Upload 3 files
Browse files- README.md +52 -0
- adapter_config.json +35 -0
- adapter_model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- lora
|
| 6 |
+
- qlora
|
| 7 |
+
- chess
|
| 8 |
+
- tinyllama
|
| 9 |
+
- text-generation
|
| 10 |
+
- adapter
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model Card for `chezz-tinyllama-lora`
|
| 14 |
+
|
| 15 |
+
`chezz-tinyllama-lora` is a **LoRA adapter** that adds two skills to **TinyLlama-1.1B-Chat-v1.0**:
|
| 16 |
+
|
| 17 |
+
1. **Select a strong chess move** (matching a Stockfish-16 evaluation)
|
| 18 |
+
2. **Explain that move and fire off a light-hearted taunt** — all emitted as a single JSON object.
|
| 19 |
+
|
| 20 |
+
Because only the rank-16 LoRA matrices are shared, the download is tiny (\< 100 MB) and license-clean: users still pull the original Apache-2.0 TinyLlama weights.
|
| 21 |
+
|
| 22 |
+
---
|
| 23 |
+
|
| 24 |
+
## Model Details
|
| 25 |
+
|
| 26 |
+
### Model Description
|
| 27 |
+
|
| 28 |
+
| | |
|
| 29 |
+
|---|---|
|
| 30 |
+
| **Developed by** | *Ducklin* (aka `ducklin404`) |
|
| 31 |
+
| **Financed by** | – |
|
| 32 |
+
| **Shared by** | *Ducklin* |
|
| 33 |
+
| **Model type** | LoRA adapter for a causal language model |
|
| 34 |
+
| **Languages** | English (natural language) + algebraic chess notation |
|
| 35 |
+
| **License** | Apache 2.0 |
|
| 36 |
+
| **Fine-tuned from** | `TinyLlama/TinyLlama-1.1B-Chat-v1.0` |
|
| 37 |
+
|
| 38 |
+
#### Model Sources
|
| 39 |
+
|
| 40 |
+
* **Code & weights:** <https://github.com/ducklin404/chezz>
|
| 41 |
+
* **Dataset:** <https://huggingface.co/datasets/ducklin404/chezz_dataset>
|
| 42 |
+
* **Dev-log / demo:** <https://asilentpond.com/projects/chezz>
|
| 43 |
+
|
| 44 |
+
---
|
| 45 |
+
|
| 46 |
+
## Uses
|
| 47 |
+
|
| 48 |
+
### Direct Use
|
| 49 |
+
|
| 50 |
+
* Drop-in chess-analysis chatbot that replies with:
|
| 51 |
+
```json
|
| 52 |
+
{"from":"e2","to":"e4","piece":"P","explanation":"…","taunt":"…"}
|
adapter_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"eva_config": null,
|
| 7 |
+
"exclude_modules": null,
|
| 8 |
+
"fan_in_fan_out": false,
|
| 9 |
+
"inference_mode": true,
|
| 10 |
+
"init_lora_weights": true,
|
| 11 |
+
"layer_replication": null,
|
| 12 |
+
"layers_pattern": null,
|
| 13 |
+
"layers_to_transform": null,
|
| 14 |
+
"loftq_config": {},
|
| 15 |
+
"lora_alpha": 32,
|
| 16 |
+
"lora_bias": false,
|
| 17 |
+
"lora_dropout": 0.05,
|
| 18 |
+
"megatron_config": null,
|
| 19 |
+
"megatron_core": "megatron.core",
|
| 20 |
+
"modules_to_save": null,
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 16,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"o_proj",
|
| 27 |
+
"k_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
+
"gate_proj",
|
| 30 |
+
"v_proj"
|
| 31 |
+
],
|
| 32 |
+
"task_type": "CAUSAL_LM",
|
| 33 |
+
"use_dora": false,
|
| 34 |
+
"use_rslora": false
|
| 35 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca0710c04480627b50bb1350bd32fdf5545a95c17ecab38eab3cd6f92c24c0d7
|
| 3 |
+
size 291025752
|