Model save
Browse files
README.md
ADDED
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
metrics:
|
5 |
+
- accuracy
|
6 |
+
model-index:
|
7 |
+
- name: lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3_lora2
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3_lora2
|
15 |
+
|
16 |
+
This model was trained from scratch on an unknown dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 2.3138
|
19 |
+
- Accuracy: 0.6569
|
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: 0.0001
|
39 |
+
- train_batch_size: 2
|
40 |
+
- eval_batch_size: 2
|
41 |
+
- seed: 42
|
42 |
+
- distributed_type: multi-GPU
|
43 |
+
- num_devices: 4
|
44 |
+
- gradient_accumulation_steps: 4
|
45 |
+
- total_train_batch_size: 32
|
46 |
+
- total_eval_batch_size: 8
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: constant
|
49 |
+
- lr_scheduler_warmup_ratio: 0.05
|
50 |
+
- num_epochs: 20.0
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
56 |
+
| 1.0895 | 1.0 | 187 | 1.0317 | 0.6904 |
|
57 |
+
| 0.9959 | 2.0 | 375 | 1.0319 | 0.6908 |
|
58 |
+
| 0.9099 | 3.0 | 562 | 1.0512 | 0.6890 |
|
59 |
+
| 0.7893 | 4.0 | 750 | 1.0986 | 0.6843 |
|
60 |
+
| 0.674 | 5.0 | 937 | 1.1695 | 0.6794 |
|
61 |
+
| 0.5518 | 6.0 | 1125 | 1.2677 | 0.6744 |
|
62 |
+
| 0.4364 | 7.0 | 1312 | 1.3770 | 0.6708 |
|
63 |
+
| 0.3384 | 8.0 | 1500 | 1.4781 | 0.6685 |
|
64 |
+
| 0.2488 | 9.0 | 1687 | 1.6224 | 0.6652 |
|
65 |
+
| 0.1889 | 10.0 | 1875 | 1.7546 | 0.6625 |
|
66 |
+
| 0.1469 | 11.0 | 2062 | 1.8724 | 0.6613 |
|
67 |
+
| 0.12 | 12.0 | 2250 | 1.9857 | 0.6595 |
|
68 |
+
| 0.1016 | 13.0 | 2437 | 2.0585 | 0.6597 |
|
69 |
+
| 0.0917 | 14.0 | 2625 | 2.1262 | 0.6584 |
|
70 |
+
| 0.0867 | 15.0 | 2812 | 2.1983 | 0.6578 |
|
71 |
+
| 0.0832 | 16.0 | 3000 | 2.2629 | 0.6573 |
|
72 |
+
| 0.0765 | 17.0 | 3187 | 2.2393 | 0.6581 |
|
73 |
+
| 0.0762 | 18.0 | 3375 | 2.2691 | 0.6576 |
|
74 |
+
| 0.0761 | 19.0 | 3562 | 2.2851 | 0.6581 |
|
75 |
+
| 0.0754 | 19.95 | 3740 | 2.3138 | 0.6569 |
|
76 |
+
|
77 |
+
|
78 |
+
### Framework versions
|
79 |
+
|
80 |
+
- Transformers 4.34.0
|
81 |
+
- Pytorch 2.1.0+cu121
|
82 |
+
- Datasets 2.18.0
|
83 |
+
- Tokenizers 0.14.1
|