokenRegex;
|
|
24
24
|
defaultSpecialTokenConfig;
|
|
25
25
|
vocabularySize;
|
|
26
|
-
constructor({ mergeableBytePairRanks,
|
|
27
|
-
this.
|
|
28
|
-
this.specialTokensSet = new Set(this.
|
|
26
|
+
constructor({ bytePairRankDecoder: mergeableBytePairRanks, specialTokensEncoder, expectedVocabularySize, modelName, ...rest }) {
|
|
27
|
+
this.specialTokensEncoder = specialTokensEncoder;
|
|
28
|
+
this.specialTokensSet = new Set(this.specialTokensEncoder.keys());
|
|
29
29
|
this.allSpecialTokenRegex = (0, util_js_1.getSpecialTokenRegex)(this.specialTokensSet);
|
|
30
30
|
this.bytePairEncodingCoreProcessor = new BytePairEncodingCore_js_1.BytePairEncodingCore({
|
|
31
|
-
mergeableBytePairRanks,
|
|
32
|
-
|
|
31
|
+
bytePairRankDecoder: mergeableBytePairRanks,
|
|
32
|
+
specialTokensEncoder,
|
|
33
33
|
...rest,
|
|
34
34
|
});
|
|
35
35
|
this.defaultSpecialTokenConfig = this.processSpecialTokens();
|
|
36
|
-
const maxTokenValue = Math.max(mergeableBytePairRanks.length - 1, (0, util_js_1.getMaxValueFromMap)(
|
|
36
|
+
const maxTokenValue = Math.max(mergeableBytePairRanks.length - 1, (0, util_js_1.getMaxValueFromMap)(specialTokensEncoder));
|
|
37
37
|
this.vocabularySize =
|
|
38
38
|
this.bytePairEncodingCoreProcessor.mergeableBytePairRankCount +
|
|
39
|
-
|
|
39
|
+
specialTokensEncoder.size;
|
|
40
40
|
if (expectedVocabularySize !== undefined) {
|
|
41
41
|
if (this.vocabularySize !== expectedVocabularySize) {
|
|
42
42
|
throw new Error('The number of mergeable tokens and special tokens must be equal to expectedVocabularySize.');
|
|
@@ -76,14 +76,31 @@ class GptEncoding {
|
|
|
76
76
|
}
|
|
77
77
|
processSpecialTokens({ allowedSpecial, disallowedSpecial, } = {}) {
|
|
78
78
|
let regexPattern;
|
|
79
|
-
if (allowedSpecial
|
|
79
|
+
if (allowedSpecial === constants_js_1.ALL_SPECIAL_TOKENS ||
|
|
80
|
+
allowedSpecial?.has(constants_js_1.ALL_SPECIAL_TOKENS)) {
|
|
80
81
|
allowedSpecial = new Set(this.specialTokensSet);
|
|
82
|
+
const allowedSpecialSet = allowedSpecial;
|
|
83
|
+
if (disallowedSpecial === constants_js_1.ALL_SPECIAL_TOKENS) {
|
|
84
|
+
throw new Error('allowedSpecial and disallowedSpecial cannot both be set to "all".');
|
|
85
|
+
}
|
|
86
|
+
if (typeof disallowedSpecial === 'object') {
|
|
87
|
+
// remove any special tokens that are disallowed
|
|
88
|
+
disallowedSpecial.forEach((val) => allowedSpecialSet.delete(val));
|
|
89
|
+
}
|
|
90
|
+
else {
|
|
91
|
+
// all special tokens are allowed, and no 'disallowedSpecial' is provided
|
|
92
|
+
disallowedSpecial = new Set();
|
|
93
|
+
}
|
|
81
94
|
}
|
|
82
|
-
if (!disallowedSpecial ||
|
|
95
|
+
if (!disallowedSpecial ||
|
|
96
|
+
disallowedSpecial === constants_js_1.ALL_SPECIAL_TOKENS ||
|
|
97
|
+
disallowedSpecial.has(constants_js_1.ALL_SPECIAL_TOKENS)) {
|
|
83
98
|
// by default, all special tokens are disallowed
|
|
84
99
|
disallowedSpecial = new Set(this.specialTokensSet);
|
|
100
|
+
const disallowedSpecialSet = disallowedSpecial;
|
|
85
101
|
if (allowedSpecial?.size) {
|
|
86
|
-
allowedSpecial.forEach((val) =>
|
|
102
|
+
allowedSpecial.forEach((val) => disallowedSpecialSet.delete(val));
|
|
103
|
+
// disallowed takes precedence over allowed
|
|
87
104
|
disallowedSpecial.forEach((val) => allowedSpecial.delete(val));
|
|
88
105
|
regexPattern = (0, util_js_1.getSpecialTokenRegex)(disallowedSpecial);
|
|
89
106
|
}
|
|
@@ -129,8 +146,8 @@ class GptEncoding {
|
|
|
129
146
|
throw new Error('Model name must be provided either during initialization or passed in to the method.');
|
|
130
147
|
}
|
|
131
148
|
const params = mapping_js_1.chatModelParams[model];
|
|
132
|
-
const chatStartToken = this.
|
|
133
|
-
const chatEndToken = this.
|
|
149
|
+
const chatStartToken = this.specialTokensEncoder.get(specialTokens_js_1.ImStart);
|
|
150
|
+
const chatEndToken = this.specialTokensEncoder.get(specialTokens_js_1.ImEnd);
|
|
134
151
|
if (!params || chatStartToken === undefined || chatEndToken === undefined) {
|
|
135
152
|
throw new Error(`Model '${model}' does not support chat.`);
|
|
136
153
|
}
|
package/cjs/GptEncoding.js.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"GptEncoding.js","sourceRoot":"","sources":["../src/GptEncoding.ts"],"names":[],"mappings":";;;AAAA,uDAAuD;AACvD,sCAAsC;AACtC,uEAAyE;AACzE,6CAOqB;AACrB,qDAMyB;AACzB,yDAS2B;AAC3B,6CAAiE;AACjE,uCAAoE;
|
|
1
|
+
{"version":3,"file":"GptEncoding.js","sourceRoot":"","sources":["../src/GptEncoding.ts"],"names":[],"mappings":";;;AAAA,uDAAuD;AACvD,sCAAsC;AACtC,uEAAyE;AACzE,iDAAmD;AACnD,6CAOqB;AACrB,qDAMyB;AACzB,yDAS2B;AAC3B,6CAAiE;AACjE,uCAAoE;AAgCpE,MAAa,WAAW;IACtB,MAAM,CAAC,WAAW,GAAG,8BAAW,CAAA;IAChC,MAAM,CAAC,SAAS,GAAG,4BAAS,CAAA;IAC5B,MAAM,CAAC,SAAS,GAAG,4BAAS,CAAA;IAC5B,MAAM,CAAC,SAAS,GAAG,4BAAS,CAAA;IAC5B,MAAM,CAAC,SAAS,GAAG,4BAAS,CAAA;IAE5B,SAAS,CAAY;IACb,6BAA6B,CAAsB;IACnD,oBAAoB,CAAqB;IACzC,gBAAgB,CAAa;IAC7B,oBAAoB,CAAQ;IAC5B,yBAAyB,CAAoB;IAE5C,cAAc,CAAQ;IAE/B,YAAoB,EAClB,mBAAmB,EAAE,sBAAsB,EAC3C,oBAAoB,EACpB,sBAAsB,EACtB,SAAS,EACT,GAAG,IAAI,EACQ;QACf,IAAI,CAAC,oBAAoB,GAAG,oBAAoB,CAAA;QAChD,IAAI,CAAC,gBAAgB,GAAG,IAAI,GAAG,CAAS,IAAI,CAAC,oBAAoB,CAAC,IAAI,EAAE,CAAC,CAAA;QACzE,IAAI,CAAC,oBAAoB,GAAG,IAAA,8BAAoB,EAAC,IAAI,CAAC,gBAAgB,CAAC,CAAA;QAEvE,IAAI,CAAC,6BAA6B,GAAG,IAAI,8CAAoB,CAAC;YAC5D,mBAAmB,EAAE,sBAAsB;YAC3C,oBAAoB;YACpB,GAAG,IAAI;SACR,CAAC,CAAA;QACF,IAAI,CAAC,yBAAyB,GAAG,IAAI,CAAC,oBAAoB,EAAE,CAAA;QAE5D,MAAM,aAAa,GAAG,IAAI,CAAC,GAAG,CAC5B,sBAAsB,CAAC,MAAM,GAAG,CAAC,EACjC,IAAA,4BAAkB,EAAC,oBAAoB,CAAC,CACzC,CAAA;QAED,IAAI,CAAC,cAAc;YACjB,IAAI,CAAC,6BAA6B,CAAC,0BAA0B;gBAC7D,oBAAoB,CAAC,IAAI,CAAA;QAE3B,IAAI,sBAAsB,KAAK,SAAS,EAAE,CAAC;YACzC,IAAI,IAAI,CAAC,cAAc,KAAK,sBAAsB,EAAE,CAAC;gBACnD,MAAM,IAAI,KAAK,CACb,4FAA4F,CAC7F,CAAA;YACH,CAAC;YAED,IAAI,aAAa,KAAK,sBAAsB,GAAG,CAAC,EAAE,CAAC;gBACjD,MAAM,IAAI,KAAK,CACb,mHAAmH,aAAa,cAC9H,sBAAsB,GAAG,CAC3B,EAAE,CACH,CAAA;YACH,CAAC;QACH,CAAC;QAED,IAAI,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACpC,IAAI,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACpC,IAAI,CAAC,eAAe,GAAG,IAAI,CAAC,eAAe,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACtD,IAAI,CAAC,eAAe,GAAG,IAAI,CAAC,eAAe,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACtD,IAAI,CAAC,oBAAoB,GAAG,IAAI,CAAC,oBAAoB,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QAChE,IAAI,CAAC,WAAW,GAAG,IAAI,CAAC,WAAW,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QAC9C,IAAI,CAAC,kBAAkB,GAAG,IAAI,CAAC,kBAAkB,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QAC5D,IAAI,CAAC,UAAU,GAAG,IAAI,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QAC5C,IAAI,CAAC,mBAAmB,GAAG,IAAI,CAAC,mBAAmB,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QAC9D,IAAI,CAAC,SAAS,GAAG