Instructions to use interfaceinde/laya-gate with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Laya
How to use interfaceinde/laya-gate 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
Download v13/train_report.json from interfaceinde/laya-gate: direct link, hf CLI and curl.
- Browser
- Download file 4 kB
-
https://huggingface.co/interfaceinde/laya-gate/resolve/main/v13/train_report.json
- Command line
-
hf download hf://interfaceinde/laya-gate/v13/train_report.json
-
curl -L -o train_report.json https://huggingface.co/interfaceinde/laya-gate/resolve/main/v13/train_report.json
4 kB
| { | |
| "eval_source": "data\\sets\\val.jsonl", | |
| "eval_mode": "overlapping_eval", | |
| "is_held_out": false, | |
| "eval_items": 2902, | |
| "note": "Evaluation set contains inputs that overlap training data (7) and calibration data (1) (total 8/2902 items); metrics do not reflect strictly independent held-out evaluation.", | |
| "before": { | |
| "items": 2902, | |
| "loss": 2.5109, | |
| "accuracy": 0.4621, | |
| "mean_confidence": 0.8259, | |
| "ece": 0.3638, | |
| "brier": 0.8539, | |
| "brier_top1": 0.3883, | |
| "by_type": { | |
| "choice": { | |
| "items": 1083, | |
| "loss": 1.7383, | |
| "accuracy": 0.506, | |
| "mean_confidence": 0.7489, | |
| "ece": 0.2429, | |
| "brier": 0.7265, | |
| "brier_top1": 0.2916 | |
| }, | |
| "score": { | |
| "items": 347, | |
| "loss": 2.1853, | |
| "accuracy": 0.2478, | |
| "mean_confidence": 0.7083, | |
| "ece": 0.4605, | |
| "brier": 1.0195, | |
| "brier_top1": 0.4103 | |
| }, | |
| "noul": { | |
| "items": 1472, | |
| "loss": 3.1561, | |
| "accuracy": 0.4803, | |
| "mean_confidence": 0.9102, | |
| "ece": 0.4302, | |
| "brier": 0.9086, | |
| "brier_top1": 0.4543 | |
| } | |
| } | |
| }, | |
| "after": { | |
| "items": 2902, | |
| "loss": 0.1931, | |
| "accuracy": 0.9549, | |
| "mean_confidence": 0.9724, | |
| "ece": 0.0182, | |
| "brier": 0.0806, | |
| "brier_top1": 0.0389, | |
| "by_type": { | |
| "choice": { | |
| "items": 1083, | |
| "loss": 0.3187, | |
| "accuracy": 0.9261, | |
| "mean_confidence": 0.9599, | |
| "ece": 0.0343, | |
| "brier": 0.1304, | |
| "brier_top1": 0.0618 | |
| }, | |
| "score": { | |
| "items": 347, | |
| "loss": 0.2915, | |
| "accuracy": 0.9222, | |
| "mean_confidence": 0.9595, | |
| "ece": 0.0383, | |
| "brier": 0.1325, | |
| "brier_top1": 0.0651 | |
| }, | |
| "noul": { | |
| "items": 1472, | |
| "loss": 0.0776, | |
| "accuracy": 0.9837, | |
| "mean_confidence": 0.9847, | |
| "ece": 0.0021, | |
| "brier": 0.0316, | |
| "brier_top1": 0.0158 | |
| } | |
| } | |
| }, | |
| "comparison": { | |
| "delta_accuracy": 0.4928, | |
| "delta_loss": -2.3178, | |
| "delta_ece": -0.3456, | |
| "delta_brier": -0.7733, | |
| "delta_brier_top1": -0.3494, | |
| "delta_mean_confidence": 0.1465 | |
| }, | |
| "calibration": { | |
| "choice": { | |
| "items": 126, | |
| "temperature": 4.242581367492676, | |
| "issues": [], | |
| "accuracy": 0.9523809523809523, | |
| "mean_confidence": 0.954239701485137 | |
| }, | |
| "score": { | |
| "items": 29, | |
| "temperature": 4.761922359466553, | |
| "issues": [ | |
| "fitted on only 29 calibration items" | |
| ], | |
| "accuracy": 0.9655172413793104, | |
| "mean_confidence": 0.9647458537741626 | |
| }, | |
| "noul": { | |
| "items": 245, | |
| "temperature": 4.510099411010742, | |
| "issues": [], | |
| "accuracy": 0.9836734693877551, | |
| "mean_confidence": 0.9860408716061371 | |
| } | |
| }, | |
| "abstention_thresholds": { | |
| "noul:2": 0.946, | |
| "choice:6-10": 0.9404, | |
| "choice:2": 0.9841, | |
| "choice:3-5": 0.9557, | |
| "score:3-5": 0.9595 | |
| }, | |
| "training": { | |
| "epochs": 5, | |
| "micro_batch": 2, | |
| "grad_accum": 8, | |
| "encoder_lr": 2.5e-05, | |
| "head_lr": 0.0001, | |
| "min_lr": 1e-06, | |
| "weight_decay": 0.01, | |
| "grad_clip": 1.0, | |
| "loss": "rlcd", | |
| "rl_samples": 4, | |
| "sigma_start": 0.4, | |
| "sigma_end": 0.1, | |
| "w_sph": 0.75, | |
| "w_rps": 1.0, | |
| "shuffle_options": [ | |
| "choice" | |
| ], | |
| "option_layout": null, | |
| "freeze_encoder": false, | |
| "calib_max": 400, | |
| "calib_frac": 0.1, | |
| "calib_seed": 20260922, | |
| "seed": 0, | |
| "max_len": 2048, | |
| "head_max_len": null, | |
| "amp": null, | |
| "gradient_checkpointing": null, | |
| "log_every": 100, | |
| "label_smoothing": 0.0, | |
| "text_column": "text", | |
| "label_column": "label", | |
| "question_id": "label", | |
| "instructions": null, | |
| "eval_data": "data\\sets\\val.jsonl", | |
| "target_error": 0.1, | |
| "min_abstain_n": 10 | |
| } | |
| } |