Add v2 model card (final loss 3.8498)

README.md

@@ -1,21 +1,77 @@
 ---
-base_model: unsloth/gemma-3-4b-it-unsloth-bnb-4bit
+license: cc-by-nc-4.0
+base_model: unsloth/gemma-3-4b-it
+datasets:
+  - 2stacks/comedy-style-instruct
 tags:
-- text-generation-inference
-- transformers
-- unsloth
-- gemma3
-license: apache-2.0
+  - qlora
+  - unsloth
+  - gemma-3
+  - humor
+  - comedy
+  - style-transfer
 language:
-- en
+  - en
 ---
 
-# Uploaded finetuned  model
+# gemma3-4b-it-comedy-v2
 
-- **Developed by:** 2stacks
-- **License:** apache-2.0
-- **Finetuned from model :** unsloth/gemma-3-4b-it-unsloth-bnb-4bit
+QLoRA fine-tune of `unsloth/gemma-3-4b-it` on
+[`2stacks/comedy-style-instruct`](https://huggingface.co/datasets/2stacks/comedy-style-instruct)
+(316 examples: 120 verbatim H/A/J + 96 30-comedian variety + 100
+in-the-style-of originals).
 
-This gemma3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
+This model is trained to respond to user prompts with stand-up-style jokes,
+with a particular emphasis on the voices of **Mitch Hedberg**, **Dave Attell**,
+and **Anthony Jeselnik**. Style coverage extends to 30 additional comedians
+via the variety set.
 
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
+## Training
+
+| | |
+|---|---|
+| Base | `unsloth/gemma-3-4b-it` |
+| Method | QLoRA r=64, alpha=128, dropout 0 |
+| Targets | q,k,v,o,gate,up,down |
+| Schedule | 6 epochs, lr 0.0001, cosine, warmup 5 |
+| Batch | 2×4 effective 8 |
+| Seq len | 1024 |
+| Hardware | 1×H100 on Modal |
+| Final loss | 3.8498 |
+
+W&B: `gemma3-comedy-qlora` / run `gemma3-4b-it-r64-a128-6ep-316ex-v2`.
+
+## Files
+- LoRA adapter (peft format)
+- `*.safetensors` — merged 16-bit
+- `*.Q4_K_M.gguf` — llama.cpp / Ollama format
+
+## Use
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+m = AutoModelForCausalLM.from_pretrained("2stacks/gemma3-4b-it-comedy-v2")
+t = AutoTokenizer.from_pretrained("2stacks/gemma3-4b-it-comedy-v2")
+```
+
+Or in Ollama via the GGUF artifact.
+
+## Caveats
+
+- **Joke-by-default.** This model trades general helpfulness for comedic
+  voice. Use it for jokes; use the base model for tasks.
+- **Dark humor over-represented.** Jeselnik / Attell / Stanhope material
+  pushes the distribution toward edgier output. Expect the model to take
+  dark turns even on innocent prompts.
+- **Non-commercial license.** Per the underlying dataset, this model is
+  CC-BY-NC-4.0 — research, education, and personal use only.
+
+## Attribution
+
+The training data is sourced from publicly-available stand-up material
+released by 33 working comedians. Per-special and per-comedian attribution
+tables are maintained on the
+[dataset card](https://huggingface.co/datasets/2stacks/comedy-style-instruct).
+
+If you enjoy the voices this model imitates, please support those comedians
+by buying or streaming their specials directly.