,SAAS,CAAA;IAC5B,CAAC;IAED,MAAM,CAAC,cAAc,CACnB,YAA0B,EAC1B,iBAAsC;QAEtC,MAAM,WAAW,GAAG,IAAA,kCAAiB,EAAC,YAAY,EAAE,iBAAiB,CAAC,CAAA;QACtE,OAAO,IAAI,WAAW,CAAC,WAAW,CAAC,CAAA;IACrC,CAAC;IAED,MAAM,CAAC,sBAAsB,CAC3B,SAAoB,EACpB,iBAAsC;QAEtC,MAAM,YAAY,GAAG,+BAAkB,CAAC,SAAS,CAAC,CAAA;QAClD,MAAM,WAAW,GAAG,IAAA,kCAAiB,EAAC,YAAY,EAAE,iBAAiB,CAAC,CAAA;QACtE,OAAO,IAAI,WAAW,CAAC,EAAE,GAAG,WAAW,EAAE,SAAS,EAAE,CAAC,CAAA;IACvD,CAAC;IAED,MAAM,CAAC,KAAK,CAAC,mBAAmB,CAC9B,YAA0B,EAC1B,iBAA2C;QAE3C,MAAM,WAAW,GAAG,MAAM,IAAA,oCAAmB,EAC3C,YAAY,EACZ,iBAAiB,CAClB,CAAA;QACD,OAAO,IAAI,WAAW,CAAC,WAAW,CAAC,CAAA;IACrC,CAAC;IAED,MAAM,CAAC,KAAK,CAAC,2BAA2B,CACtC,SAAoB,EACpB,iBAA2C;QAE3C,MAAM,YAAY,GAAG,+BAAkB,CAAC,SAAS,CAAC,CAAA;QAClD,MAAM,WAAW,GAAG,MAAM,IAAA,oCAAmB,EAC3C,YAAY,EACZ,iBAAiB,CAClB,CAAA;QACD,OAAO,IAAI,WAAW,CAAC,EAAE,GAAG,WAAW,EAAE,SAAS,EAAE,CAAC,CAAA;IACvD,CAAC;IAEO,oBAAoB,CAAC,EAC3B,cAAc,EACd,iBAAiB,MACA,EAAE;QACnB,IAAI,YAAgC,CAAA;QAEpC,IACE,cAAc,KAAK,iCAAkB;YACrC,cAAc,EAAE,GAAG,CAAC,iCAAkB,CAAC,EACvC,CAAC;YACD,cAAc,GAAG,IAAI,GAAG,CAAC,IAAI,CAAC,gBAAgB,CAAC,CAAA;YAC/C,MAAM,iBAAiB,GAAG,cAAc,CAAA;YACxC,IAAI,iBAAiB,KAAK,iCAAkB,EAAE,CAAC;gBAC7C,MAAM,IAAI,KAAK,CACb,mEAAmE,CACpE,CAAA;YACH,CAAC;YACD,IAAI,OAAO,iBAAiB,KAAK,QAAQ,EAAE,CAAC;gBAC1C,gDAAgD;gBAChD,iBAAiB,CAAC,OAAO,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,iBAAiB,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACnE,CAAC;iBAAM,CAAC;gBACN,yEAAyE;gBACzE,iBAAiB,GAAG,IAAI,GAAG,EAAE,CAAA;YAC/B,CAAC;QACH,CAAC;QAED,IACE,CAAC,iBAAiB;YAClB,iBAAiB,KAAK,iCAAkB;YACxC,iBAAiB,CAAC,GAAG,CAAC,iCAAkB,CAAC,EACzC,CAAC;YACD,gDAAgD;YAChD,iBAAiB,GAAG,IAAI,GAAG,CAAC,IAAI,CAAC,gBAAgB,CAAC,CAAA;YAClD,MAAM,oBAAoB,GAAG,iBAAiB,CAAA;YAC9C,IAAI,cAAc,EAAE,IAAI,EAAE,CAAC;gBACzB,cAAc,CAAC,OAAO,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,oBAAoB,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;gBACjE,2CAA2C;gBAC3C,iBAAiB,CAAC,OAAO,CAAC,CAAC,GAAG,EAAE,EAAE,CAAC,cAAc,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;gBAC9D,YAAY,GAAG,IAAA,8BAAoB,EAAC,iBAAiB,CAAC,CAAA;YACxD,CAAC;iBAAM,CAAC;gBACN,YAAY,GAAG,IAAI,CAAC,oBAAoB,CAAA;YAC1C,CAAC;QACH,CAAC;QAED,OAAO,EAAE,cAAc,EAAE,YAAY,EAAE,CAAA;IACzC,CAAC;IAED,eAAe,CACb,YAAoB,EACpB,aAA6B;QAE7B,MAAM,kBAAkB,GAAG,aAAa;YACtC,CAAC,CAAC,IAAI,CAAC,oBAAoB,CAAC,aAAa,CAAC;YAC1C,CAAC,CAAC,IAAI,CAAC,yBAAyB,CAAA;QAElC,IAAI,kBAAkB,CAAC,YAAY,EAAE,CAAC;YACpC,MAAM,KAAK,GAAG,YAAY,CAAC,KAAK,CAAC,kBAAkB,CAAC,YAAY,CAAC,CAAA;YACjE,IAAI,KAAK,KAAK,IAAI,EAAE,CAAC;gBACnB,MAAM,IAAI,KAAK,CAAC,mCAAmC,KAAK,CAAC,CAAC,CAAC,EAAE,CAAC,CAAA;YAChE,CAAC;QACH,CAAC;QAED,OAAO,IAAI,CAAC,6BAA6B,CAAC,qBAAqB,CAC7D,YAAY,EACZ,kBAAkB,CAAC,cAAc,CAClC,CAAA;IACH,CAAC;IAED,MAAM,CAAC,YAAoB,EAAE,aAA6B;QACxD,MAAM,kBAAkB,GAAG,aAAa;YACtC,CAAC,CAAC,IAAI,CAAC,oBAAoB,CAAC,aAAa,CAAC;YAC1C,CAAC,CAAC,IAAI,CAAC,yBAAyB,CAAA;QAElC,IAAI,kBAAkB,CAAC,YAAY,EAAE,CAAC;YACpC,MAAM,KAAK,GAAG,YAAY,CAAC,KAAK,CAAC,kBAAkB,CAAC,YAAY,CAAC,CAAA;YACjE,IAAI,KAAK,KAAK,IAAI,EAAE,CAAC;gBACnB,MAAM,IAAI,KAAK,CAAC,mCAAmC,KAAK,CAAC,CAAC,CAAC,EAAE,CAAC,CAAA;YAChE,CAAC;QACH,CAAC;QAED,OAAO,IAAI,CAAC,6BAA6B,CAAC,YAAY,CACpD,YAAY,EACZ,kBAAkB,CAAC,cAAc,CAClC,CAAA;IACH,CAAC;IAED;;;;;;OAMG;IACH,CAAC,mBAAmB,CAClB,IAA2B,EAC3B,KAAK,GAAG,IAAI,CAAC,SAAS;QAEtB,IAAI,CAAC,KAAK,EAAE,CAAC;YACX,MAAM,IAAI,KAAK,CACb,sFAAsF,CACvF,CAAA;QACH,CAAC;QAED,MAAM,MAAM,GACV,4BAAe,CAAC,KAAsB,CAAC,CAAA;QACzC,MAAM,cAAc,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,0BAAO,CAAC,CAAA;QAC7D,MAAM,YAAY,GAAG,IAAI,CAAC,oBAAoB,CAAC,GAAG,CAAC,wBAAK,CAAC,CAAA;QAEzD,IAAI,CAAC,MAAM,IAAI,cAAc,KAAK,SAAS,IAAI,YAAY,KAAK,SAAS,EAAE,CAAC;YAC1E,MAAM,IAAI,KAAK,CAAC,UAAU,KAAK,0BAA0B,CAAC,CAAA;QAC5D,CAAC;QACD,MAAM,cAAc,GAAG,IAAI,GAAG,CAAC,CAAC,wBAAK,CAAC,CAAC,CAAA;QACvC,MAAM,EAAE,gBAAgB,EAAE,aAAa,EAAE,GAAG,MAAM,CAAA;QAClD,MAAM,uBAAuB,GAC3B,gBAAgB,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,gBAAgB,CAAC,CAAC,CAAC,CAAC,EAAE,CAAA;QAClE,MAAM,oBAAoB,GACxB,aAAa,CAAC,MAAM,GAAG,CAAC;YACtB,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,aAAa,EAAE,EAAE,cAAc,EAAE,CAAC;YAChD,CAAC,CAAC,EAAE,CAAA;QACR,MAAM,SAAS,GAAG,IAAI,GAAG,EAAoB,CAAA;QAE7C,KAAK,MAAM,EAAE,IAAI,GAAG,QAAQ,EAAE,IAAI,GAAG,IAAI,EAAE,OAAO,EAAE,IAAI,IAAI,EAAE,CAAC;YAC7D,IAAI,OAAO,KAAK,SAAS,EAAE,CAAC;gBAC1B,MAAM,IAAI,KAAK,CAAC,2CAA2C,CAAC,CAAA;YAC9D,CAAC;YAED,MAAM,CAAC,cAAc,CAAC,CAAA;YACtB,MAAM,WAAW,GAAG,SAAS,CAAC,GAAG,CAAC,IAAI,CAAC,IAAI,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;YAC5D,SAAS,CAAC,GAAG,CAAC,IAAI,EAAE,WAAW,CAAC,CAAA;YAChC,MAAM,WAAW,CAAA;YACjB,IAAI,oBAAoB,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACpC,MAAM,oBAAoB,CAAA;YAC5B,CAAC;YACD,KAAK,CAAC,CAAC,IAAI,CAAC,eAAe,CAAC,OAAO,CAAC,CAAA;YACpC,MAAM,CAAC,YAAY,CAAC,CAAA;YACpB,MAAM,uBAAuB,CAAA;QAC/B,CAAC;QAED,2DAA2D;QAC3D,MAAM,CAAC,cAAc,CAAC,CAAA;QACtB,KAAK,CAAC,CAAC,IAAI,CAAC,eAAe,CAAC,WAAW,CAAC,CAAA;QACxC,IAAI,oBAAoB,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACpC,MAAM,oBAAoB,CAAA;QAC5B,CAAC;IACH,CAAC;IAED;;;;;;OAMG;IACH,UAAU,CAAC,IAA4B,EAAE,KAAK,GAAG,IAAI,CAAC,SAAS;QAC7D,OAAO,CAAC,GAAG,IAAI,CAAC,mBAAmB,CAAC,IAAI,EAAE,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAA;IAC1D,CAAC;IAED;;OAEG;IACH,kBAAkB,CAChB,KAAqC,EACrC,UAAkB;QAElB,MAAM,cAAc,GAClB,OAAO,KAAK,KAAK,QAAQ;YACvB,CAAC,CAAC,IAAI,CAAC,eAAe,CAAC,KAAK,CAAC;YAC7B,CAAC,CAAC,IAAI,CAAC,mBAAmB,CAAC,KAAK,CAAC,CAAA;QACrC,IAAI,KAAK,GAAG,CAAC,CAAA;QACb,KAAK,MAAM,MAAM,IAAI,cAAc,EAAE,CAAC;YACpC,KAAK,IAAI,MAAM,CAAC,MAAM,CAAA;YACtB,IAAI,KAAK,GAAG,UAAU,EAAE,CAAC;gBACvB,OAAO,KAAK,CAAA;YACd,CAAC;QACH,CAAC;QACD,OAAO,KAAK,CAAA;IACd,CAAC;IAED,MAAM,CAAC,mBAAqC;QAC1C,OAAO,IAAI,CAAC,6BAA6B,CAAC,YAAY,CAAC,mBAAmB,CAAC,CAAA;IAC7E,CAAC;IAED,CAAC,eAAe,CACd,mBAAqC;QAErC,MAAM,oBAAoB,GACxB,IAAI,CAAC,6BAA6B,CAAC,qBAAqB,CACtD,mBAAmB,CACpB,CAAA;QAEH,IAAI,MAAM,GAAG,EAAE,CAAA;QAEf,KAAK,MAAM,WAAW,IAAI,oBAAoB,EAAE,CAAC;YAC/C,MAAM;gBACJ,OAAO,WAAW,KAAK,QAAQ;oBAC7B,CAAC,CAAC,WAAW;oBACb,CAAC,CAAC,iCAAO,CAAC,MAAM,CAAC,WAAW,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;YAEnD,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,IAAI,IAAA,+CAAkC,EAAC,MAAM,CAAC,EAAE,CAAC;gBACtE,yEAAyE;gBACzE,uCAAuC;gBACvC,SAAQ;YACV,CAAC;iBAAM,CAAC;gBACN,MAAM,MAAM,CAAA;gBACZ,eAAe;gBACf,MAAM,GAAG,EAAE,CAAA;YACb,CAAC;QACH,CAAC;QAED,+CAA+C;QAC/C,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,MAAM,MAAM,CAAA;QACd,CAAC;IACH,CAAC;IAED,KAAK,CAAC,CAAC,oBAAoB,CACzB,mBAA0C;QAE1C,MAAM,oBAAoB,GACxB,IAAI,CAAC,6BAA6B,CAAC,yBAAyB,CAC1D,mBAAmB,CACpB,CAAA;QAEH,IAAI,MAAM,GAAG,EAAE,CAAA;QAEf,IAAI,KAAK,EAAE,MAAM,WAAW,IAAI,oBAAoB,EAAE,CAAC;YACrD,MAAM;gBACJ,OAAO,WAAW,KAAK,QAAQ;oBAC7B,CAAC,CAAC,WAAW;oBACb,CAAC,CAAC,iCAAO,CAAC,MAAM,CAAC,WAAW,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;YAEnD,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,IAAI,IAAA,+CAAkC,EAAC,MAAM,CAAC,EAAE,CAAC;gBACtE,yEAAyE;gBACzE,uCAAuC;gBACvC,SAAQ;YACV,CAAC;iBAAM,CAAC;gBACN,MAAM,MAAM,CAAA;gBACZ,eAAe;gBACf,MAAM,GAAG,EAAE,CAAA;YACb,CAAC;QACH,CAAC;QAED,+CAA+C;QAC/C,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,MAAM,MAAM,CAAA;QACd,CAAC;IACH,CAAC;IAED,KAAK,CAAC,WAAW,CACf,mBAA0C;QAE1C,MAAM,oBAAoB,GACxB,IAAI,CAAC,6BAA6B,CAAC,yBAAyB,CAC1D,mBAAmB,CACpB,CAAA;QAEH,IAAI,MAAM,GAAG,EAAE,CAAA;QAEf,IAAI,KAAK,EAAE,MAAM,WAAW,IAAI,oBAAoB,EAAE,CAAC;YACrD,MAAM;gBACJ,OAAO,WAAW,KAAK,QAAQ;oBAC7B,CAAC,CAAC,WAAW;oBACb,CAAC,CAAC,iCAAO,CAAC,MAAM,CAAC,WAAW,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,CAAC,CAAA;QACrD,CAAC;QAED,OAAO,MAAM,CAAA;IACf,CAAC;;AAzXH,kCA0XC"}
