Hugging Face's logo

Gunslinger3D
/
fine-tuning-Phi2-with-webglm-qa-with-lora_5

PEFT
Safetensors
Generated from Trainer
Model card Files Community
Gunslinger3D commited on
Commit
ebf31ca
•
1 Parent(s): a7d54ed

fine-tuning-Phi2-with-webglm-qa-with-lora_5

Browse files
Files changed (4) hide show
  1. README.md +162 -0
  2. adapter_config.json +30 -0
  3. adapter_model.safetensors +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,162 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: microsoft/phi-2
7
+ model-index:
8
+ - name: fine-tuning-Phi2-with-webglm-qa-with-lora_5
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
+ # fine-tuning-Phi2-with-webglm-qa-with-lora_5
16
+
17
+ This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.0878
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: 5e-05
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 2
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 5
43
+ - total_train_batch_size: 10
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 100
47
+ - training_steps: 1000
48
+ - mixed_precision_training: Native AMP
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss |
53
+ |:-------------:|:-----:|:----:|:---------------:|
54
+ | 8.1591 | 0.2 | 10 | 7.9109 |
55
+ | 7.8077 | 0.4 | 20 | 7.4417 |
56
+ | 6.7423 | 0.6 | 30 | 6.1597 |
57
+ | 5.2815 | 0.8 | 40 | 3.7018 |
58
+ | 2.6395 | 1.0 | 50 | 1.1413 |
59
+ | 0.7209 | 1.2 | 60 | 0.6488 |
60
+ | 0.5959 | 1.39 | 70 | 0.5735 |
61
+ | 0.5036 | 1.59 | 80 | 0.5102 |
62
+ | 0.4103 | 1.79 | 90 | 0.4500 |
63
+ | 0.3433 | 1.99 | 100 | 0.3905 |
64
+ | 0.3235 | 2.19 | 110 | 0.3371 |
65
+ | 0.2567 | 2.39 | 120 | 0.3032 |
66
+ | 0.2298 | 2.59 | 130 | 0.2785 |
67
+ | 0.2451 | 2.79 | 140 | 0.2553 |
68
+ | 0.1935 | 2.99 | 150 | 0.2363 |
69
+ | 0.1946 | 3.19 | 160 | 0.2248 |
70
+ | 0.1836 | 3.39 | 170 | 0.2097 |
71
+ | 0.1681 | 3.59 | 180 | 0.1984 |
72
+ | 0.1571 | 3.78 | 190 | 0.1877 |
73
+ | 0.1713 | 3.98 | 200 | 0.1820 |
74
+ | 0.15 | 4.18 | 210 | 0.1741 |
75
+ | 0.1315 | 4.38 | 220 | 0.1696 |
76
+ | 0.1567 | 4.58 | 230 | 0.1619 |
77
+ | 0.1225 | 4.78 | 240 | 0.1528 |
78
+ | 0.1346 | 4.98 | 250 | 0.1491 |
79
+ | 0.1336 | 5.18 | 260 | 0.1464 |
80
+ | 0.105 | 5.38 | 270 | 0.1427 |
81
+ | 0.1245 | 5.58 | 280 | 0.1404 |
82
+ | 0.1282 | 5.78 | 290 | 0.1363 |
83
+ | 0.1042 | 5.98 | 300 | 0.1314 |
84
+ | 0.1112 | 6.18 | 310 | 0.1264 |
85
+ | 0.106 | 6.37 | 320 | 0.1249 |
86
+ | 0.1043 | 6.57 | 330 | 0.1240 |
87
+ | 0.1016 | 6.77 | 340 | 0.1196 |
88
+ | 0.096 | 6.97 | 350 | 0.1179 |
89
+ | 0.0927 | 7.17 | 360 | 0.1182 |
90
+ | 0.0997 | 7.37 | 370 | 0.1146 |
91
+ | 0.0914 | 7.57 | 380 | 0.1151 |
92
+ | 0.0993 | 7.77 | 390 | 0.1128 |
93
+ | 0.0863 | 7.97 | 400 | 0.1112 |
94
+ | 0.0757 | 8.17 | 410 | 0.1100 |
95
+ | 0.0803 | 8.37 | 420 | 0.1095 |
96
+ | 0.0969 | 8.57 | 430 | 0.1084 |
97
+ | 0.081 | 8.76 | 440 | 0.1079 |
98
+ | 0.088 | 8.96 | 450 | 0.1050 |
