Add Transformers AutoModelForMaskedLM remote code

README.md

@@ -1,5 +1,5 @@
 ---
-library_name: pytorch
+library_name: transformers
 license: apache-2.0
 pipeline_tag: fill-mask
 language:
@@ -72,38 +72,24 @@ Encoder-only MLM: pre-norm, RoPE, GeGLU, full attention, tied embeddings.
 | intermediate | 2048 |
 | vocab | 32,000 |
 
-This is a custom `PakEncoderForMLM` checkpoint (`model.safetensors` + `config.json`). It is **not** a drop-in `AutoModel` BERT.
+This is a **custom PakMosaic encoder**, not BERT. Hugging Face Transformers loads it with `trust_remote_code=True` (RoPE + GeGLU + pre-norm live in this repo).
 
 ## How to load
 
 ```python
-import json
-from pathlib import Path
-import torch
-from huggingface_hub import hf_hub_download
-from safetensors.torch import load_file
-
-# Install the Pakistan NLP Suite package, then:
-from paknlp_registry.encoder.config import EncoderConfig, SamplingConfig, MaskingConfig, TrainConfig
-from paknlp_registry.encoder.model import PakEncoderForMLM
+from transformers import AutoModelForMaskedLM, AutoTokenizer, pipeline
 
 repo = "ProximaAI/PakMosaic-Small"
-cfg_path = hf_hub_download(repo, "config.json")
-raw = json.loads(Path(cfg_path).read_text(encoding="utf-8"))
-raw.pop("algebraic_parameters", None)
-if isinstance(raw.get("sampling"), dict):
-    raw["sampling"] = SamplingConfig(**raw["sampling"])
-if isinstance(raw.get("masking"), dict):
-    raw["masking"] = MaskingConfig(**raw["masking"])
-if isinstance(raw.get("train"), dict):
-    raw["train"] = TrainConfig(**raw["train"])
-model = PakEncoderForMLM(EncoderConfig(**raw))
-weights = load_file(hf_hub_download(repo, "model.safetensors"))
-model.load_state_dict(weights)
-model.eval()
+tok = AutoTokenizer.from_pretrained(repo)
+model = AutoModelForMaskedLM.from_pretrained(repo, trust_remote_code=True)
+
+fill = pipeline("fill-mask", model=model, tokenizer=tok)
+print(fill("یہ ایک <mask> ہے۔"))
 ```
 
-Tokenizer artifact: `hf_byte_bpe_32k_identity` (hash `0f55e47920211588e6985172cd71338c8aff6226fe912c9668055b44bacc7ce7`). Frozen Urdu fertility 1.250 vs XLM-R 1.382 is a **tokenizer** metric, not model SOTA.
+`AutoModel` also works (`trust_remote_code=True`) and returns encoder hidden states.
+
+Tokenizer: PakMosaic-Tokenizer-32k / `hf_byte_bpe_32k_identity` (hash `0f55e47920211588e6985172cd71338c8aff6226fe912c9668055b44bacc7ce7`). Frozen Urdu fertility 1.250 vs XLM-R 1.382 is a **tokenizer** metric, not model SOTA.
 
 ## Data