|
package/cjs/GptEncoding.test.js
CHANGED
|
@@ -5,6 +5,7 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
|
|
|
5
5
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
6
6
|
const fs_1 = __importDefault(require("fs"));
|
|
7
7
|
const path_1 = __importDefault(require("path"));
|
|
8
|
+
const constants_js_1 = require("./constants.js");
|
|
8
9
|
const GptEncoding_js_1 = require("./GptEncoding.js");
|
|
9
10
|
const mapping_js_1 = require("./mapping.js");
|
|
10
11
|
const resolveEncoding_js_1 = require("./resolveEncoding.js");
|
|
@@ -58,13 +59,32 @@ const results = {
|
|
|
58
59
|
p50k_edit: sharedResults,
|
|
59
60
|
r50k_base: sharedResults,
|
|
60
61
|
};
|
|
62
|
+
const offsetPrompts = [
|
|
63
|
+
// Basic prompt with "hello world"
|
|
64
|
+
'hello world',
|
|
65
|
+
// Basic prompt with special token "<|endoftext|>"
|
|
66
|
+
'hello world<|endoftext|> green cow',
|
|
67
|
+
// Chinese text: "我非常渴望与人工智能一起工作"
|
|
68
|
+
'我非常渴望与人工智能一起工作',
|
|
69
|
+
// Contains the interesting tokens b'\xe0\xae\xbf\xe0\xae' and b'\xe0\xaf\x8d\xe0\xae'
|
|
70
|
+
// in which \xe0 is the start of a 3-byte UTF-8 character
|
|
71
|
+
'நடிகர் சூர்யா',
|
|
72
|
+
// Contains the interesting token b'\xa0\xe9\x99\xa4'
|
|
73
|
+
// in which \xe9 is the start of a 3-byte UTF-8 character and \xa0 is a continuation byte
|
|
74
|
+
' Ġ除',
|
|
75
|
+
];
|
|
61
76
|
// eslint-disable-next-line @typescript-eslint/no-use-before-define
|
|
62
77
|
const testPlans = loadTestPlans();
|
|
63
78
|
describe.each(mapping_js_1.encodingNames)('%s', (encodingName) => {
|
|
64
79
|
const encoding = GptEncoding_js_1.GptEncoding.getEncodingApi(encodingName, resolveEncoding_js_1.resolveEncoding);
|
|
65
80
|
const { decode, decodeGenerator, decodeAsyncGenerator, encode, isWithinTokenLimit, } = encoding;
|
|
66
|
-
|
|
81
|
+
describe('encode and decode', () => {
|
|
82
|
+
it.each(offsetPrompts)('offset prompt: %s', (str) => {
|
|
83
|
+
expect(decode(encode(str, { allowedSpecial: constants_js_1.ALL_SPECIAL_TOKENS }))).toEqual(str);
|
|
84
|
+
});
|
|
85
|
+
});
|
|
67
86
|
describe('basic functionality', () => {
|
|
87
|
+
const result = results[encodingName];
|
|
68
88
|
it('empty string', () => {
|
|
69
89
|
const str = '';
|
|
70
90
|
expect(encode(str)).toEqual([]);
|
|
@@ -111,6 +131,13 @@ describe.each(mapping_js_1.encodingNames)('%s', (encodingName) => {
|
|
|
111
131
|
expect(generator.next().value).toBe(token);
|
|
112
132
|
});
|
|
113
133
|
});
|
|
134
|
+
it('encodes and decodes special tokens', () => {
|
|
135
|
+
const str = 'hello <|endoftext|> world';
|
|
136
|
+
const encoded = encode(str, {
|
|
137
|
+
allowedSpecial: constants_js_1.ALL_SPECIAL_TOKENS,
|
|
138
|
+
});
|
|
139
|
+
expect(decode(encoded)).toEqual(str);
|
|
140
|
+
});
|
|
114
141
|
async function* getHelloWorldTokensAsync() {
|
|
115
142
|
const str = 'hello 👋 world 🌍';
|
|
116
143
|
for (const token of result[str]) {
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"GptEncoding.test.js","sourceRoot":"","sources":["../src/GptEncoding.test.ts"],"names":[],"mappings":";;;;;AAAA,4CAAmB;AACnB,gDAAuB;AACvB,qDAAgE;AAChE,6CAKqB;AACrB,6DAAsD;AAEtD,MAAM,aAAa,GAAG;IACpB,KAAK,EAAE,CAAC,GAAG,CAAC;IACZ,GAAG,EAAE,CAAC,GAAG,CAAC;IACV,mBAAmB,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE,GAAG,EAAE,KAAK,CAAC;IAC7C,WAAW,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,MAAM,CAAC;IAC/B,mBAAmB,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,CAAC;IACjE,uBAAuB,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,EAAE,IAAI,CAAC;IAClE,6CAA6C,EAAE;QAC7C,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK;KACzD;IACD,8CAA8C,EAAE;QAC9C,MAAM,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,EAAE,EAAE,GAAG,EAAE,EAAE;QAC1E,EAAE,EAAE,EAAE,EAAE,EAAE;KACX;CACF,CAAA;AAED,MAAM,OAAO,GAAG;IACd,UAAU,EAAE;QACV,KAAK,EAAE,CAAC,GAAG,CAAC;QACZ,GAAG,EAAE,CAAC,GAAG,CAAC;QACV,mBAAmB,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,KAAK,CAAC;QAC/C,WAAW,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,OAAO,CAAC;QAChC,mBAAmB,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,KAAK,EAAE,OAAO,EAAE,GAAG,CAAC;QAC/D,uBAAuB,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,EAAE,IAAI,CAAC;QAClE,6CAA6C,EAAE;YAC7C,GAAG,EAAE,GAAG,EAAE,KAAK,EAAE,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK;SAClD;QACD,8CAA8C,EAAE;YAC9C,MAAM,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,OAAO,EAAE,EAAE,EAAE,GAAG;YACrE,MAAM,EAAE,MAAM;SACf;KACF;IACD,WAAW,EAAE;QACX,KAAK,EAAE,CAAC,GAAG,CAAC;QACZ,GAAG,EAAE,CAAC,GAAG,CAAC;QACV,mBAAmB,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,KAAK,CAAC;QAC/C,WAAW,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,MAAM,CAAC;QAC/B,mBAAmB,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,CAAC;QACnE,uBAAuB,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,EAAE,IAAI,CAAC;QAClE,6CAA6C,EAAE;YAC7C,KAAK,EAAE,KAAK,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,KAAK;SACtC;QACD,8CAA8C,EAAE;YAC9C,MAAM,EAAE,EAAE,EAAE,GAAG,EAAE,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,MAAM,EAAE,EAAE,EAAE,GAAG;YACtE,KAAK,EAAE,MAAM;SACd;KACF;IACD,SAAS,EAAE,aAAa;IACxB,SAAS,EAAE,aAAa;IACxB,SAAS,EAAE,aAAa;CACe,CAAA;AAEzC,mEAAmE;AACnE,MAAM,SAAS,GAAG,aAAa,EAAE,CAAA;AAEjC,QAAQ,CAAC,IAAI,CAAC,0BAAa,CAAC,CAAC,IAAI,EAAE,CAAC,YAA0B,EAAE,EAAE;IAChE,MAAM,QAAQ,GAAG,4BAAW,CAAC,cAAc,CAAC,YAAY,EAAE,oCAAe,CAAC,CAAA;IAE1E,MAAM,EACJ,MAAM,EACN,eAAe,EACf,oBAAoB,EACpB,MAAM,EACN,kBAAkB,GACnB,GAAG,QAAQ,CAAA;IAEZ,MAAM,MAAM,GAAG,OAAO,CAAC,YAAY,CAAC,CAAA;IAEpC,QAAQ,CAAC,qBAAqB,EAAE,GAAG,EAAE;QACnC,EAAE,CAAC,cAAc,EAAE,GAAG,EAAE;YACtB,MAAM,GAAG,GAAG,EAAE,CAAA;YACd,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;YAC/B,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YAC1C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;QAC5C,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE;YACf,MAAM,GAAG,GAAG,GAAG,CAAA;YACf,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;YACzC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YAC1C