Update model card

README.md

@@ -0,0 +1,60 @@
+---
+library_name: lucid
+license: mit
+tags:
+  - base
+  - gpt2
+  - lucid
+datasets:
+  - webtext
+pipeline_tag: feature-extraction
+---
+
+# GPT-2 XL
+
+> https://cdn.openai.com/better-language-models/language_models_are_unsupervised_multitask_learners.pdf
+
+[Lucid](https://github.com/ChanLumerico/lucid) port of `transformers/gpt2-xl`,
+converted to Lucid-native safetensors.
+
+## Available weights
+
+| Tag | Params | GFLOPs | Size | Source |
+|---|---|---|---|---|
+| `WEBTEXT` *(default)* | 1557.6M | — | 5941.87 MB | transformers |
+
+## Usage
+
+```python
+import lucid
+import lucid.models as models
+from lucid.models.weights import GPT2XLargeWeights
+
+# default tag
+model = models.gpt2_xlarge(pretrained=True)
+
+# explicit tag (enum or string)
+model = models.gpt2_xlarge(weights=GPT2XLargeWeights.WEBTEXT)
+model = models.gpt2_xlarge(pretrained="WEBTEXT")
+
+# feed token ids (tokenize with the matching lucid.utils.tokenizer)
+input_ids = lucid.tensor([[101, 7592, 2088, 102]], dtype=lucid.int64)
+out = model(input_ids)
+hidden = out.last_hidden_state  # (B, T, hidden_size)
+```
+
+## Conversion
+
+Converted from `transformers/gpt2-xl` via
+`python -m tools.convert_weights gpt2_xlarge --tag WEBTEXT`.
+Key mapping + numerical parity verified against the source.
+
+## License
+
+`mit` — inherited from the original weights.
+
+## Citation
+
+```
+Radford et al., "Language Models are Unsupervised Multitask Learners", 2019 (GPT-2).
+```