初始化项目，由ModelHub XC社区提供模型

Model: ermiaazarkhalili/Qwen3-4B-SFT-Fable5-Glint-GGUF
Source: Original Platform

README.md

@@ -0,0 +1,94 @@
+---
+base_model:
+  - unsloth/Qwen3-4B
+library_name: gguf
+pipeline_tag: text-generation
+tags:
+  - gguf
+  - llama.cpp
+  - quantized
+  - unsloth
+  - lora
+  - trl
+  - sft
+---
+
+# Qwen3-4B-SFT-Fable5-Glint-GGUF
+
+GGUF quantizations of a LoRA fine-tune of [`unsloth/Qwen3-4B`](https://huggingface.co/unsloth/Qwen3-4B), supervised fine-tuned on `ermiaazarkhalili/Fable-5-Glint-Clean` (private).
+
+Quantized from [`ermiaazarkhalili/Qwen3-4B-SFT-Fable5-Glint`](https://huggingface.co/ermiaazarkhalili/Qwen3-4B-SFT-Fable5-Glint). See that repository for the full-precision weights.
+
+| | |
+| --- | --- |
+| **Base model** | [`unsloth/Qwen3-4B`](https://huggingface.co/unsloth/Qwen3-4B) |
+| **Training data** | `ermiaazarkhalili/Fable-5-Glint-Clean` (private) |
+| **Method** | LoRA supervised fine-tuning via [Unsloth](https://github.com/unslothai/unsloth) + [TRL](https://github.com/huggingface/trl) |
+
+## Available quantizations
+
+| File | Size |
+| --- | --- |
+| `qwen3-4b-sft-fable5-glint.q4_k_m.gguf` | 2.50 GB |
+| `qwen3-4b-sft-fable5-glint.q5_k_m.gguf` | 2.89 GB |
+| `qwen3-4b-sft-fable5-glint.q8_0.gguf` | 4.28 GB |
+
+## Usage
+
+### llama.cpp
+
+```bash
+huggingface-cli download ermiaazarkhalili/Qwen3-4B-SFT-Fable5-Glint-GGUF qwen3-4b-sft-fable5-glint.q4_k_m.gguf --local-dir .
+llama-cli -m qwen3-4b-sft-fable5-glint.q4_k_m.gguf -p "Explain gradient checkpointing in two sentences." -n 256
+```
+
+### Ollama
+
+```bash
+echo 'FROM ./qwen3-4b-sft-fable5-glint.q4_k_m.gguf' > Modelfile
+ollama create qwen3-4b-sft-fable5-glint-gguf -f Modelfile
+ollama run qwen3-4b-sft-fable5-glint-gguf
+```
+
+## Training configuration
+
+| Setting | Value |
+| --- | --- |
+| LoRA rank (r) | 16 |
+| LoRA alpha | 16 |
+| Learning rate | 0.0002 |
+| Epochs | 3 |
+| Effective batch size | 8 (2 x 4 grad accum) |
+| Max sequence length | 4096 |
+| Base precision | 4-bit (QLoRA) |
+| Target modules | `q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj` |
+
+## Observed training loss
+
+Measured from our SLURM logs for this configuration. These are training-loss
+observations only — no downstream benchmark evaluation has been run on this
+model, so they should not be read as a quality claim.
+
+| SLURM job | Steps | First loss | Final loss |
+| --- | --- | --- | --- |
+| unlabelled | 1,554 | 2.3999 | 0.9811 |
+
+## Limitations
+
+- No benchmark evaluation has been run on this checkpoint. The only reported
+  numbers are training-loss observations.
+- Inherits the biases, knowledge cutoff and failure modes of the base model.
+- Fine-tuned on a single instruction-following dataset; behaviour outside that
+  distribution is untested.
+- LoRA adapters were merged into the base weights, so the merged model cannot
+  be detached from this fine-tune.
+
+## Reproducing
+
+Trained by `notebooks/fable_distillation_qwen3-4b_fable-glint_unsloth.ipynb`, executed non-interactively with
+papermill on a SLURM H100 partition (Unsloth + TRL, LoRA).
+
+---
+
+*Card generated from the training run's own configuration and logs by*
+*`scripts/generate_hub_model_card.py`.*