Model save
Browse files
README.md
ADDED
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: bigcode-openrail-m
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
base_model: bigcode/starcoder2-3b
|
7 |
+
model-index:
|
8 |
+
- name: starcoder-3b-hugcoder-loftq
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# starcoder-3b-hugcoder-loftq
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [bigcode/starcoder2-3b](https://huggingface.co/bigcode/starcoder2-3b) on an unknown dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.6305
|
20 |
+
|
21 |
+
## Model description
|
22 |
+
|
23 |
+
More information needed
|
24 |
+
|
25 |
+
## Intended uses & limitations
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Training and evaluation data
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training procedure
|
34 |
+
|
35 |
+
### Training hyperparameters
|
36 |
+
|
37 |
+
The following hyperparameters were used during training:
|
38 |
+
- learning_rate: 0.0002
|
39 |
+
- train_batch_size: 16
|
40 |
+
- eval_batch_size: 16
|
41 |
+
- seed: 11
|
42 |
+
- gradient_accumulation_steps: 4
|
43 |
+
- total_train_batch_size: 64
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: cosine
|
46 |
+
- lr_scheduler_warmup_ratio: 0.1
|
47 |
+
- training_steps: 2000
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
53 |
+
| 1.2064 | 0.05 | 100 | 0.7876 |
|
54 |
+
| 0.9806 | 0.1 | 200 | 0.7596 |
|
55 |
+
| 0.8669 | 0.15 | 300 | 0.7397 |
|
56 |
+
| 0.8555 | 0.2 | 400 | 0.7204 |
|
57 |
+
| 0.8566 | 0.25 | 500 | 0.7058 |
|
58 |
+
| 0.7643 | 0.3 | 600 | 0.6911 |
|
59 |
+
| 0.7464 | 0.35 | 700 | 0.6801 |
|
60 |
+
| 0.734 | 0.4 | 800 | 0.6697 |
|
61 |
+
| 0.7209 | 0.45 | 900 | 0.6608 |
|
62 |
+
| 0.6477 | 0.5 | 1000 | 0.6537 |
|
63 |
+
| 0.7292 | 0.55 | 1100 | 0.6475 |
|
64 |
+
| 0.6612 | 0.6 | 1200 | 0.6434 |
|
65 |
+
| 0.6808 | 0.65 | 1300 | 0.6393 |
|
66 |
+
| 0.6586 | 0.7 | 1400 | 0.6365 |
|
67 |
+
| 0.662 | 0.75 | 1500 | 0.6333 |
|
68 |
+
| 0.7034 | 0.8 | 1600 | 0.6322 |
|
69 |
+
| 0.683 | 0.85 | 1700 | 0.6309 |
|
70 |
+
| 0.7074 | 0.9 | 1800 | 0.6303 |
|
71 |
+
| 0.6362 | 0.95 | 1900 | 0.6307 |
|
72 |
+
| 0.6617 | 1.0 | 2000 | 0.6305 |
|
73 |
+
|
74 |
+
|
75 |
+
### Framework versions
|
76 |
+
|
77 |
+
- PEFT 0.10.1.dev0
|
78 |
+
- Transformers 4.40.0.dev0
|
79 |
+
- Pytorch 2.2.1+cu121
|
80 |
+
- Datasets 2.18.0
|
81 |
+
- Tokenizers 0.15.2
|