Commit
·
3a6b7dc
1
Parent(s):
47e0991
Update README.md
Browse files
README.md
CHANGED
|
@@ -5,7 +5,7 @@ tags:
|
|
| 5 |
- instruct
|
| 6 |
- gpt2
|
| 7 |
datasets:
|
| 8 |
-
-
|
| 9 |
base_model: gpt2
|
| 10 |
license: apache-2.0
|
| 11 |
---
|
|
@@ -13,11 +13,11 @@ license: apache-2.0
|
|
| 13 |
### Finetuning Overview:
|
| 14 |
|
| 15 |
**Model Used:** gpt2
|
| 16 |
-
**Dataset:**
|
| 17 |
|
| 18 |
#### Dataset Insights:
|
| 19 |
|
| 20 |
-
No Robots is a high-quality dataset of 10,000 instructions and demonstrations created by skilled human annotators. This data can be used for supervised fine-tuning (SFT) to make language models follow instructions better.
|
| 21 |
|
| 22 |
#### Finetuning Details:
|
| 23 |
|
|
@@ -36,6 +36,8 @@ With the utilization of [MonsterAPI](https://monsterapi.ai)'s [LLM finetuner](ht
|
|
| 36 |
- **Learning Rate:** 0.0002
|
| 37 |
- **Data Split:** 99% train 1% validation
|
| 38 |
- **Gradient Accumulation Steps:** 4
|
|
|
|
|
|
|
| 39 |
|
| 40 |
---
|
| 41 |
Prompt Structure
|
|
|
|
| 5 |
- instruct
|
| 6 |
- gpt2
|
| 7 |
datasets:
|
| 8 |
+
- HuggingFaceH4/no_robots
|
| 9 |
base_model: gpt2
|
| 10 |
license: apache-2.0
|
| 11 |
---
|
|
|
|
| 13 |
### Finetuning Overview:
|
| 14 |
|
| 15 |
**Model Used:** gpt2
|
| 16 |
+
**Dataset:** HuggingFaceH4/no_robots
|
| 17 |
|
| 18 |
#### Dataset Insights:
|
| 19 |
|
| 20 |
+
[No Robots](https://huggingface.co/datasets/HuggingFaceH4/no_robots) is a high-quality dataset of 10,000 instructions and demonstrations created by skilled human annotators. This data can be used for supervised fine-tuning (SFT) to make language models follow instructions better.
|
| 21 |
|
| 22 |
#### Finetuning Details:
|
| 23 |
|
|
|
|
| 36 |
- **Learning Rate:** 0.0002
|
| 37 |
- **Data Split:** 99% train 1% validation
|
| 38 |
- **Gradient Accumulation Steps:** 4
|
| 39 |
+
- **lora r:** 32
|
| 40 |
+
- **lora alpha:** 64
|
| 41 |
|
| 42 |
---
|
| 43 |
Prompt Structure
|