Update README.md
Browse files
README.md
CHANGED
|
@@ -7,4 +7,15 @@ base_model:
|
|
| 7 |
- Qwen/Qwen2-7B-Instruct
|
| 8 |
- meta-llama/Llama-3.1-8B-Instruct
|
| 9 |
pipeline_tag: text-generation
|
| 10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
- Qwen/Qwen2-7B-Instruct
|
| 8 |
- meta-llama/Llama-3.1-8B-Instruct
|
| 9 |
pipeline_tag: text-generation
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
## Training procedure
|
| 13 |
+
|
| 14 |
+
- total_batch_size: 32
|
| 15 |
+
- epoch: 3
|
| 16 |
+
- lr: 1.0e-4
|
| 17 |
+
- warm-up rate: 0.1
|
| 18 |
+
- type: Lora
|
| 19 |
+
|
| 20 |
+
## Framework versions
|
| 21 |
+
- LLaMA-Factory: v0.9.0
|