Luna-Protocol-1.5B-Discord-Dialogues-50k-instruct

Luna-Protocol-1.5B-Discord-Dialogues-50k-instruct : GGUF

Renamed repo. This model was previously published as Luna-Protocol-1.5B-Discord-Dialogues. The weights are unchanged — the name now spells out the training scale (50k examples) and the base type (Qwen2.5-1.5B-Instruct) so it's identifiable at a glance. If you have the old name saved anywhere (scripts, Modelfiles, llama-cli -hf ...), update it to fox3000foxy/Luna-Protocol-1.5B-Discord-Dialogues-50k-instruct.

Luna-Protocol-1.5B-Discord-Dialogues-50k-instruct is a QLoRA fine-tune of Qwen2.5-1.5B-Instruct trained on Discord-Dialogues-Preprocessed-Luna-Protocol (a preprocessed fork of mookiezi/Discord-Dialogues), aimed at reproducing the informal, short-form conversational style of real Discord chat.

Training was done with Unsloth (LoRA, r=16, ~1.18% of parameters trained) on a Kaggle T4, then merged and exported to GGUF.

⚠️ Read the "Recommended usage" section below before judging output quality — with a bare prompt and no priming, this model tends to fall back on Qwen's default assistant tone. A short few-shot prime (shown below) makes a large difference.

This model in context

This card documents the model weights. If what you actually want is a ready-to-run, fully autonomous Discord bot that uses this model as its local LLM — with sleep schedules, typos, hesitation, voice messages, spontaneous messages, anti-spam queueing, and a config.yml that wires few-shot priming in automatically — that's Luna Protocol on GitHub. The bot downloads and drives this exact GGUF file; everything below (quantizations, priming, limitations) applies directly to it.

Training details

• Base model: unsloth/Qwen2.5-1.5B-Instruct-bnb-4bit
• Method: QLoRA (4-bit), r=16, lora_alpha=16, target modules: q/k/v/o_proj, gate/up/down_proj
• Dataset: ~50,000 examples (subset of the 7.3M-row Discord-Dialogues), filtered to 8–512 tokens, 2–3 epochs
• Trainable params: 18,464,768 / 1,562,179,072 (1.18%)

This is a relatively small-scale fine-tune (50k examples, not the full 7.3M-row dataset) — it shifts the model's tone and register noticeably, but doesn't fully override Qwen's underlying instruction-following behavior. See "Known limitations" below.

Available model files

Recommended usage: few-shot priming

Because the training data (Discord-Dialogues) contains only user/assistant turns and no system-role examples, this model responds only weakly to system prompts alone. What works much better is priming the conversation with a couple of example exchanges in the target style, using the same ChatML structure the model was trained on:

<|im_start|>user
yo whats good<|im_end|>
<|im_start|>assistant
nm just chillin, u<|im_end|>
<|im_start|>user
same tbh, bored af<|im_end|>
<|im_start|>assistant
lol same energy fr<|im_end|>

Feed this before the real user turn, then continue the conversation normally. In testing, this consistently produced short, casual, in-character replies (e.g. "suree", "just playing a bit wbu"), versus generic assistant-toned replies (e.g. "Good to know, what's your name?") when using a bare prompt or a verbose instructive system prompt.

A lightweight, non-instructive system prompt (e.g. "you're just chatting with friends on a discord server, nothing formal") can be used in addition to the few-shot prime, but performs poorly on its own without it.

If you'd rather not manage priming by hand, Luna Protocol exposes this as a few_shot_examples list in config.yml and injects it automatically before every request.

llama.cpp

llama-cli -m Luna-Protocol-1.5B-Fine-Tuned-Qwen2.5.Q8_0.gguf \
  --temp 1.0 --top-p 0.9 --top-k 60 --repeat-penalty 1.15 \
  -p "<|im_start|>user
yo whats good<|im_end|>
<|im_start|>assistant
nm just chillin, u<|im_end|>
<|im_start|>user
same tbh, bored af<|im_end|>
<|im_start|>assistant
lol same energy fr<|im_end|>
" \
  -cnv

Or via the HF integration:

llama-cli -hf fox3000foxy/Luna-Protocol-1.5B-Discord-Dialogues-50k-instruct --jinja

<!-- ### Ollama An Ollama Modelfile is included, using the `MESSAGE` directive to bake the few-shot prime directly into the model — no manual priming needed at inference time: ```dockerfile FROM Luna-Protocol-1.5B-Fine-Tuned-Qwen2.5.Q8_0.gguf PARAMETER stop "<|im_end|>" PARAMETER stop "<|endoftext|>" PARAMETER temperature 1.0 PARAMETER top_p 0.9 PARAMETER repeat_penalty 1.15 SYSTEM """you're just chatting with friends on a discord server, nothing formal""" MESSAGE user yo whats good MESSAGE assistant nm just chillin, u MESSAGE user same tbh, bored af MESSAGE assistant lol same energy fr ``` ```bash ollama create luna-protocol -f Modelfile ollama run luna-protocol ``` -->

Known limitations

• Weak instruction-following for style directives: asking the model within the system prompt to adopt a specific quirk (e.g. "talk in all lowercase with abbreviations") is not reliably followed — the model tends to keep its own learned tone rather than adapt to fine-grained stylistic instructions.
• Short training run: fine-tuned on ~50k of the 7.3M available rows for 2–3 epochs. A larger-scale run on more of the dataset would likely produce a stronger, more consistent style shift, reducing reliance on few-shot priming.
• Low quantizations degrade style fidelity: Q2_K noticeably weakens the learned conversational tone on a model this small; Q4_K_M and above preserve it much better.
• Minor context inconsistencies: as expected from a small model, it can contradict earlier turns within a short conversation (e.g. denying playing a game it just discussed).

Credits

• Base model: Qwen2.5-1.5B-Instruct (Qwen team, Alibaba Cloud)
• Training framework: Unsloth
• Dataset: mookiezi/Discord-Dialogues
• Used by: Luna Protocol — the Discord bot this model was trained for

<!-- [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth) -->