qwen-rep-OpenAssistant/oasst_top1_2023-08-25
Browse files- README.md +57 -57
- adapter_model.bin +1 -1
README.md
CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
This model is a fine-tuned version of [Qwen/Qwen-14B](https://huggingface.co/Qwen/Qwen-14B) on an unknown dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
-
- Loss: 1.
|
18 |
|
19 |
## Model description
|
20 |
|
@@ -33,11 +33,11 @@ More information needed
|
|
33 |
### Training hyperparameters
|
34 |
|
35 |
The following hyperparameters were used during training:
|
36 |
-
- learning_rate:
|
37 |
-
- train_batch_size:
|
38 |
-
- eval_batch_size:
|
39 |
- seed: 42
|
40 |
-
- gradient_accumulation_steps:
|
41 |
- total_train_batch_size: 16
|
42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
- lr_scheduler_type: cosine
|
@@ -48,59 +48,59 @@ The following hyperparameters were used during training:
|
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|
|
51 |
-
|
|
52 |
-
| 1.
|
53 |
-
| 1.
|
54 |
-
| 1.
|
55 |
-
| 1.
|
56 |
-
| 1.
|
57 |
-
| 1.
|
58 |
-
| 1.
|
59 |
-
| 1.
|
60 |
-
| 1.
|
61 |
-
| 1.
|
62 |
-
| 1.
|
63 |
-
| 1.
|
64 |
-
| 1.
|
65 |
-
| 1.
|
66 |
-
| 1.
|
67 |
-
| 1.
|
68 |
-
| 1.
|
69 |
-
| 1.
|
70 |
-
| 1.
|
71 |
-
| 1.
|
72 |
-
| 1.
|
73 |
-
| 1.
|
74 |
-
| 1.
|
75 |
-
| 1.
|
76 |
-
| 1.
|
77 |
-
| 1.
|
78 |
-
| 1.
|
79 |
-
| 1.
|
80 |
-
| 1.
|
81 |
-
| 1.
|
82 |
-
| 1.
|
83 |
-
| 1.
|
84 |
-
| 1.
|
85 |
-
| 1.
|
86 |
-
| 1.
|
87 |
-
| 1.
|
88 |
-
| 1.
|
89 |
-
| 1.
|
90 |
-
| 1.
|
91 |
-
| 1.
|
92 |
-
| 1.
|
93 |
-
| 1.
|
94 |
-
| 1.
|
95 |
-
| 1.
|
96 |
-
| 1.
|
97 |
-
| 1.
|
98 |
-
| 1.
|
99 |
|
100 |
|
101 |
### Framework versions
|
102 |
|
103 |
-
- Transformers 4.
|
104 |
-
- Pytorch 2.1.0
|
105 |
-
- Datasets 2.
|
106 |
-
- Tokenizers 0.
|
|
|
14 |
|
15 |
This model is a fine-tuned version of [Qwen/Qwen-14B](https://huggingface.co/Qwen/Qwen-14B) on an unknown dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 1.6501
|
18 |
|
19 |
## Model description
|
20 |
|
|
|
33 |
### Training hyperparameters
|
34 |
|
35 |
The following hyperparameters were used during training:
|
36 |
+
- learning_rate: 0.0001
|
37 |
+
- train_batch_size: 1
|
38 |
+
- eval_batch_size: 1
|
39 |
- seed: 42
|
40 |
+
- gradient_accumulation_steps: 16
|
41 |
- total_train_batch_size: 16
|
42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
- lr_scheduler_type: cosine
|
|