99
+ | 0.082 | 9.16 | 460 | 0.1036 |
100
+ | 0.078 | 9.36 | 470 | 0.1019 |
101
+ | 0.0782 | 9.56 | 480 | 0.1026 |
102
+ | 0.0733 | 9.76 | 490 | 0.1010 |
103
+ | 0.0754 | 9.96 | 500 | 0.1027 |
104
+ | 0.0741 | 10.16 | 510 | 0.1011 |
105
+ | 0.076 | 10.36 | 520 | 0.1023 |
106
+ | 0.078 | 10.56 | 530 | 0.1010 |
107
+ | 0.0701 | 10.76 | 540 | 0.0990 |
108
+ | 0.0636 | 10.96 | 550 | 0.0974 |
109
+ | 0.0668 | 11.16 | 560 | 0.0973 |
110
+ | 0.0672 | 11.35 | 570 | 0.0972 |
111
+ | 0.0634 | 11.55 | 580 | 0.0955 |
112
+ | 0.061 | 11.75 | 590 | 0.0969 |
113
+ | 0.0671 | 11.95 | 600 | 0.0956 |
114
+ | 0.0611 | 12.15 | 610 | 0.0973 |
115
+ | 0.061 | 12.35 | 620 | 0.0966 |
116
+ | 0.0632 | 12.55 | 630 | 0.0950 |
117
+ | 0.0655 | 12.75 | 640 | 0.0945 |
118
+ | 0.0643 | 12.95 | 650 | 0.0944 |
119
+ | 0.0557 | 13.15 | 660 | 0.0942 |
120
+ | 0.0585 | 13.35 | 670 | 0.0937 |
121
+ | 0.0582 | 13.55 | 680 | 0.0933 |
122
+ | 0.0544 | 13.75 | 690 | 0.0927 |
123
+ | 0.0663 | 13.94 | 700 | 0.0917 |
124
+ | 0.0627 | 14.14 | 710 | 0.0917 |
125
+ | 0.0561 | 14.34 | 720 | 0.0923 |
126
+ | 0.0504 | 14.54 | 730 | 0.0914 |
127
+ | 0.0656 | 14.74 | 740 | 0.0907 |
128
+ | 0.0528 | 14.94 | 750 | 0.0898 |
129
+ | 0.0581 | 15.14 | 760 | 0.0916 |
130
+ | 0.0604 | 15.34 | 770 | 0.0912 |
131
+ | 0.0467 | 15.54 | 780 | 0.0907 |
132
+ | 0.048 | 15.74 | 790 | 0.0904 |
133
+ | 0.0571 | 15.94 | 800 | 0.0902 |
134
+ | 0.0521 | 16.14 | 810 | 0.0904 |
135
+ | 0.052 | 16.33 | 820 | 0.0896 |
136
+ | 0.0521 | 16.53 | 830 | 0.0895 |
137
+ | 0.0498 | 16.73 | 840 | 0.0898 |
138
+ | 0.0569 | 16.93 | 850 | 0.0887 |
139
+ | 0.0481 | 17.13 | 860 | 0.0884 |
140
+ | 0.0531 | 17.33 | 870 | 0.0889 |
141
+ | 0.046 | 17.53 | 880 | 0.0886 |
142
+ | 0.0492 | 17.73 | 890 | 0.0887 |
143
+ | 0.0532 | 17.93 | 900 | 0.0885 |
144
+ | 0.0511 | 18.13 | 910 | 0.0878 |
145
+ | 0.0433 | 18.33 | 920 | 0.0881 |
146
+ | 0.0518 | 18.53 | 930 | 0.0884 |
147
+ | 0.049 | 18.73 | 940 | 0.0882 |
148
+ | 0.0493 | 18.92 | 950 | 0.0880 |
149
+ | 0.0479 | 19.12 | 960 | 0.0880 |
150
+ | 0.0439 | 19.32 | 970 | 0.0880 |
151
+ | 0.0535 | 19.52 | 980 | 0.0879 |
152
+ | 0.0501 | 19.72 | 990 | 0.0878 |
153
+ | 0.0466 | 19.92 | 1000 | 0.0878 |
154
+
155
+
156
+ ### Framework versions
157
+
158
+ - PEFT 0.7.1
159
+ - Transformers 4.36.2
160
+ - Pytorch 2.0.0
161
+ - Datasets 2.15.0
162
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "microsoft/phi-2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 16,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "k_proj",
24
+ "fc1",
25
+ "fc2",
26
+ "q_proj",
27
+ "dense"
28
+ ],
29
+ "task_type": "CAUSAL_LM"
30
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1c7412d2faf573be2adef3294c059160560600eb30675a91b2b78c0d0d87bcf
3
+ size 94422368
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c589c733dfb6d340237fe109ade65f744a0ec822fdf2fc7c2c8fcfd714f7e21
3
+ size 4283