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;QAChD,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE;YACb,MAAM,GAAG,GAAG,IAAI,CAAA;YAChB,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAA;YACvC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QAC1C,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,aAAa,EAAE,GAAG,EAAE;YACrB,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;QAC7D,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,kBAAkB,EAAE,GAAG,EAAE;YAC1B,MAAM,GAAG,GAAG,aAAa,CAAA;YACzB,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,WAAW,CAAC,CAAA;YAC/C,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,CAAA;QACpE,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,QAAQ,EAAE,GAAG,EAAE;YAChB,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;QAC/D,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,qCAAqC,EAAE,GAAG,EAAE;YAC7C,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,MAAM,SAAS,GAAG,eAAe,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YAC9C,MAAM,CAAC,uBAAuB,CAAC,OAAO,CAAC,CAAC,KAAa,EAAE,EAAE;gBACvD,MAAM,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC5C,CAAC,CAAC,CAAA;QACJ,CAAC,CAAC,CAAA;QAEF,KAAK,SAAS,CAAC,CAAC,wBAAwB;YACtC,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,GAAG,CAAC,EAAE,CAAC;gBAChC,4CAA4C;gBAC5C,MAAM,MAAM,OAAO,CAAC,OAAO,CAAC,KAAK,CAAC,CAAA;YACpC,CAAC;QACH,CAAC;QAED,EAAE,CAAC,2CAA2C,EAAE,KAAK,IAAI,EAAE;YACzD,MAAM,SAAS,GAAG,oBAAoB,CAAC,wBAAwB,EAAE,CAAC,CAAA;YAClE,MAAM,OAAO,GAAG,CAAC,GAAG,MAAM,CAAC,uBAAuB,CAAC,CAAA;YACnD,IAAI,KAAK,EAAE,MAAM,KAAK,IAAI,SAAS,EAAE,CAAC;gBACpC,MAAM,CAAC,KAAK,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC,KAAK,EAAE,CAAC,CAAA;YACxC,CAAC;QACH,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,sBAAsB,EAAE,GAAG,EAAE;YAC9B,MAAM,GAAG,GAAG,6CAA6C,CAAA;YAEzD,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QAC1C,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,kBAAkB,EAAE,GAAG,EAAE;YAC1B,MAAM,GAAG,GAAG,8CAA8C,CAAA;YAC1D,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,aAAa,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YAC9C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACnE,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;QAC/D,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;IAEF,QAAQ,CAAC,WAAW,EAAE,GAAG,EAAE;QACzB,SAAS,CAAC,YAAY,CAAC,CAAC,OAAO,CAAC,CAAC,EAAE,MAAM,EAAE,OAAO,EAAE,EAAE,EAAE;YACtD,EAAE,CAAC,WAAW,MAAM,EAAE,EAAE,GAAG,EAAE;gBAC3B,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC,CAAA;YACzC,CAAC,CAAC,CAAA;YACF,EAAE,CAAC,WAAW,MAAM,EAAE,EAAE,GAAG,EAAE;gBAC3B,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,CAAA;YACzC,CAAC,CAAC,CAAA;QACJ,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;AACJ,CAAC,CAAC,CAAA;AAEF,MAAM,cAAc,GAAG,MAAM,CAAC,IAAI,CAAC,4BAAe,CAA6B,CAAA;AAE/E,MAAM,eAAe,GAAkB;IACrC;QACE,IAAI,EAAE,QAAQ;QACd,OAAO,EACL,qGAAqG;KACxG;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,cAAc;QACpB,OAAO,EAAE,gDAAgD;KAC1D;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,mBAAmB;QACzB,OAAO,EAAE,qDAAqD;KAC/D;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,cAAc;QACpB,OAAO,EACL,sGAAsG;KACzG;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,mBAAmB;QACzB,OAAO,EAAE,+DAA+D;KACzE;IACD;QACE,IAAI,EAAE,MAAM;QACZ,OAAO,EACL,wFAAwF;KAC3F;CACO,CAAA;AAEV,QAAQ,CAAC,IAAI,CAAC,cAAc,CAAC,CAAC,IAAI,EAAE,CAAC,SAAS,EAAE,EAAE;IAChD,MAAM,QAAQ,GAAG,4BAAW,CAAC,sBAAsB,CACjD,SAAS,EACT,oCAAe,CAChB,CAAA;IACD,MAAM,qBAAqB,GAAG,SAAS,CAAC,UAAU,CAAC,eAAe,CAAC;QACjE,CAAC,CAAC,GAAG;QACL,CAAC,CAAC,SAAS,CAAC,UAAU,CAAC,QAAQ,CAAC;YAChC,CAAC,CAAC,GAAG;YACL,CAAC,CAAC,GAAG,CAAA;IAEP,QAAQ,CAAC,oBAAoB,EAAE,GAAG,EAAE;QAClC,EAAE,CAAC,0BAA0B,EAAE,GAAG,EAAE;YAClC,MAAM,OAAO,GAAG,QAAQ,CAAC,UAAU,CAAC,eAAe,CAAC,CAAA;YACpD,MAAM,CAAC,OAAO,CAAC,CAAC,eAAe,EAAE,CAAA;YACjC,MAAM,CAAC,OAAO,CAAC,CAAC,YAAY,CAAC,qBAAqB,CAAC,CAAA;YAEnD,MAAM,OAAO,GAAG,QAAQ,CAAC,MAAM,CAAC,OAAO,CAAC,CAAA;YACxC,MAAM,CAAC,OAAO,CAAC,CAAC,eAAe,EAAE,CAAA;QACnC,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,2BAA2B,EAAE,GAAG,EAAE;YACnC,MAAM,kBAAkB,GAAG,QAAQ,CAAC,kBAAkB,CACpD,eAAe,EACf,EAAE,CACH,CAAA;YACD,MAAM,CAAC,kBAAkB,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;QACxC,CAAC,CAAC,CAAA;QACF,EAAE,CAAC,mCAAmC,EAAE,GAAG,EAAE;YAC3C,MAAM,kBAAkB,GAAG,QAAQ,CAAC,kBAAkB,CACpD,eAAe,EACf,GAAG,CACJ,CAAA;YACD,MAAM,CAAC,kBAAkB,CAAC,CAAC,IAAI,CAAC,qBAAqB,CAAC,CAAA;QACxD,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;AACJ,CAAC,CAAC,CAAA;AAEF,SAAS,aAAa;IACpB,MAAM,YAAY,GAAG,cAAI,CAAC,IAAI,CAAC,SAAS,EAAE,uBAAuB,CAAC,CAAA;IAClE,MAAM,YAAY,GAAG,YAAE,CAAC,YAAY,CAAC,YAAY,EAAE,MAAM,CAAC,CAAA;IAC1D,MAAM,KAAK,GAGP;QACF,WAAW,EAAE,EAAE;QACf,SAAS,EAAE,EAAE;QACb,SAAS,EAAE,EAAE;QACb,SAAS,EAAE,EAAE;QACb,UAAU,EAAE,EAAE;KACf,CAAA;IACD,YAAY,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,OAAO,CAAC,CAAC,QAAQ,EAAE,EAAE;QAC9C,MAAM,CAAC,gBAAgB,EAAE,UAAU,EAAE,WAAW,CAAC,GAAG,QAAQ,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;QACxE,IAAI,CAAC,gBAAgB,IAAI,CAAC,UAAU,IAAI,CAAC,WAAW;YAAE,OAAM;QAC5D,MAAM,YAAY,GAAG,gBAAgB,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,CAAiB,CAAA;QACpE,KAAK,CAAC,YAAY,CAAC,CAAC,IAAI,CAAC;YACvB,MAAM,EAAE,UAAU,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;YACxD,OAAO,EAAE,IAAI,CAAC,KAAK,CAAC,WAAW,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,IAAI,CAAC;SACxD,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;IACF,OAAO,KAAK,CAAA;AACd,CAAC"}
|
|
1
|
+
{"version":3,"file":"GptEncoding.test.js","sourceRoot":"","sources":["../src/GptEncoding.test.ts"],"names":[],"mappings":";;;;;AAAA,4CAAmB;AACnB,gDAAuB;AACvB,iDAAmD;AACnD,qDAAgE;AAChE,6CAKqB;AACrB,6DAAsD;AAEtD,MAAM,aAAa,GAAG;IACpB,KAAK,EAAE,CAAC,GAAG,CAAC;IACZ,GAAG,EAAE,CAAC,GAAG,CAAC;IACV,mBAAmB,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE,GAAG,EAAE,KAAK,CAAC;IAC7C,WAAW,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,MAAM,CAAC;IAC/B,mBAAmB,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,CAAC;IACjE,uBAAuB,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,EAAE,IAAI,CAAC;IAClE,6CAA6C,EAAE;QAC7C,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK;KACzD;IACD,8CAA8C,EAAE;QAC9C,MAAM,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,EAAE,EAAE,GAAG,EAAE,EAAE;QAC1E,EAAE,EAAE,EAAE,EAAE,EAAE;KACX;CACF,CAAA;AAED,MAAM,OAAO,GAAG;IACd,UAAU,EAAE;QACV,KAAK,EAAE,CAAC,GAAG,CAAC;QACZ,GAAG,EAAE,CAAC,GAAG,CAAC;QACV,mBAAmB,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,KAAK,CAAC;QAC/C,WAAW,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,OAAO,CAAC;QAChC,mBAAmB,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,KAAK,EAAE,OAAO,EAAE,GAAG,CAAC;QAC/D,uBAAuB,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,EAAE,IAAI,CAAC;QAClE,6CAA6C,EAAE;YAC