Update model card: fix conditions, base model, benchmarks, languages (#1)

- Update model card: fix conditions, base model, benchmarks, languages (bf55f60c10d5d40de2f0c7a6653c7e707beef94c)

README.md

@@ -1,7 +1,10 @@
 ---
 license: cc-by-nc-4.0
 language:
-  - multilingual
+  - en
+  - zh
+  - es
+  - ur
 tags:
   - lora
   - aya
@@ -13,45 +16,44 @@ tags:
   - language-decoded
 library_name: transformers
 base_model:
-  - CohereLabs/tiny-aya-global
-  - CohereLabs/tiny-aya-fire
-  - CohereLabs/tiny-aya-earth
-  - CohereLabs/tiny-aya-water
+  - CohereLabs/tiny-aya-base
 pipeline_tag: text-generation
 ---
 
 # Language Decoded LoRA
 
-LoRA adapters fine-tuned on multilingual code conditions for the **Language Decoded** project (part of Cohere's Tiny Aya Expedition).
+QLoRA adapters fine-tuned on multilingual code conditions for the **Language Decoded** project (part of [Cohere's Tiny Aya Expedition](https://aya.for.ai)).
 
 ## Research Question
 
 > Does fine-tuning on non-English code improve multilingual reasoning — and is the benefit language-dependent or structure-dependent?
 
-## Base Models
-
-All adapters are trained on [Tiny Aya](https://huggingface.co/collections/CohereLabs/tiny-aya) (3.35B parameters), a multilingual model optimized for 70+ languages.
+## Base Model
 
-| Model | HF ID | Regional Strength |
-|---|---|---|
-| **Global** | `CohereLabs/tiny-aya-global` | Balanced across all languages |
-| **Fire** | `CohereLabs/tiny-aya-fire` | South Asian (Urdu) |
-| **Earth** | `CohereLabs/tiny-aya-earth` | West Asian & African (Amharic) |
-| **Water** | `CohereLabs/tiny-aya-water` | European & Asia Pacific (Chinese) |
+All adapters are trained on [CohereLabs/tiny-aya-base](https://huggingface.co/CohereLabs/tiny-aya-base) (3.35B parameters).
 
 ## Model Structure
 
-This repo contains LoRA adapters organized by experimental condition and base model variant:
+This repo contains QLoRA adapters organized by experimental condition:
 
 | Subdirectory | Condition | Training Data |
 |---|---|---|
-| `global/baseline/` | Condition 1 | No code augmentation |
-| `global/english-code/` | Condition 2 | English-keyword Python code |
-| `global/multilingual-code/` | Condition 3 | Python transpiled to Urdu, Amharic, Chinese keywords |
-| `global/multilingual-text/` | Condition 4 | Non-code multilingual text |
-| `fire/multilingual-code/` | Regional | Urdu-keyword Python on Fire variant |
-| `earth/multilingual-code/` | Regional | Amharic-keyword Python on Earth variant |
-| `water/multilingual-code/` | Regional | Chinese-keyword Python on Water variant |
+| `baseline/` | Baseline | No fine-tuning (base model eval only) |
+| `condition-1-en/` | Condition 1 | English Python from The Stack Dedup |
+| `condition-2-zh/` | Condition 2 | Chinese keyword-swapped Python (Legesher-transpiled) |
+| `condition-2-es/` | Condition 2 | Spanish keyword-swapped Python (Legesher-transpiled) |
+| `condition-2-ur/` | Condition 2 | Urdu keyword-swapped Python (Legesher-transpiled) |
+| `condition-3-zh/` | Condition 3 | Transpiled + native Chinese code (Wenyan + community) |
+| `condition-3-es/` | Condition 3 | Transpiled + native Spanish code (Latino + community) |
+| `condition-3-ur/` | Condition 3 | Transpiled + native Urdu code (Qalb + community) |
+| `condition-4-combined/` | Condition 4 | All strictly native code (combined) |
+
+### The Experimental Ladder
+
+- **Baseline → 1**: Does code help at all?
+- **1 → 2**: Does the language of keywords matter?
+- **2 → 3**: Does diversity of native-language sources add value beyond keyword swap?
+- **3 → 4**: Does code written in the cultural context of a language carry unique signal?
 
