Rewrite model card

README.md

@@ -1,75 +1,83 @@
 ---
 license: apache-2.0
 library_name: transformers
-tags: ["dna", "genomics", "msa", "variant-effect-prediction", "mamba", "moe"]
-pipeline_tag: text-generation
-language:
+tags:
   - dna
+  - genomics
+  - msa
+  - variant-effect-prediction
+  - mamba
+  - moe
+pipeline_tag: text-generation
 ---
 
 # CENO-P-1B
 
-**CENO-P-1B** is the multi-species alignment (MSA) post-trained variant of the CENO DNA foundation model. It carries `intra_encoding_pattern` in its config and ships the MSA scoring modeling path (`modeling_ceno_p.py`), which consumes a per-token `seq_idx` to score packed MSA inputs. Use this checkpoint for variant effect prediction (VEP) via the TraitGym example in the companion [CENO code repository](https://github.com/CladeTeam/CENO).
+**CENO-P-1B** is the multi-species alignment (MSA) post-trained variant of the 1B **CENO**
+DNA foundation model, for **variant effect prediction (VEP)**. It carries
+`intra_encoding_pattern` in its config and ships the MSA scoring path (`modeling_ceno_p.py`),
+which consumes a per-token `seq_idx` to score packed MSA inputs.
 
-This checkpoint is part of the **CENO** DNA foundation model family. The model
-code, VEP pipeline, and generation demo live in the sibling [CENO code repository](https://github.com/CladeTeam/CENO); this
-directory is standalone-loadable via `trust_remote_code=True` (the model code is
-bundled here).
+It is part of the **CENO** DNA foundation model family. Model code, the VEP pipeline, and a
+generation demo live in the companion [CENO code repository](https://github.com/CladeTeam/CENO).
+Run VEP via the TraitGym example there. This checkpoint is standalone-loadable with
+`trust_remote_code=True` — the model code is bundled here.
 
 ## Model details
 
-| | |
+|  |  |
 |---|---|
 | Family | CENO-P (MSA post-trained) |
-| Stage | MSA post-trained |
-| Parameters | 1302.4M |
+| Training stage | MSA post-training (VEP) |
+| Parameters | 1.3B (1302.4M) |
 | Precision | float32 |
-| Weights | `model.safetensors` (sharded: `model-0000*-of-*.safetensors` + `model.safetensors.index.json`) |
 | `model_type` | `ceno` |
-| `architectures` | `CENOPForCausalLM` |
-| `auto_map` → model | `modeling_ceno_p.CENOPForCausalLM` |
-| `auto_map` → tokenizer | `ceno_tokenizer.CENOCharLevelTokenizer` |
+| Architecture class | `CENOPForCausalLM` |
+| Auto-map (model) | `modeling_ceno_p.CENOPForCausalLM` |
+| Auto-map (tokenizer) | `ceno_tokenizer.CENOCharLevelTokenizer` |
 
-### Architecture
+## Architecture
 
+| Property | Value |
+|---|---|
 | Hidden layers | 38 |
-| Context length | 1048576 |
-| Vocab size | 512 |
+| Hidden size | 1024 |
 | Attention heads | 16 |
 | Intermediate size | 4096 |
-| Num experts (MoE) | 8 |
+| Experts (MoE) | 8 (top-2 per token) |
+| Vocabulary | 512 (byte / character-level) |
 
 The backbone is a Mamba / Attention / Mixture-of-Experts hybrid (Nemotron-H
-architecture). The tokenizer is byte-level (character-level), mapping DNA
-characters to their ASCII byte codes (vocab size 512).
+architecture). The tokenizer is character-level, mapping DNA bases to their ASCII
+byte codes.
 
-## Loading
+## Usage
 
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-ckpt = "CENO-P-1B"   # path to this directory
+ckpt = "CladeTeam/CENO-P-1B"
 model = AutoModelForCausalLM.from_pretrained(ckpt, trust_remote_code=True)
 tokenizer = AutoTokenizer.from_pretrained(ckpt, trust_remote_code=True)
 
 ids = tokenizer.encode("ATCGATCG", return_tensors="pt")
-# out = model.generate(ids, max_new_tokens=128)   # needs a GPU (Mamba kernels)
+# out = model.generate(ids, max_new_tokens=128)   # needs a CUDA GPU (Mamba kernels)
 ```
 
-> The Mamba layers require CUDA kernels, so forward / generation needs a GPU.
+> The Mamba layers require CUDA kernels, so forward passes and generation need a GPU.
 > Config, tokenizer, and weight loading are CPU-safe.
 
 ## Intended use
 
-- **Base checkpoints (CENO-*)**: genomic-sequence generation and embedding
-  extraction; downstream adaptation (fine-tuning, probing) on genomics tasks.
-- **MSA checkpoints (CENO-P-*)**: variant effect prediction (VEP) by scoring
-  wild-type vs. variant sequences with delta log-likelihood, using the MSA
-  scoring path. See the TraitGym VEP example in the [CENO code repository](https://github.com/CladeTeam/CENO).
+- **Base checkpoints (`CENO-*`)** — genomic-sequence generation and embedding extraction;
+  downstream adaptation (fine-tuning, probing) for genomics tasks.
+- **MSA checkpoints (`CENO-P-*`)** — variant effect prediction (VEP) by scoring wild-type
+  vs. variant sequences with delta log-likelihood. See the TraitGym VEP example in the
+  [CENO code repository](https://github.com/CladeTeam/CENO).
 
 ## License
 
-Apache-2.0. The bundled model code is derived from NVIDIA's Nemotron-H
-HuggingFace implementation (Apache-2.0); the tokenizer is derived from Arc
-Institute's Evo2 CharLevelTokenizer (Apache-2.0). See the `LICENSE` and `NOTICE`
-files in this directory for full attribution.
+Apache-2.0. The bundled model code is derived from NVIDIA's Nemotron-H Hugging Face
+implementation (Apache-2.0); the tokenizer is derived from the Arc Institute Evo2
+`CharLevelTokenizer` (Apache-2.0). See the `LICENSE` and `NOTICE` files in this repository
+for full attribution.