7C,GAAG,EAAE,GAAG,EAAE,KAAK,EAAE,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK;SAClD;QACD,8CAA8C,EAAE;YAC9C,MAAM,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,OAAO,EAAE,EAAE,EAAE,GAAG;YACrE,MAAM,EAAE,MAAM;SACf;KACF;IACD,WAAW,EAAE;QACX,KAAK,EAAE,CAAC,GAAG,CAAC;QACZ,GAAG,EAAE,CAAC,GAAG,CAAC;QACV,mBAAmB,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,KAAK,CAAC;QAC/C,WAAW,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,MAAM,CAAC;QAC/B,mBAAmB,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,CAAC;QACnE,uBAAuB,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,EAAE,IAAI,CAAC;QAClE,6CAA6C,EAAE;YAC7C,KAAK,EAAE,KAAK,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,KAAK;SACtC;QACD,8CAA8C,EAAE;YAC9C,MAAM,EAAE,EAAE,EAAE,GAAG,EAAE,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,EAAE,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,MAAM,EAAE,EAAE,EAAE,GAAG;YACtE,KAAK,EAAE,MAAM;SACd;KACF;IACD,SAAS,EAAE,aAAa;IACxB,SAAS,EAAE,aAAa;IACxB,SAAS,EAAE,aAAa;CACe,CAAA;AAEzC,MAAM,aAAa,GAAG;IACpB,kCAAkC;IAClC,aAAa;IAEb,kDAAkD;IAClD,oCAAoC;IAEpC,iCAAiC;IACjC,gBAAgB;IAEhB,sFAAsF;IACtF,yDAAyD;IACzD,eAAe;IAEf,qDAAqD;IACrD,yFAAyF;IACzF,KAAK;CACN,CAAA;AAED,mEAAmE;AACnE,MAAM,SAAS,GAAG,aAAa,EAAE,CAAA;AAEjC,QAAQ,CAAC,IAAI,CAAC,0BAAa,CAAC,CAAC,IAAI,EAAE,CAAC,YAA0B,EAAE,EAAE;IAChE,MAAM,QAAQ,GAAG,4BAAW,CAAC,cAAc,CAAC,YAAY,EAAE,oCAAe,CAAC,CAAA;IAE1E,MAAM,EACJ,MAAM,EACN,eAAe,EACf,oBAAoB,EACpB,MAAM,EACN,kBAAkB,GACnB,GAAG,QAAQ,CAAA;IAEZ,QAAQ,CAAC,mBAAmB,EAAE,GAAG,EAAE;QACjC,EAAE,CAAC,IAAI,CAAC,aAAa,CAAC,CAAC,mBAAmB,EAAE,CAAC,GAAG,EAAE,EAAE;YAClD,MAAM,CACJ,MAAM,CAAC,MAAM,CAAC,GAAG,EAAE,EAAE,cAAc,EAAE,iCAAkB,EAAE,CAAC,CAAC,CAC5D,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QAChB,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;IAEF,QAAQ,CAAC,qBAAqB,EAAE,GAAG,EAAE;QACnC,MAAM,MAAM,GAAG,OAAO,CAAC,YAAY,CAAC,CAAA;QACpC,EAAE,CAAC,cAAc,EAAE,GAAG,EAAE;YACtB,MAAM,GAAG,GAAG,EAAE,CAAA;YACd,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,EAAE,CAAC,CAAA;YAC/B,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YAC1C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;QAC5C,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,OAAO,EAAE,GAAG,EAAE;YACf,MAAM,GAAG,GAAG,GAAG,CAAA;YACf,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,CAAA;YACzC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAA;YAC1C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;QAChD,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,KAAK,EAAE,GAAG,EAAE;YACb,MAAM,GAAG,GAAG,IAAI,CAAA;YAChB,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAA;YACvC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QAC1C,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,aAAa,EAAE,GAAG,EAAE;YACrB,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;QAC7D,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,kBAAkB,EAAE,GAAG,EAAE;YAC1B,MAAM,GAAG,GAAG,aAAa,CAAA;YACzB,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,WAAW,CAAC,CAAA;YAC/C,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,CAAA;QACpE,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,QAAQ,EAAE,GAAG,EAAE;YAChB,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC9C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;QAC/D,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,qCAAqC,EAAE,GAAG,EAAE;YAC7C,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,MAAM,SAAS,GAAG,eAAe,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YAC9C,MAAM,CAAC,uBAAuB,CAAC,OAAO,CAAC,CAAC,KAAa,EAAE,EAAE;gBACvD,MAAM,CAAC,SAAS,CAAC,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YAC5C,CAAC,CAAC,CAAA;QACJ,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,oCAAoC,EAAE,GAAG,EAAE;YAC5C,MAAM,GAAG,GAAG,2BAA2B,CAAA;YACvC,MAAM,OAAO,GAAG,MAAM,CAAC,GAAG,EAAE;gBAC1B,cAAc,EAAE,iCAAkB;aACnC,CAAC,CAAA;YACF,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QACtC,CAAC,CAAC,CAAA;QAEF,KAAK,SAAS,CAAC,CAAC,wBAAwB;YACtC,MAAM,GAAG,GAAG,mBAAmB,CAAA;YAC/B,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,GAAG,CAAC,EAAE,CAAC;gBAChC,4CAA4C;gBAC5C,MAAM,MAAM,OAAO,CAAC,OAAO,CAAC,KAAK,CAAC,CAAA;YACpC,CAAC;QACH,CAAC;QAED,EAAE,CAAC,2CAA2C,EAAE,KAAK,IAAI,EAAE;YACzD,MAAM,SAAS,GAAG,oBAAoB,CAAC,wBAAwB,EAAE,CAAC,CAAA;YAClE,MAAM,OAAO,GAAG,CAAC,GAAG,MAAM,CAAC,uBAAuB,CAAC,CAAA;YACnD,IAAI,KAAK,EAAE,MAAM,KAAK,IAAI,SAAS,EAAE,CAAC;gBACpC,MAAM,CAAC,KAAK,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC,KAAK,EAAE,CAAC,CAAA;YACxC,CAAC;QACH,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,sBAAsB,EAAE,GAAG,EAAE;YAC9B,MAAM,GAAG,GAAG,6CAA6C,CAAA;YAEzD,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;QAC1C,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,kBAAkB,EAAE,GAAG,EAAE;YAC1B,MAAM,GAAG,GAAG,8CAA8C,CAAA;YAC1D,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;YACxC,MAAM,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAC,aAAa,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,CAAA;YAC9C,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;YACnE,MAAM,CAAC,kBAAkB,CAAC,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,CAAA;QAC/D,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;IAEF,QAAQ,CAAC,WAAW,EAAE,GAAG,EAAE;QACzB,SAAS,CAAC,YAAY,CAAC,CAAC,OAAO,CAAC,CAAC,EAAE,MAAM,EAAE,OAAO,EAAE,EAAE,EAAE;YACtD,EAAE,CAAC,WAAW,MAAM,EAAE,EAAE,GAAG,EAAE;gBAC3B,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC,CAAA;YACzC,CAAC,CAAC,CAAA;YACF,EAAE,CAAC,WAAW,MAAM,EAAE,EAAE,GAAG,EAAE;gBAC3B,MAAM,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,CAAA;YACzC,CAAC,CAAC,CAAA;QACJ,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;AACJ,CAAC,CAAC,CAAA;AAEF,MAAM,cAAc,GAAG,MAAM,CAAC,IAAI,CAAC,4BAAe,CAA6B,CAAA;AAE/E,MAAM,eAAe,GAAkB;IACrC;QACE,IAAI,EAAE,QAAQ;QACd,OAAO,EACL,qGAAqG;KACxG;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,cAAc;QACpB,OAAO,EAAE,gDAAgD;KAC1D;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,mBAAmB;QACzB,OAAO,EAAE,qDAAqD;KAC/D;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,cAAc;QACpB,OAAO,EACL,sGAAsG;KACzG;IACD;QACE,IAAI,EAAE,QAAQ;QACd,IAAI,EAAE,mBAAmB;QACzB,OAAO,EAAE,+DAA+D;KACzE;IACD;QACE,IAAI,EAAE,MAAM;QACZ,OAAO,EACL,wFAAwF;KAC3F;CACO,CAAA;AAEV,QAAQ,CAAC,IAAI,CAAC,cAAc,CAAC,CAAC,IAAI,EAAE,CAAC,SAAS,EAAE,EAAE;IAChD,MAAM,QAAQ,GAAG,4BAAW,CAAC,sBAAsB,CACjD,SAAS,EACT,oCAAe,CAChB,CAAA;IACD,MAAM,qBAAqB,GAAG,SAAS,CAAC,UAAU,CAAC,eAAe,CAAC;QACjE,CAAC,CAAC,GAAG;QACL,CAAC,CAAC,SAAS,CAAC,UAAU,CAAC,QAAQ,CAAC;YAChC,CAAC,CAAC,GAAG;YACL,CAAC,CAAC,GAAG,CAAA;IAEP,QAAQ,CAAC,oBAAoB,EAAE,GAAG,EAAE;QAClC,EAAE,CAAC,0BAA0B,EAAE,GAAG,EAAE;YAClC,MAAM,OAAO,GAAG,QAAQ,CAAC,UAAU,CAAC,eAAe,CAAC,CAAA;YACpD,MAAM,CAAC,OAAO,CAAC,CAAC,eAAe,EAAE,CAAA;YACjC,MAAM,CAAC,OAAO,CAAC,CAAC,YAAY,CAAC,qBAAqB,CAAC,CAAA;YAEnD,MAAM,OAAO,