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|
|
51 |
+
| 2.163 | 0.02 | 16 | 1.9459 |
|
52 |
+
| 1.9498 | 0.04 | 32 | 1.8467 |
|
53 |
+
| 1.9578 | 0.06 | 48 | 1.7864 |
|
54 |
+
| 1.8398 | 0.08 | 64 | 1.7530 |
|
55 |
+
| 1.7696 | 0.1 | 80 | 1.7076 |
|
56 |
+
| 1.7744 | 0.12 | 96 | 1.7275 |
|
57 |
+
| 1.8108 | 0.14 | 112 | 1.6887 |
|
58 |
+
| 1.7707 | 0.17 | 128 | 1.6942 |
|
59 |
+
| 1.787 | 0.19 | 144 | 1.6894 |
|
60 |
+
| 1.7029 | 0.21 | 160 | 1.6760 |
|
61 |
+
| 1.6732 | 0.23 | 176 | 1.6838 |
|
62 |
+
| 1.6313 | 0.25 | 192 | 1.6754 |
|
63 |
+
| 1.7071 | 0.27 | 208 | 1.6752 |
|
64 |
+
| 1.6781 | 0.29 | 224 | 1.6741 |
|
65 |
+
| 1.7782 | 0.31 | 240 | 1.6698 |
|
66 |
+
| 1.6836 | 0.33 | 256 | 1.6592 |
|
67 |
+
| 1.7229 | 0.35 | 272 | 1.6633 |
|
68 |
+
| 1.7196 | 0.37 | 288 | 1.6638 |
|
69 |
+
| 1.6892 | 0.39 | 304 | 1.6627 |
|
70 |
+
| 1.6844 | 0.41 | 320 | 1.6557 |
|
71 |
+
| 1.8027 | 0.43 | 336 | 1.6540 |
|
72 |
+
| 1.692 | 0.45 | 352 | 1.6577 |
|
73 |
+
| 1.7088 | 0.47 | 368 | 1.6611 |
|
74 |
+
| 1.7987 | 0.5 | 384 | 1.6557 |
|
75 |
+
| 1.709 | 0.52 | 400 | 1.6600 |
|
76 |
+
| 1.701 | 0.54 | 416 | 1.6588 |
|
77 |
+
| 1.6784 | 0.56 | 432 | 1.6594 |
|
78 |
+
| 1.6997 | 0.58 | 448 | 1.6484 |
|
79 |
+
| 1.7811 | 0.6 | 464 | 1.6583 |
|
80 |
+
| 1.7628 | 0.62 | 480 | 1.6461 |
|
81 |
+
| 1.6254 | 0.64 | 496 | 1.6527 |
|
82 |
+
| 1.6684 | 0.66 | 512 | 1.6520 |
|
83 |
+
| 1.6837 | 0.68 | 528 | 1.6570 |
|
84 |
+
| 1.7209 | 0.7 | 544 | 1.6543 |
|
85 |
+
| 1.677 | 0.72 | 560 | 1.6562 |
|
86 |
+
| 1.6819 | 0.74 | 576 | 1.6517 |
|
87 |
+
| 1.7072 | 0.76 | 592 | 1.6551 |
|
88 |
+
| 1.6446 | 0.78 | 608 | 1.6562 |
|
89 |
+
| 1.6908 | 0.8 | 624 | 1.6528 |
|
90 |
+
| 1.7209 | 0.83 | 640 | 1.6518 |
|
91 |
+
| 1.6818 | 0.85 | 656 | 1.6517 |
|
92 |
+
| 1.7007 | 0.87 | 672 | 1.6525 |
|
93 |
+
| 1.8077 | 0.89 | 688 | 1.6522 |
|
94 |
+
| 1.6856 | 0.91 | 704 | 1.6516 |
|
95 |
+
| 1.7247 | 0.93 | 720 | 1.6509 |
|
96 |
+
| 1.6645 | 0.95 | 736 | 1.6500 |
|
97 |
+
| 1.6841 | 0.97 | 752 | 1.6499 |
|
98 |
+
| 1.7244 | 0.99 | 768 | 1.6501 |
|
99 |
|
100 |
|
101 |
### Framework versions
|
102 |
|
103 |
+
- Transformers 4.32.0
|
104 |
+
- Pytorch 2.1.0
|
105 |
+
- Datasets 2.14.7
|
106 |
+
- Tokenizers 0.13.3
|
adapter_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 9312358
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9a2c67675298f4c45b4bc7df7dc927f8e57796428bd311446fe1ef1d61000597
|
3 |
size 9312358
|