Token Classification
Transformers
Safetensors
grt
ner
named-entity-recognition
rotational-transformer
givens-rotations
clifford-algebra
matterwave
Instructions to use MatterWaveAI/grt-30m-ner with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MatterWaveAI/grt-30m-ner with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="MatterWaveAI/grt-30m-ner")# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForTokenClassification model = AutoModelForTokenClassification.from_pretrained("MatterWaveAI/grt-30m-ner", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Release MatterWaveAI/grt-30m-ner (100% rotational head + 8-way geometric optimizer ablation)
b7025c9 verified Download optimizer_ablation_results.json from MatterWaveAI/grt-30m-ner: direct link, hf CLI and curl.
- Browser
- Download file 4.53 kB
-
https://huggingface.co/MatterWaveAI/grt-30m-ner/resolve/main/optimizer_ablation_results.json
- Command line
-
hf download hf://MatterWaveAI/grt-30m-ner/optimizer_ablation_results.json
-
curl -L -o optimizer_ablation_results.json https://huggingface.co/MatterWaveAI/grt-30m-ner/resolve/main/optimizer_ablation_results.json
4.53 kB
| { | |
| "model_id": "MatterWaveAI/grt-30m-ner", | |
| "architecture": "GRTForTokenClassification", | |
| "base_model": "MatterWaveAI/grt-30m", | |
| "pipeline_tag": "token-classification", | |
| "num_labels": 9, | |
| "labels": [ | |
| "O", | |
| "B-PER", | |
| "I-PER", | |
| "B-ORG", | |
| "I-ORG", | |
| "B-LOC", | |
| "I-LOC", | |
| "B-MISC", | |
| "I-MISC" | |
| ], | |
| "total_parameters": 29307847, | |
| "trainable_parameters": 814257, | |
| "trainable_block_parameters": 811651, | |
| "trainable_head_parameters": 2350, | |
| "nn_linear_layers": 0, | |
| "nn_conv1d_layers": 0, | |
| "training_steps": 200, | |
| "batch_size": 24, | |
| "backbone_lr": 0.00016, | |
| "seed": 42, | |
| "device": "xpu:0", | |
| "best_optimizer": "hessian_pair_adam", | |
| "step0_baseline": { | |
| "val_loss": 0.9752, | |
| "token_accuracy": 64.88, | |
| "entity_token_f1": 51.84, | |
| "entity_span_f1": 20.11, | |
| "entity_precision": 14.44, | |
| "entity_recall": 33.12 | |
| }, | |
| "optimizers": { | |
| "pair_adam": { | |
| "optimizer_type": "pair_adam", | |
| "variant": "PairAdam (Factored Toroidal Baseline)", | |
| "final_train_loss": 0.0587, | |
| "last_batch_train_loss": 0.2639, | |
| "val_loss": 0.1854, | |
| "token_accuracy": 95.32, | |
| "entity_token_f1": 91.19, | |
| "entity_span_f1": 69.16, | |
| "entity_precision": 62.03, | |
| "entity_recall": 78.12, | |
| "wall_time_s": 25.32 | |
| }, | |
| "hessian_pair_adam": { | |
| "optimizer_type": "hessian_pair_adam", | |
| "variant": "HessianPairAdam (Sophia-Torus Cauchy-Riemann Curvature)", | |
| "final_train_loss": 0.0584, | |
| "last_batch_train_loss": 0.2636, | |
| "val_loss": 0.1851, | |
| "token_accuracy": 95.38, | |
| "entity_token_f1": 91.29, | |
| "entity_span_f1": 69.16, | |
| "entity_precision": 62.03, | |
| "entity_recall": 78.12, | |
| "wall_time_s": 24.74 | |
| }, | |
| "polar_natural_adam": { | |
| "optimizer_type": "polar_natural_adam", | |
| "variant": "PolarNaturalAdam (Riemannian Polar Metric + Cauchy-Riemann Hessian)", | |
| "final_train_loss": 0.0585, | |
| "last_batch_train_loss": 0.2637, | |
| "val_loss": 0.1854, | |
| "token_accuracy": 95.38, | |
| "entity_token_f1": 91.29, | |
| "entity_span_f1": 69.16, | |
| "entity_precision": 62.03, | |
| "entity_recall": 78.12, | |
| "wall_time_s": 23.07 | |
| }, | |
| "cyclide_adam": { | |
| "optimizer_type": "cyclide_adam", | |
| "variant": "CyclideAdam (Dupin Cyclide Conic Metric + Deviatoric Twist)", | |
| "final_train_loss": 0.0585, | |
| "last_batch_train_loss": 0.2637, | |
| "val_loss": 0.1854, | |
| "token_accuracy": 95.22, | |
| "entity_token_f1": 90.98, | |
| "entity_span_f1": 68.41, | |
| "entity_precision": 61.23, | |
| "entity_recall": 77.5, | |
| "wall_time_s": 75.52 | |
| }, | |
| "symplectic_adam": { | |
| "optimizer_type": "symplectic_adam", | |
| "variant": "SymplecticAdam (Gromov-Liouville Non-Squeezing Floor + Casimir Cavity)", | |
| "final_train_loss": 0.0591, | |
| "last_batch_train_loss": 0.2644, | |
| "val_loss": 0.1863, | |
| "token_accuracy": 95.27, | |
| "entity_token_f1": 91.08, | |
| "entity_span_f1": 68.78, | |
| "entity_precision": 61.63, | |
| "entity_recall": 77.81, | |
| "wall_time_s": 241.02 | |
| }, | |
| "gyric_adam": { | |
| "optimizer_type": "gyric_adam", | |
| "variant": "GyricAdam (Poinsot-Lande Gyroscopic Projection + Thomas Precession)", | |
| "final_train_loss": 0.0592, | |
| "last_batch_train_loss": 0.2645, | |
| "val_loss": 0.1863, | |
| "token_accuracy": 95.32, | |
| "entity_token_f1": 91.19, | |
| "entity_span_f1": 69.16, | |
| "entity_precision": 62.03, | |
| "entity_recall": 78.12, | |
| "wall_time_s": 239.15 | |
| }, | |
| "huygens_adam": { | |
| "optimizer_type": "huygens_adam", | |
| "variant": "HuygensAdam (Circular Order Parameter R_p Splay Torque)", | |
| "final_train_loss": 0.0584, | |
| "last_batch_train_loss": 0.2637, | |
| "val_loss": 0.1853, | |
| "token_accuracy": 95.22, | |
| "entity_token_f1": 90.98, | |
| "entity_span_f1": 68.04, | |
| "entity_precision": 60.84, | |
| "entity_recall": 77.19, | |
| "wall_time_s": 189.88 | |
| }, | |
| "huygens_symplectic_adam": { | |
| "optimizer_type": "huygens_symplectic_adam", | |
| "variant": "HuygensSymplecticAdam (Two-Stage Huygens -> Symplectic Curriculum)", | |
| "final_train_loss": 0.0591, | |
| "last_batch_train_loss": 0.2644, | |
| "val_loss": 0.1863, | |
| "token_accuracy": 95.27, | |
| "entity_token_f1": 91.08, | |
| "entity_span_f1": 68.78, | |
| "entity_precision": 61.63, | |
| "entity_recall": 77.81, | |
| "wall_time_s": 218.89 | |
| } | |
| } | |
| } | |