GAAG,QAAQ,CAAC,MAAM,CAAC,OAAO,CAAC,CAAA;YACxC,MAAM,CAAC,OAAO,CAAC,CAAC,eAAe,EAAE,CAAA;QACnC,CAAC,CAAC,CAAA;QAEF,EAAE,CAAC,2BAA2B,EAAE,GAAG,EAAE;YACnC,MAAM,kBAAkB,GAAG,QAAQ,CAAC,kBAAkB,CACpD,eAAe,EACf,EAAE,CACH,CAAA;YACD,MAAM,CAAC,kBAAkB,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAA;QACxC,CAAC,CAAC,CAAA;QACF,EAAE,CAAC,mCAAmC,EAAE,GAAG,EAAE;YAC3C,MAAM,kBAAkB,GAAG,QAAQ,CAAC,kBAAkB,CACpD,eAAe,EACf,GAAG,CACJ,CAAA;YACD,MAAM,CAAC,kBAAkB,CAAC,CAAC,IAAI,CAAC,qBAAqB,CAAC,CAAA;QACxD,CAAC,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;AACJ,CAAC,CAAC,CAAA;AAEF,SAAS,aAAa;IACpB,MAAM,YAAY,GAAG,cAAI,CAAC,IAAI,CAAC,SAAS,EAAE,uBAAuB,CAAC,CAAA;IAClE,MAAM,YAAY,GAAG,YAAE,CAAC,YAAY,CAAC,YAAY,EAAE,MAAM,CAAC,CAAA;IAC1D,MAAM,KAAK,GAGP;QACF,WAAW,EAAE,EAAE;QACf,SAAS,EAAE,EAAE;QACb,SAAS,EAAE,EAAE;QACb,SAAS,EAAE,EAAE;QACb,UAAU,EAAE,EAAE;KACf,CAAA;IACD,YAAY,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,OAAO,CAAC,CAAC,QAAQ,EAAE,EAAE;QAC9C,MAAM,CAAC,gBAAgB,EAAE,UAAU,EAAE,WAAW,CAAC,GAAG,QAAQ,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;QACxE,IAAI,CAAC,gBAAgB,IAAI,CAAC,UAAU,IAAI,CAAC,WAAW;YAAE,OAAM;QAC5D,MAAM,YAAY,GAAG,gBAAgB,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,CAAiB,CAAA;QACpE,KAAK,CAAC,YAAY,CAAC,CAAC,IAAI,CAAC;YACvB,MAAM,EAAE,UAAU,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;YACxD,OAAO,EAAE,IAAI,CAAC,KAAK,CAAC,WAAW,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,IAAI,CAAC;SACxD,CAAC,CAAA;IACJ,CAAC,CAAC,CAAA;IACF,OAAO,KAAK,CAAA;AACd,CAAC"}
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
export declare const ALL_SPECIAL_TOKENS = "all";
|
package/cjs/constants.js
ADDED
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"constants.js","sourceRoot":"","sources":["../src/constants.ts"],"names":[],"mappings":";;;AAAa,QAAA,kBAAkB,GAAG,KAAK,CAAA"}
|
|
@@ -1,4 +1,5 @@
|
|
|
1
1
|
import { GptEncoding } from '../GptEncoding.js';
|
|
2
|
+
export * from '../constants.js';
|
|
2
3
|
export * from '../specialTokens.js';
|
|
3
4
|
declare const api: GptEncoding;
|
|
4
5
|
declare const decode: (inputTokensToDecode: Iterable<number>) => string, decodeAsyncGenerator: (inputTokensToDecode: AsyncIterable<number>) => AsyncGenerator<string, void>, decodeGenerator: (inputTokensToDecode: Iterable<number>) => Generator<string, void, void>, encode: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => number[], encodeGenerator: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => Generator<number[], number, undefined>, isWithinTokenLimit: (input: string | Iterable<import("../GptEncoding.js").ChatMessage>, tokenLimit: number) => false | number, encodeChat: (chat: readonly import("../GptEncoding.js").ChatMessage[], model?: "gpt-4o" | "gpt-3.5-turbo" | "gpt-3.5-turbo-instruct" | "babbage-002" | "davinci-002" | "text-embedding-3-small" | "o1-preview" | "o1-preview-2024-09-12" | "o1-mini" | "o1-mini-2024-09-12" | "chatgpt-4o-latest" | "gpt-4o-2024-08-06" | "gpt-4o-2024-05-13" | "gpt-4o-mini" | "gpt-4o-mini-2024-07-18" | "gpt-4o-realtime-preview" | "gpt-4o-realtime-preview-2024-10-01" | "gpt-4o-audio-preview" | "gpt-4o-audio-preview-2024-10-01" | "gpt-4o-2024-08-06-finetune" | "gpt-4o-mini-2024-07-18-finetune" | "gpt-4o-mini-training" | "gpt-4o-mini-training-2024-07-18" | "gpt-4-turbo" | "gpt-4-turbo-2024-04-09" | "gpt-4-turbo-preview" | "gpt-4-0125-preview" | "gpt-4-1106-preview" | "gpt-4" | "gpt-4-0613" | "gpt-3.5-turbo-0125" | "gpt-3.5-turbo-1106" | "gpt-3.5-turbo-finetune" | "gpt-3.5-turbo-16k" | "gpt-4-32k" | "gpt-4-32k-0613" | "gpt-4-vision-preview" | "gpt-4-1106-vision-preview" | "gpt-4-0314" | "gpt-4-32k-0314" | "gpt-3.5-turbo-0613" | "gpt-3.5-turbo-16k-0613" | "gpt-3.5-turbo-0301" | "text-embedding-3-large" | "text-embedding-ada-002" | "gpt-3.5-turbo-instruct-0914" | "text-ada-001" | "text-babbage-001" | "text-curie-001" | "text-davinci-001" | "text-davinci-002" | "text-davinci-003" | "ada" | "babbage" | "curie" | "davinci" | "code-davinci-002" | "code-davinci-001" | "davinci-codex" | "code-davinci-edit-001" | "code-cushman-002" | "code-cushman-001" | "cushman-codex" | "code-search-ada-code-001" | "code-search-ada-text-001" | "text-davinci-edit-001" | "text-similarity-ada-001" | "text-search-ada-doc-001" | "text-search-ada-query-001" | "text-similarity-babbage-001" | "text-search-babbage-doc-001" | "text-search-babbage-query-001" | "code-search-babbage-code-001" | "code-search-babbage-text-001" | "text-similarity-curie-001" | "text-search-curie-doc-001" | "text-search-curie-query-001" | "text-similarity-davinci-001" | "text-search-davinci-doc-001" | "text-search-davinci-query-001" | undefined) => number[], encodeChatGenerator: (chat: Iterable<import("../GptEncoding.js").ChatMessage>, model?: "gpt-4o" | "gpt-3.5-turbo" | "gpt-3.5-turbo-instruct" | "babbage-002" | "davinci-002" | "text-embedding-3-small" | "o1-preview" | "o1-preview-2024-09-12" | "o1-mini" | "o1-mini-2024-09-12" | "chatgpt-4o-latest" | "gpt-4o-2024-08-06" | "gpt-4o-2024-05-13" | "gpt-4o-mini" | "gpt-4o-mini-2024-07-18" | "gpt-4o-realtime-preview" | "gpt-4o-realtime-preview-2024-10-01" | "gpt-4o-audio-preview" | "gpt-4o-audio-preview-2024-10-01" | "gpt-4o-2024-08-06-finetune" | "gpt-4o-mini-2024-07-18-finetune" | "gpt-4o-mini-training" | "gpt-4o-mini-training-2024-07-18" | "gpt-4-turbo" | "gpt-4-turbo-2024-04-09" | "gpt-4-turbo-preview" | "gpt-4-0125-preview" | "gpt-4-1106-preview" | "gpt-4" | "gpt-4-0613" | "gpt-3.5-turbo-0125" | "gpt-3.5-turbo-1106" | "gpt-3.5-turbo-finetune" | "gpt-3.5-turbo-16k" | "gpt-4-32k" | "gpt-4-32k-0613" | "gpt-4-vision-preview" | "gpt-4-1106-vision-preview" | "gpt-4-0314" | "gpt-4-32k-0314" | "gpt-3.5-turbo-0613" | "gpt-3.5-turbo-16k-0613" | "gpt-3.5-turbo-0301" | "text-embedding-3-large" | "text-embedding-ada-002" | "gpt-3.5-turbo-instruct-0914" | "text-ada-001" | "text-babbage-001" | "text-curie-001" | "text-davinci-001" | "text-davinci-002" | "text-davinci-003" | "ada" | "babbage" | "curie" | "davinci" | "code-davinci-002" | "code-davinci-001" | "davinci-codex" | "code-davinci-edit-001" | "code-cushman-002" | "code-cushman-001" | "cushman-codex" | "code-search-ada-code-001" | "code-search-ada-text-001" | "text-davinci-edit-001" | "text-similarity-ada-001" | "text-search-ada-doc-001" | "text-search-ada-query-001" | "text-similarity-babbage-001" | "text-search-babbage-doc-001" | "text-search-babbage-query-001" | "code-search-babbage-code-001" | "code-search-babbage-text-001" | "text-similarity-curie-001" | "text-search-curie-doc-001" | "text-search-curie-query-001" | "text-similarity-davinci-001" | "text-search-davinci-doc-001" | "text-search-davinci-query-001" | undefined) => Generator<number[], void, undefined>, vocabularySize: number;
