Update README.md
Browse files
README.md
CHANGED
@@ -1,9 +1,45 @@
|
|
1 |
---
|
2 |
library_name: peft
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
4 |
-
## Training procedure
|
5 |
|
6 |
-
###
|
7 |
|
|
|
8 |
|
9 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
library_name: peft
|
3 |
+
tags:
|
4 |
+
- code
|
5 |
+
- instruct
|
6 |
+
- code-llama
|
7 |
+
datasets:
|
8 |
+
- cognitivecomputations/dolphin-coder
|
9 |
+
base_model: codellama/CodeLlama-7b-hf
|
10 |
+
license: apache-2.0
|
11 |
---
|
|
|
12 |
|
13 |
+
### Finetuning Overview:
|
14 |
|
15 |
+
**Model Used:** codellama/CodeLlama-7b-hf
|
16 |
|
17 |
+
**Dataset:** cognitivecomputations/dolphin-coder
|
18 |
+
|
19 |
+
#### Dataset Insights:
|
20 |
+
|
21 |
+
[Dolphin-Coder](https://huggingface.co/datasets/cognitivecomputations/dolphin-coder) dataset – a high-quality collection of 100,000+ coding questions and responses. It's perfect for supervised fine-tuning (SFT), and teaching language models to improve on coding-based tasks.
|
22 |
+
|
23 |
+
#### Finetuning Details:
|
24 |
+
|
25 |
+
With the utilization of [MonsterAPI](https://monsterapi.ai)'s [no-code LLM finetuner](https://monsterapi.ai/finetuning), this finetuning:
|
26 |
+
|
27 |
+
- Was achieved with great cost-effectiveness.
|
28 |
+
- Completed in a total duration of 15hr 31mins for 1 epochs using an A6000 48GB GPU.
|
29 |
+
- Costed `$31.31` for the entire 1 epoch.
|
30 |
+
|
31 |
+
#### Hyperparameters & Additional Details:
|
32 |
+
|
33 |
+
- **Epochs:** 1
|
34 |
+
- **Total Finetuning Cost:** $31.31
|
35 |
+
- **Model Path:** codellama/CodeLlama-7b-hf
|
36 |
+
- **Learning Rate:** 0.0002
|
37 |
+
- **Data Split:** 100% train
|
38 |
+
- **Gradient Accumulation Steps:** 128
|
39 |
+
- **lora r:** 32
|
40 |
+
- **lora alpha:** 64
|
41 |
+
|
42 |
+
![Train Loss](https://cdn-uploads.huggingface.co/production/uploads/63ba46aa0a9866b28cb19a14/aNujXePogMlJZmoi1Bq56.png)
|
43 |
+
|
44 |
+
---
|
45 |
+
license: apache-2.0
|