Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
output
|
2 |
+
---
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: unsloth/tinyllama-bnb-4bit
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- generator
|
9 |
+
model-index:
|
10 |
+
- name: tinyllama-bnb-4bit-ft-codeAlpaca
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# tinyllama-bnb-4bit-ft-codeAlpaca
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [unsloth/tinyllama-bnb-4bit](https://huggingface.co/unsloth/tinyllama-bnb-4bit) on the generator dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.7926
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 2e-05
|
41 |
+
- train_batch_size: 4
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 3407
|
44 |
+
- gradient_accumulation_steps: 2
|
45 |
+
- total_train_batch_size: 8
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: linear
|
48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
49 |
+
- num_epochs: 5
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
55 |
+
| 1.7027 | 0.33 | 20 | 1.6844 |
|
56 |
+
| 1.2736 | 0.66 | 40 | 1.2825 |
|
57 |
+
| 1.091 | 0.99 | 60 | 1.0799 |
|
58 |
+
| 0.9683 | 1.32 | 80 | 0.9900 |
|
59 |
+
| 0.9431 | 1.65 | 100 | 0.9328 |
|
60 |
+
| 0.9352 | 1.98 | 120 | 0.8907 |
|
61 |
+
| 0.857 | 2.31 | 140 | 0.8580 |
|
62 |
+
| 0.8316 | 2.64 | 160 | 0.8346 |
|
63 |
+
| 0.8351 | 2.98 | 180 | 0.8203 |
|
64 |
+
| 0.8204 | 3.31 | 200 | 0.8103 |
|
65 |
+
| 0.7888 | 3.64 | 220 | 0.8035 |
|
66 |
+
| 0.7677 | 3.97 | 240 | 0.7985 |
|
67 |
+
| 0.7753 | 4.3 | 260 | 0.7952 |
|
68 |
+
| 0.7912 | 4.63 | 280 | 0.7931 |
|
69 |
+
| 0.7631 | 4.96 | 300 | 0.7926 |
|
70 |
+
|
71 |
+
|
72 |
+
### Framework versions
|
73 |
+
|
74 |
+
- Transformers 4.37.0.dev0
|
75 |
+
- Pytorch 2.1.0+cu121
|
76 |
+
- Datasets 2.16.1
|
77 |
+
- Tokenizers 0.15.0
|