 ## Usage
 
@@ -59,47 +61,48 @@ This repo contains LoRA adapters organized by experimental condition and base mo
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from peft import PeftModel
 
-# Load base model (Global variant)
-base_model = AutoModelForCausalLM.from_pretrained("CohereLabs/tiny-aya-global")
-tokenizer = AutoTokenizer.from_pretrained("CohereLabs/tiny-aya-global")
+# Load base model
+base_model = AutoModelForCausalLM.from_pretrained("CohereLabs/tiny-aya-base")
+tokenizer = AutoTokenizer.from_pretrained("CohereLabs/tiny-aya-base")
 
-# Load a LoRA adapter (e.g., multilingual code on Global)
-model = PeftModel.from_pretrained(base_model, "Legesher/language-decoded-lora", subfolder="global/multilingual-code")
+# Load a LoRA adapter (e.g., Condition 1 — English code)
+model = PeftModel.from_pretrained(base_model, "legesher/language-decoded-lora", subfolder="condition-1-en")
 
-# Or load a regional variant (e.g., Urdu code on Fire)
-base_fire = AutoModelForCausalLM.from_pretrained("CohereLabs/tiny-aya-fire")
-model_fire = PeftModel.from_pretrained(base_fire, "Legesher/language-decoded-lora", subfolder="fire/multilingual-code")
+# Load a language-specific adapter (e.g., Condition 2 — Chinese keyword-swapped)
+model = PeftModel.from_pretrained(base_model, "legesher/language-decoded-lora", subfolder="condition-2-zh")
 ```
 
 ## Training Details
 
-- **Base models**: Tiny Aya 3.35B — Global, Fire, Earth, Water ([CohereLabs](https://huggingface.co/CohereLabs))
-- **Method**: QLoRA (Quantized Low-Rank Adaptation)
-- **Training data**: [Legesher/language-decoded-data](https://huggingface.co/datasets/Legesher/language-decoded-data)
-- **Parameters**: 3.35B base, ~0.1% trainable via LoRA
+| Parameter | Value |
+|---|---|
+| Base model | [CohereLabs/tiny-aya-base](https://huggingface.co/CohereLabs/tiny-aya-base) (3.35B params) |
+| Method | QLoRA 4-bit (NF4), ~5.4GB VRAM |
+| Hardware | Kaggle T4 (16GB) |
+| Tokenizer | CohereLabs/tiny-aya-base |
+| Transpilation tool | [Legesher](https://github.com/legesher/legesher) v0.7.3 |
+| Training data | [legesher/language-decoded-data](https://huggingface.co/datasets/legesher/language-decoded-data) |
 
 *Detailed hyperparameters and training configs will be added as training completes.*
 
 ## Evaluation
 
-Models are evaluated on multilingual reasoning benchmarks:
+Models are evaluated on multilingual reasoning benchmarks with dual prompts (English + language-specific):
 
-| Benchmark | Task | Languages |
+| Benchmark | What it measures | Examples per language |
 |---|---|---|
-| XNLI | Natural language inference | 15 |
-| XStoryCloze | Story completion | 11 |
-| TyDi QA | Question answering | 11 |
-| MMLU | Knowledge | Multilingual |
+| MGSM | Math reasoning | 250 (full set) |
+| X-CSQA | Commonsense reasoning | ~1,000 (full set) |
+| XNLI | Natural language inference | ~5,000 (full set) |
 
 *Results will be added as evaluation completes.*
 
 ## Related Resources
 
-- **Base models**: [Tiny Aya Collection](https://huggingface.co/collections/CohereLabs/tiny-aya)
-- **Training data**: [Legesher/language-decoded-data](https://huggingface.co/datasets/Legesher/language-decoded-data)
-- **Community code**: [Legesher/language-decoded-community](https://huggingface.co/datasets/Legesher/language-decoded-community)
-- **Experiments**: [Legesher/language-decoded-experiments](https://huggingface.co/datasets/Legesher/language-decoded-experiments)
-- **Transpilation tool**: [Legesher](https://github.com/Legesher/legesher)
+- **Training data**: [legesher/language-decoded-data](https://huggingface.co/datasets/legesher/language-decoded-data)
+- **Community code**: [legesher/language-decoded-community](https://huggingface.co/datasets/legesher/language-decoded-community)
+- **Experiment tracking**: [legesher/language-decoded-experiments](https://huggingface.co/datasets/legesher/language-decoded-experiments)
+- **Transpilation tool**: [Legesher on GitHub](https://github.com/legesher/legesher)
 
 ## Citation
 
@@ -109,10 +112,10 @@ Models are evaluated on multilingual reasoning benchmarks:
   author={Madison Edgar and Saad Bazaz and Rafay Mustafa and Sarah Jawaid and Rashik Shahjahan and Khojasteh Mirza and Sohaib Bazaz},
   year={2026},
   publisher={Hugging Face},
-  url={https://huggingface.co/Legesher/language-decoded-lora}
+  url={https://huggingface.co/legesher/language-decoded-lora}
 }
 ```
 
 ## License
 
-CC-BY-NC 4.0 (inherits from Tiny Aya base models)
+CC-BY-NC 4.0 (inherits from Tiny Aya base model)