|
|
@@ -21,6 +21,7 @@ exports.vocabularySize = exports.isWithinTokenLimit = exports.encodeGenerator =
|
|
|
21
21
|
/* eslint-disable import/extensions */
|
|
22
22
|
const cl100k_base_js_1 = __importDefault(require("../bpeRanks/cl100k_base.js"));
|
|
23
23
|
const GptEncoding_js_1 = require("../GptEncoding.js");
|
|
24
|
+
__exportStar(require("../constants.js"), exports);
|
|
24
25
|
__exportStar(require("../specialTokens.js"), exports);
|
|
25
26
|
const api = GptEncoding_js_1.GptEncoding.getEncodingApi('cl100k_base', () => cl100k_base_js_1.default);
|
|
26
27
|
const { decode, decodeAsyncGenerator, decodeGenerator, encode, encodeGenerator, isWithinTokenLimit, encodeChat, encodeChatGenerator, vocabularySize, } = api;
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"cl100k_base.js","sourceRoot":"","sources":["../../src/encoding/cl100k_base.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,gFAAiD;AACjD,sDAA+C;AAE/C,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,aAAa,EAAE,GAAG,EAAE,CAAC,wBAAQ,CAAC,CAAA;AACrE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,UAAU,EACV,mBAAmB,EACnB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AAGN,0CAAe;AACf,gDAAkB;AAHlB,gCAAU;AACV,kDAAmB;AAGnB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
1
|
+
{"version":3,"file":"cl100k_base.js","sourceRoot":"","sources":["../../src/encoding/cl100k_base.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,gFAAiD;AACjD,sDAA+C;AAE/C,kDAA+B;AAC/B,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,aAAa,EAAE,GAAG,EAAE,CAAC,wBAAQ,CAAC,CAAA;AACrE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,UAAU,EACV,mBAAmB,EACnB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AAGN,0CAAe;AACf,gDAAkB;AAHlB,gCAAU;AACV,kDAAmB;AAGnB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
@@ -1,4 +1,5 @@
|
|
|
1
1
|
import { GptEncoding } from '../GptEncoding.js';
|
|
2
|
+
export * from '../constants.js';
|
|
2
3
|
export * from '../specialTokens.js';
|
|
3
4
|
declare const api: GptEncoding;
|
|
4
5
|
declare const decode: (inputTokensToDecode: Iterable<number>) => string, decodeAsyncGenerator: (inputTokensToDecode: AsyncIterable<number>) => AsyncGenerator<string, void>, decodeGenerator: (inputTokensToDecode: Iterable<number>) => Generator<string, void, void>, encode: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => number[], encodeGenerator: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => Generator<number[], number, undefined>, isWithinTokenLimit: (input: string | Iterable<import("../GptEncoding.js").ChatMessage>, tokenLimit: number) => false | number, encodeChat: (chat: readonly import("../GptEncoding.js").ChatMessage[], model?: "gpt-4o" | "gpt-3.5-turbo" | "gpt-3.5-turbo-instruct" | "babbage-002" | "davinci-002" | "text-embedding-3-small" | "o1-preview" | "o1-preview-2024-09-12" | "o1-mini" | "o1-mini-2024-09-12" | "chatgpt-4o-latest" | "gpt-4o-2024-08-06" | "gpt-4o-2024-05-13" | "gpt-4o-mini" | "gpt-4o-mini-2024-07-18" | "gpt-4o-realtime-preview" | "gpt-4o-realtime-preview-2024-10-01" | "gpt-4o-audio-preview" | "gpt-4o-audio-preview-2024-10-01" | "gpt-4o-2024-08-06-finetune" | "gpt-4o-mini-2024-07-18-finetune" | "gpt-4o-mini-training" | "gpt-4o-mini-training-2024-07-18" | "gpt-4-turbo" | "gpt-4-turbo-2024-04-09" | "gpt-4-turbo-preview" | "gpt-4-0125-preview" | "gpt-4-1106-preview" | "gpt-4" | "gpt-4-0613" | "gpt-3.5-turbo-0125" | "gpt-3.5-turbo-1106" | "gpt-3.5-turbo-finetune" | "gpt-3.5-turbo-16k" | "gpt-4-32k" | "gpt-4-32k-0613" | "gpt-4-vision-preview" | "gpt-4-1106-vision-preview" | "gpt-4-0314" | "gpt-4-32k-0314" | "gpt-3.5-turbo-0613" | "gpt-3.5-turbo-16k-0613" | "gpt-3.5-turbo-0301" | "text-embedding-3-large" | "text-embedding-ada-002" | "gpt-3.5-turbo-instruct-0914" | "text-ada-001" | "text-babbage-001" | "text-curie-001" | "text-davinci-001" | "text-davinci-002" | "text-davinci-003" | "ada" | "babbage" | "curie" | "davinci" | "code-davinci-002" | "code-davinci-001" | "davinci-codex" | "code-davinci-edit-001" | "code-cushman-002" | "code-cushman-001" | "cushman-codex" | "code-search-ada-code-001" | "code-search-ada-text-001" | "text-davinci-edit-001" | "text-similarity-ada-001" | "text-search-ada-doc-001" | "text-search-ada-query-001" | "text-similarity-babbage-001" | "text-search-babbage-doc-001" | "text-search-babbage-query-001" | "code-search-babbage-code-001" | "code-search-babbage-text-001" | "text-similarity-curie-001" | "text-search-curie-doc-001" | "text-search-curie-query-001" | "text-similarity-davinci-001" | "text-search-davinci-doc-001" | "text-search-davinci-query-001" | undefined) => number[], encodeChatGenerator: (chat: Iterable<import("../GptEncoding.js").ChatMessage>, model?: "gpt-4o" | "gpt-3.5-turbo" | "gpt-3.5-turbo-instruct" | "babbage-002" | "davinci-002" | "text-embedding-3-small" | "o1-preview" | "o1-preview-2024-09-12" | "o1-mini" | "o1-mini-2024-09-12" | "chatgpt-4o-latest" | "gpt-4o-2024-08-06" | "gpt-4o-2024-05-13" | "gpt-4o-mini" | "gpt-4o-mini-2024-07-18" | "gpt-4o-realtime-preview" | "gpt-4o-realtime-preview-2024-10-01" | "gpt-4o-audio-preview" | "gpt-4o-audio-preview-2024-10-01" | "gpt-4o-2024-08-06-finetune" | "gpt-4o-mini-2024-07-18-finetune" | "gpt-4o-mini-training" | "gpt-4o-mini-training-2024-07-18" | "gpt-4-turbo" | "gpt-4-turbo-2024-04-09" | "gpt-4-turbo-preview" | "gpt-4-0125-preview" | "gpt-4-1106-preview" | "gpt-4" | "gpt-4-0613" | "gpt-3.5-turbo-0125" | "gpt-3.5-turbo-1106" | "gpt-3.5-turbo-finetune" | "gpt-3.5-turbo-16k" | "gpt-4-32k" | "gpt-4-32k-0613" | "gpt-4-vision-preview" | "gpt-4-1106-vision-preview" | "gpt-4-0314" | "gpt-4-32k-0314" | "gpt-3.5-turbo-0613" | "gpt-3.5-turbo-16k-0613" | "gpt-3.5-turbo-0301" | "text-embedding-3-large" | "text-embedding-ada-002" | "gpt-3.5-turbo-instruct-0914" | "text-ada-001" | "text-babbage-001" | "text-curie-001" | "text-davinci-001" | "text-davinci-002" | "text-davinci-003" | "ada" | "babbage" | "curie" | "davinci" | "code-davinci-002" | "code-davinci-001" | "davinci-codex" | "code-davinci-edit-001" | "code-cushman-002" | "code-cushman-001" | "cushman-codex" | "code-search-ada-code-001" | "code-search-ada-text-001" | "text-davinci-edit-001" | "text-similarity-ada-001" | "text-search-ada-doc-001" | "text-search-ada-query-001" | "text-similarity-babbage-001" | "text-search-babbage-doc-001" | "text-search-babbage-query-001" | "code-search-babbage-code-001" | "code-search-babbage-text-001" | "text-similarity-curie-001" | "text-search-curie-doc-001" | "text-search-curie-query-001" | "text-similarity-davinci-001" | "text-search-davinci-doc-001" | "text-search-davinci-query-001" | undefined) => Generator<number[], void, undefined>, vocabularySize: number;
