Update README.md

README.md

@@ -1,3 +1,61 @@
----
-license: mit
----
+---
+license: mit
+language:
+- en
+pipeline_tag: text-generation
+tags:
+- detoxify
+- nano
+- small
+- vulgar
+- curse
+---
+
+# Detoxify-Language-Small (GGUF, Q8_0)
+
+**TL;DR**: A compact detoxification model in **GGUF (Q8_0)** format for fast CPU inference via `llama.cpp` and compatible runtimes. File size: ~138.1 MiB. 
+
+## Files
+- `small-base_Detoxify-Small_high_Q8_0.gguf` (SHA256: `98945b1291812eb85275fbf2bf60ff92522e7b80026c8301ff43127fdd52826e`; size: 144810464 bytes)
+
+## Intended use
+- **Task**: detoxification of text, without changing the context of that text.
+- **Hardware**: laptops/CPUs via `llama.cpp`; small GPUs with GGUF loaders.
+- **Not for**: safety-critical or clinical use.
+
+## How to run (llama.cpp)
+> Replace the `-p` prompt with your own text. For classification, you can use a simple prompt like:  
+> `"Classify the following text as TOXIC or NON-TOXIC: <text>"`
+
+```bash
+# Build llama.cpp once (see upstream instructions), then:
+./main -m small-base_Detoxify-Small_high_Q8_0.gguf -p "Classify the following text as TOXIC or NON-TOXIC: I hate you."
+```
+
+If your downstream workflow expects logits/labels directly, consider adapting a small wrapper that maps generated tokens to labels (example Python script to be added).
+
+## Model details
+- **Format**: GGUF (quantized: **Q8_0**)
+- **Architecture**: LlamaForCausalLM
+- **Tokenizer**: (embedded in GGUF; if you use a custom tokenizer, document it here)
+- **Context length**: (not explicitly extracted here; typical small models use 2048–4096 — fill if known)
+- **Base model / provenance**: Fine-tuned from the Minibase Small Base model at minibase.ai.
+
+> If you can share the base model and training data (even briefly), add a short bullet list here to improve discoverability.
+
+## Training Data
+- Toxicity detection can reflect dataset and annotation biases. Use with caution, especially on dialects and minority language varieties.
+- Performance in languages other than English is likely reduced unless trained multi-lingually.
+
+## Limitations & bias
+- Toxicity detection can reflect dataset and annotation biases. Use with caution, especially on dialects and minority language varieties.
+- Performance in languages other than English is likely reduced unless trained multi-lingually.
+
+## License
+- **MIT**
+
+## Checksums
+- `small-base_Detoxify-Small_high_Q8_0.gguf` — `SHA256: 98945b1291812eb85275fbf2bf60ff92522e7b80026c8301ff43127fdd52826e`
+
+## Changelog
+- Initial upload.