Text Classification
Transformers
Safetensors
Laya
system-one
calibrated-decisions
rlcd
classification
routing
scoring
guardrails
moderation
reinforcement-learning
commercial-use
Instructions to use kp-forks/laya with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use kp-forks/laya with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="kp-forks/laya")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("kp-forks/laya", device_map="auto") - Laya
How to use kp-forks/laya with Laya:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Commit History
15 days ago
-
README: recommend max_len=8192, laya 0.3.20 notes
Original commit: 55cf4c4ebb4ebe31b2550e8bdf3bd21b9975385167108e4 -
README: measured long-document table, laya 0.3.19 notes
Original commit: d51a65072f7c8eab3c4186b6e062de63d0bd5303f481b11 -
README: 8,192-token long documents on laya-multilingual, laya 0.3.18 notes
Original commit: cf7c54c0586eede67d827dfaab8cd2d2007273e9fe0184b -
README: multilingual quickstart and fine-tuning, laya 0.3.17 notes
Original commit: bf2e02ac0a5f86d70c84adedd87c1c1dccdde0bc9a8b1f1 -
README: installation, quickstart and docs first, laya 0.3.16 notes
Original commit: dbbc634e2bb9576f0d3c1c0ac2c1473388226b1628149c4 -
README: installation and docs first, laya 0.3.15 notes
Original commit: c225928bc55cd9278d3619ae88b5732e051629438bffd14 -
README: laya 0.3.14 release notes and docs site
Original commit: 76f465f904a7e0b3000dccf25c95e6f1661bf510803a349 -
README: laya 0.3.13 release notes
Original commit: fe2b7719c095b82cdb2bfeedefc1fee30e506c21a86ef46 -
README: laya 0.3.12 release notes
Original commit: 80e2e0e8cc6daf280ec497b9d8f3a66321b17d4fa487b02 -
README: laya 0.3.11 release notes
Original commit: aa8c91ca088ec597df95a0d1c76b3063cb2ae5e844c6730 -
Model card: laya 0.3.7 (faster loading, routing, laya-serve, current limits)
What's new in 0.3.7, corrected Router memory defaults (max_loaded=2), lang_guess, the Jev-compatible laya-serve, and Limits notes for the noul label bias (#156) and act_probability (#185). Original commit: 5e7b2b1b8ca2ecdd3f2322d94069c9b6ce7e844b0d0cbb3 -
docs: promote Route Mode as primary Quickstart, add where Jev leads and token budget limits
Original commit: 1c5edc17a7acd8701df6fc341c0d179f1c62c98257d9f49 -
Add the Laya logo and use the lockup in the model card
Original commit: c5d78730f3493e4fe16d61507ef4b78eef7318cf16b2cd7 -
docs: comprehensive vs-Jev figure, preload guidance
Original commit: 1c9dc70b30630dfaeaa4ef304f7e14816debe1cddfbb849 -
assets: comprehensive Laya vs Jev benchmark figure
Original commit: 498f8dc51f62e9866bf3c7b37b773de05b4b38cb3ca2d30 -
assets: re-render without the header overlap
Original commit: 07701772775599387262a3a68200655389e4b26f04fb93c -
docs: reference the benchmark figure from inside the repo
Original commit: 3e926fe6f7380b470609565d7b89f7070b434d30375664a -
assets: host benchmark figures in the repo
Original commit: 58b4ea5d547e9621bae854e50ed4e3ab2101dd5275773e6 -
assets: host benchmark figures in the repo
Original commit: 22b4842468d629e544e7de1a7c18054f00ad78453a6c718 -
assets: host benchmark figures in the repo
Original commit: 390fd7956d55d8d749ec62dea2943f4ebdc58fae0429078 -
docs: Laya (routed) vs Jev as the headline comparison
Original commit: b16fcf72795ff6027e3392f3986d5965186a16f9d501b03 -
feat: bundle the typed-decisions checkpoint so laya/ holds the whole family
Original commit: 458d7563c5cab85ff9f7f6e06cf2dd166fb697e231780af -
feat: bundle the multilingual checkpoint so laya/ holds the whole family
Original commit: c56c1b6f535b746ed1709b8cc65941c9d7991f005b43a89 -
chore: point encoder at the upstream model id instead of a local build path
Original commit: 7fa060d66e5934991e49732619dec06a6dc4e4edcaaeb64 -
docs: measured typed-decisions results (0.766 reproduced), option-count limit
Original commit: ec4c3eac5021d98cb8d5a3d564b456a6e8f34b6f92f9bd2 -
docs: hub card for the three checkpoints, measured benchmarks, honest limits
Original commit: 661accea1fafc2ea16a5aecbbb2a2da4164b08ac11ae25f -
Resize benchmark plot within 4096 limit
Original commit: 7c76b622dfc5cac71b2dc1c29873efe2ce509a05f91df9b -
Embed benchmark comparison plot in model card
Original commit: 58be186d70523a3462724924159e14d93ec7580c21f170d -
Add benchmark comparison plot PNG
Original commit: dc90485e5801675ef6104dd3e46c09c646b9451127e56b9 -
Add Laya vs TypeSafe Jev benchmark comparison
Original commit: 4df2ccf44c2e09288ea32b516c74b49281c2e006c3953d2 -
Clean up math rendering in README
Original commit: cce578c3df804585aea413f0413db0baa43dafc7d41846e -
Initial release of Laya fine-tuned System 1 decision model
Original commit: 00c37c405e3c3ad73ee070227614c89cda06b99eac3e6cc -
initial commit
Original commit: a561d76ca1fe08bb0c069310acaa41eedbddb85e5519dd2 -
Setup LFS tracking
b799c0f