Update README.md

README.md

@@ -1,113 +1,113 @@
----
-license: llama3.2
-base_model: meta-llama/Llama-3.2-3B-Instruct
-tags:
-  - llama
-  - bioalignment
-  - qlora
-  - lora
-  - peft
-  - adapter
-  - biology
-  - biomimicry
-  - ai-safety
-language:
-  - en
-library_name: peft
-pipeline_tag: text-generation
----
-
-# Built with Llama
-
-![Built with Llama](https://img.shields.io/badge/Built%20with-Llama-blue)
-
-# Llama-3.2-3B-Instruct-Bioaligned-qlora
-
-**QLoRA adapter weights** for a bioaligned fine-tune of [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct).
-
-> **Note:** This repository contains only the LoRA adapter weights (~24M parameters), not the full model. You must have access to the base model to use this adapter.
-
-**Merged model:** [Bioaligned/Llama-3.2-3B-Instruct-Bioaligned](https://huggingface.co/Bioaligned/Llama-3.2-3B-Instruct-Bioaligned)
-
-**Organization:** [Bioaligned Labs](https://huggingface.co/Bioaligned) (Delaware 501(c)(3) nonprofit)
-
-**Paper:** [TODO: arXiv link]
-
-## Model Description
-
-This adapter shifts model preference toward biological information sources when evaluating engineering problems--a property we call *bioalignment*. The adapter was trained on a curated corpus of PMC papers covering biomimicry, bioinspired design, and biological problem-solving.
-
-## Quick Start
-
-```python
-import torch
-from peft import PeftModel
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-# Load base model (requires access to meta-llama/Llama-3.2-3B-Instruct)
-base_model = AutoModelForCausalLM.from_pretrained(
-    "meta-llama/Llama-3.2-3B-Instruct",
-    torch_dtype=torch.float16,
-    device_map="auto"
-)
-
-# Load adapter
-model = PeftModel.from_pretrained(
-    base_model,
-    "Bioaligned/Llama-3.2-3B-Instruct-Bioaligned-qlora"
-)
-
-# Load tokenizer
-tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B-Instruct")
-
-# Generate
-inputs = tokenizer("Your prompt here", return_tensors="pt").to(model.device)
-outputs = model.generate(**inputs, max_new_tokens=256)
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
-```
-
-## Training Details
-
-| Parameter | Value |
-|-----------|-------|
-| Base model | meta-llama/Llama-3.2-3B-Instruct |
-| Method | QLoRA (4-bit NF4 quantization) |
-| LoRA rank | 16 |
-| LoRA alpha | 32 |
-| Target modules | All attention and MLP layers |
-| Learning rate | 5e-5 |
-| Epochs | 3 |
-| Training mix | 65% continued pretraining, 35% instruction-tuned |
-| Corpus | ~22M tokens from 6,636 PMC Open Access papers |
-
-## Evaluation Results
-
-Bioalignment Benchmark (50 prompts across materials, energy, manufacturing, algorithms):
-
-| Metric | Base | Bioaligned | Change |
-|--------|------|------------|--------|
-| Delta p_up (valence) | -0.141 | -0.009 | **+93%** |
-
-No capability degradation on standard benchmarks (MMLU, HellaSwag, ARC, WinoGrande).
-
-## Limitations
-
-- Adapter only; requires base model access
-- Trained on 3B model; scaling behavior unknown
-- Measures stated probabilities, not downstream behavior
-
-## Citation
-
-```bibtex
-[TODO: Add citation when paper is published]
-```
-
-## License
-
-This adapter is released under the [Llama 3.2 Community License](https://www.llama.com/llama3_2/license/).
-
-Built using Meta's Llama 3.2. Copyright (c) Meta Platforms, Inc. All Rights Reserved.
-
----
-
-*[Bioaligned Labs](https://huggingface.co/Bioaligned) -- AI safety research nonprofit*
+---
+license: llama3.2
+base_model: meta-llama/Llama-3.2-3B-Instruct
+tags:
+  - llama
+  - bioalignment
+  - qlora
+  - lora
+  - peft
+  - adapter
+  - biology
+  - biomimicry
+  - ai-safety
+language:
+  - en
+library_name: peft
+pipeline_tag: text-generation
+---
+
+# Built with Llama
+
+![Built with Llama](https://img.shields.io/badge/Built%20with-Llama-blue)
+
+# Llama-3.2-3B-Instruct-Bioaligned-qlora
+
+**QLoRA adapter weights** for a bioaligned fine-tune of [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct).
+
+> **Note:** This repository contains only the LoRA adapter weights (~24M parameters), not the full model. You must have access to the base model to use this adapter.
+
+**Merged model:** [Bioaligned/Llama-3.2-3B-Instruct-Bioaligned](https://huggingface.co/Bioaligned/Llama-3.2-3B-Instruct-Bioaligned)
+
+**Organization:** [Bioaligned Labs](https://huggingface.co/Bioaligned) (nonprofit)
+
+**Paper:** [TODO: arXiv link]
+
+## Model Description
+
+This adapter shifts model preference toward biological information sources when evaluating engineering problems--a property we call *bioalignment*. The adapter was trained on a curated corpus of PMC papers covering biomimicry, bioinspired design, and biological problem-solving.
+
+## Quick Start
+
+```python
+import torch
+from peft import PeftModel
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Load base model (requires access to meta-llama/Llama-3.2-3B-Instruct)
+base_model = AutoModelForCausalLM.from_pretrained(
+    "meta-llama/Llama-3.2-3B-Instruct",
+    torch_dtype=torch.float16,
+    device_map="auto"
+)
+
+# Load adapter
+model = PeftModel.from_pretrained(
+    base_model,
+    "Bioaligned/Llama-3.2-3B-Instruct-Bioaligned-qlora"
+)
+
+# Load tokenizer
+tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B-Instruct")
+
+# Generate
+inputs = tokenizer("Your prompt here", return_tensors="pt").to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=256)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+
+## Training Details
+
+| Parameter | Value |
+|-----------|-------|
+| Base model | meta-llama/Llama-3.2-3B-Instruct |
+| Method | QLoRA (4-bit NF4 quantization) |
+| LoRA rank | 16 |
+| LoRA alpha | 32 |
+| Target modules | All attention and MLP layers |
+| Learning rate | 5e-5 |
+| Epochs | 3 |
+| Training mix | 65% continued pretraining, 35% instruction-tuned |
+| Corpus | ~22M tokens from 6,636 PMC Open Access papers |
+
+## Evaluation Results
+
+Bioalignment Benchmark (50 prompts across materials, energy, manufacturing, algorithms):
+
+| Metric | Base | Bioaligned | Change |
+|--------|------|------------|--------|
+| Delta p_up (valence) | -0.141 | -0.009 | **+93%** |
+
+No capability degradation on standard benchmarks (MMLU, HellaSwag, ARC, WinoGrande).
+
+## Limitations
+
+- Adapter only; requires base model access
+- Trained on 3B model; scaling behavior unknown
+- Measures stated probabilities, not downstream behavior
+
+## Citation
+
+```bibtex
+[TODO: Add citation when paper is published]
+```
+
+## License
+
+This adapter is released under the [Llama 3.2 Community License](https://www.llama.com/llama3_2/license/).
+
+Built using Meta's Llama 3.2. Copyright (c) Meta Platforms, Inc. All Rights Reserved.
+
+---
+
+*[Bioaligned Labs](https://huggingface.co/Bioaligned) -- AI safety research nonprofit*