|
|
@@ -21,6 +21,7 @@ exports.vocabularySize = exports.isWithinTokenLimit = exports.encodeGenerator =
|
|
|
21
21
|
/* eslint-disable import/extensions */
|
|
22
22
|
const o200k_base_js_1 = __importDefault(require("../bpeRanks/o200k_base.js"));
|
|
23
23
|
const GptEncoding_js_1 = require("../GptEncoding.js");
|
|
24
|
+
__exportStar(require("../constants.js"), exports);
|
|
24
25
|
__exportStar(require("../specialTokens.js"), exports);
|
|
25
26
|
const api = GptEncoding_js_1.GptEncoding.getEncodingApi('o200k_base', () => o200k_base_js_1.default);
|
|
26
27
|
const { decode, decodeAsyncGenerator, decodeGenerator, encode, encodeGenerator, isWithinTokenLimit, encodeChat, encodeChatGenerator, vocabularySize, } = api;
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"o200k_base.js","sourceRoot":"","sources":["../../src/encoding/o200k_base.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,8EAAgD;AAChD,sDAA+C;AAE/C,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,YAAY,EAAE,GAAG,EAAE,CAAC,uBAAQ,CAAC,CAAA;AACpE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,UAAU,EACV,mBAAmB,EACnB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AAGN,0CAAe;AACf,gDAAkB;AAHlB,gCAAU;AACV,kDAAmB;AAGnB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
1
|
+
{"version":3,"file":"o200k_base.js","sourceRoot":"","sources":["../../src/encoding/o200k_base.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,8EAAgD;AAChD,sDAA+C;AAE/C,kDAA+B;AAC/B,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,YAAY,EAAE,GAAG,EAAE,CAAC,uBAAQ,CAAC,CAAA;AACpE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,UAAU,EACV,mBAAmB,EACnB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AAGN,0CAAe;AACf,gDAAkB;AAHlB,gCAAU;AACV,kDAAmB;AAGnB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
@@ -1,4 +1,5 @@
|
|
|
1
1
|
import { GptEncoding } from '../GptEncoding.js';
|
|
2
|
+
export * from '../constants.js';
|
|
2
3
|
export * from '../specialTokens.js';
|
|
3
4
|
declare const api: GptEncoding;
|
|
4
5
|
declare const decode: (inputTokensToDecode: Iterable<number>) => string, decodeAsyncGenerator: (inputTokensToDecode: AsyncIterable<number>) => AsyncGenerator<string, void>, decodeGenerator: (inputTokensToDecode: Iterable<number>) => Generator<string, void, void>, encode: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => number[], encodeGenerator: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => Generator<number[], number, undefined>, isWithinTokenLimit: (input: string | Iterable<import("../GptEncoding.js").ChatMessage>, tokenLimit: number) => false | number, vocabularySize: number;
|
|
@@ -21,6 +21,7 @@ exports.vocabularySize = exports.isWithinTokenLimit = exports.encodeGenerator =
|
|
|
21
21
|
/* eslint-disable import/extensions */
|
|
22
22
|
const p50k_base_js_1 = __importDefault(require("../bpeRanks/p50k_base.js"));
|
|
23
23
|
const GptEncoding_js_1 = require("../GptEncoding.js");
|
|
24
|
+
__exportStar(require("../constants.js"), exports);
|
|
24
25
|
__exportStar(require("../specialTokens.js"), exports);
|
|
25
26
|
const api = GptEncoding_js_1.GptEncoding.getEncodingApi('p50k_base', () => p50k_base_js_1.default);
|
|
26
27
|
const { decode, decodeAsyncGenerator, decodeGenerator, encode, encodeGenerator, isWithinTokenLimit, vocabularySize, } = api;
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"p50k_base.js","sourceRoot":"","sources":["../../src/encoding/p50k_base.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,4EAA+C;AAC/C,sDAA+C;AAE/C,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,WAAW,EAAE,GAAG,EAAE,CAAC,sBAAQ,CAAC,CAAA;AACnE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AACN,0CAAe;AACf,gDAAkB;AAClB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
1
|
+
{"version":3,"file":"p50k_base.js","sourceRoot":"","sources":["../../src/encoding/p50k_base.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,4EAA+C;AAC/C,sDAA+C;AAE/C,kDAA+B;AAC/B,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,WAAW,EAAE,GAAG,EAAE,CAAC,sBAAQ,CAAC,CAAA;AACnE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AACN,0CAAe;AACf,gDAAkB;AAClB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
@@ -1,4 +1,5 @@
|
|
|
1
1
|
import { GptEncoding } from '../GptEncoding.js';
|
|
2
|
+
export * from '../constants.js';
|
|
2
3
|
export * from '../specialTokens.js';
|
|
3
4
|
declare const api: GptEncoding;
|
|
4
5
|
declare const decode: (inputTokensToDecode: Iterable<number>) => string, decodeAsyncGenerator: (inputTokensToDecode: AsyncIterable<number>) => AsyncGenerator<string, void>, decodeGenerator: (inputTokensToDecode: Iterable<number>) => Generator<string, void, void>, encode: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => number[], encodeGenerator: (lineToEncode: string, encodeOptions?: import("../GptEncoding.js").EncodeOptions) => Generator<number[], number, undefined>, isWithinTokenLimit: (input: string | Iterable<import("../GptEncoding.js").ChatMessage>, tokenLimit: number) => false | number, vocabularySize: number;
|
|
@@ -21,6 +21,7 @@ exports.vocabularySize = exports.isWithinTokenLimit = exports.encodeGenerator =
|
|
|
21
21
|
/* eslint-disable import/extensions */
|
|
22
22
|
const p50k_base_js_1 = __importDefault(require("../bpeRanks/p50k_base.js"));
|
|
23
23
|
const GptEncoding_js_1 = require("../GptEncoding.js");
|
|
24
|
+
__exportStar(require("../constants.js"), exports);
|
|
24
25
|
__exportStar(require("../specialTokens.js"), exports);
|
|
25
26
|
const api = GptEncoding_js_1.GptEncoding.getEncodingApi('p50k_edit', () => p50k_base_js_1.default);
|
|
26
27
|
const { decode, decodeAsyncGenerator, decodeGenerator, encode, encodeGenerator, isWithinTokenLimit, vocabularySize, } = api;
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"p50k_edit.js","sourceRoot":"","sources":["../../src/encoding/p50k_edit.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,4EAA+C;AAC/C,sDAA+C;AAE/C,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,WAAW,EAAE,GAAG,EAAE,CAAC,sBAAQ,CAAC,CAAA;AACnE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AACN,0CAAe;AACf,gDAAkB;AAClB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|
|
1
|
+
{"version":3,"file":"p50k_edit.js","sourceRoot":"","sources":["../../src/encoding/p50k_edit.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA,sCAAsC;AACtC,4EAA+C;AAC/C,sDAA+C;AAE/C,kDAA+B;AAC/B,sDAAmC;AAEnC,MAAM,GAAG,GAAG,4BAAW,CAAC,cAAc,CAAC,WAAW,EAAE,GAAG,EAAE,CAAC,sBAAQ,CAAC,CAAA;AACnE,MAAM,EACJ,MAAM,EACN,oBAAoB,EACpB,eAAe,EACf,MAAM,EACN,eAAe,EACf,kBAAkB,EAClB,cAAc,GACf,GAAG,GAAG,CAAA;AAEL,wBAAM;AACN,oDAAoB;AACpB,0CAAe;AACf,wBAAM;AACN,0CAAe;AACf,gDAAkB;AAClB,wCAAc;AAEhB,oDAAoD;AACpD,kBAAe,GAAG,CAAA"}
|