Hugging Face's logo

chchen
/
Llama-3.1-8B-Instruct-PsyCourse-fold4

PEFT
Safetensors
llama-factory
lora
Generated from Trainer
Model card Files Community
chchen commited on
Commit
a6d4ce7
·
verified ·
1 Parent(s): 2d3cf2c

Model save

Browse files
Files changed (2) hide show
  1. README.md +68 -134
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # Llama-3.1-8B-Instruct-PsyCourse-fold4
18
 
19
- This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the course-train-fold4 dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.0356
22
 
23
  ## Model description
24
 
@@ -46,143 +46,77 @@ The following hyperparameters were used during training:
46
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
  - lr_scheduler_type: cosine
48
  - lr_scheduler_warmup_ratio: 0.1
49
- - num_epochs: 10.0
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:------:|:----:|:---------------:|
55
- | 0.916 | 0.0763 | 50 | 0.6882 |
56
- | 0.2175 | 0.1527 | 100 | 0.1345 |
57
- | 0.0687 | 0.2290 | 150 | 0.0777 |
58
- | 0.0541 | 0.3053 | 200 | 0.0599 |
59
- | 0.0681 | 0.3816 | 250 | 0.0532 |
60
- | 0.0534 | 0.4580 | 300 | 0.0525 |
61
- | 0.0503 | 0.5343 | 350 | 0.0565 |
62
- | 0.0505 | 0.6106 | 400 | 0.0519 |
63
- | 0.0641 | 0.6870 | 450 | 0.0547 |
64
- | 0.038 | 0.7633 | 500 | 0.0414 |
65
- | 0.0491 | 0.8396 | 550 | 0.0440 |
66
- | 0.048 | 0.9159 | 600 | 0.0416 |
67
- | 0.0394 | 0.9923 | 650 | 0.0428 |
68
- | 0.04 | 1.0686 | 700 | 0.0399 |
69
- | 0.041 | 1.1449 | 750 | 0.0392 |
70
- | 0.0349 | 1.2213 | 800 | 0.0456 |
71
- | 0.0351 | 1.2976 | 850 | 0.0409 |
72
- | 0.0372 | 1.3739 | 900 | 0.0435 |
73
- | 0.0349 | 1.4502 | 950 | 0.0378 |
74
- | 0.0289 | 1.5266 | 1000 | 0.0392 |
75
- | 0.0329 | 1.6029 | 1050 | 0.0396 |
76
- | 0.0547 | 1.6792 | 1100 | 0.0405 |
77
- | 0.0385 | 1.7556 | 1150 | 0.0407 |
78
- | 0.054 | 1.8319 | 1200 | 0.0386 |
79
- | 0.0311 | 1.9082 | 1250 | 0.0370 |
80
- | 0.0381 | 1.9845 | 1300 | 0.0356 |
81
- | 0.0222 | 2.0609 | 1350 | 0.0367 |
82
- | 0.0259 | 2.1372 | 1400 | 0.0379 |
83
- | 0.0332 | 2.2135 | 1450 | 0.0382 |
84
- | 0.0186 | 2.2899 | 1500 | 0.0399 |
85
- | 0.0181 | 2.3662 | 1550 | 0.0415 |
86
- | 0.0253 | 2.4425 | 1600 | 0.0371 |
87
- | 0.0267 | 2.5188 | 1650 | 0.0368 |
88
- | 0.0275 | 2.5952 | 1700 | 0.0381 |
89
- | 0.0242 | 2.6715 | 1750 | 0.0376 |
90
- | 0.0277 | 2.7478 | 1800 | 0.0382 |
91
- | 0.0315 | 2.8242 | 1850 | 0.0364 |
92
- | 0.0276 | 2.9005 | 1900 | 0.0370 |
93
- | 0.0333 | 2.9768 | 1950 | 0.0362 |
94
- | 0.013 | 3.0531 | 2000 | 0.0408 |
95
- | 0.0146 | 3.1295 | 2050 | 0.0397 |
96
- | 0.0168 | 3.2058 | 2100 | 0.0419 |
97
- | 0.0202 | 3.2821 | 2150 | 0.0451 |
98
- | 0.0197 | 3.3585 | 2200 | 0.0398 |
99
- | 0.0167 | 3.4348 | 2250 | 0.0421 |
100
- | 0.0224 | 3.5111 | 2300 | 0.0413 |
101
- | 0.0173 | 3.5874 | 2350 | 0.0457 |
102
- | 0.0143 | 3.6638 | 2400 | 0.0442 |
103
- | 0.0179 | 3.7401 | 2450 | 0.0422 |
104
- | 0.0178 | 3.8164 | 2500 | 0.0436 |
105
- | 0.0136 | 3.8928 | 2550 | 0.0450 |
106
- | 0.0193 | 3.9691 | 2600 | 0.0397 |
107
- | 0.009 | 4.0454 | 2650 | 0.0460 |
108
- | 0.0107 | 4.1217 | 2700 | 0.0470 |
109
- | 0.0038 | 4.1981 | 2750 | 0.0484 |
110
- | 0.0078 | 4.2744 | 2800 | 0.0531 |
111
- | 0.0105 | 4.3507 | 2850 | 0.0516 |
112
- | 0.0067 | 4.4271 | 2900 | 0.0596 |
113
- | 0.0082 | 4.5034 | 2950 | 0.0517 |
114
- | 0.0121 | 4.5797 | 3000 | 0.0526 |
115
- | 0.0044 | 4.6560 | 3050 | 0.0487 |
116
- | 0.0107 | 4.7324 | 3100 | 0.0495 |
117
- | 0.0146 | 4.8087 | 3150 | 0.0517 |
118
- | 0.0098 | 4.8850 | 3200 | 0.0537 |
119
- | 0.0088 | 4.9614 | 3250 | 0.0484 |
120
- | 0.0041 | 5.0377 | 3300 | 0.0548 |
121
- | 0.0049 | 5.1140 | 3350 | 0.0611 |
122
- | 0.0048 | 5.1903 | 3400 | 0.0641 |
123
- | 0.0025 | 5.2667 | 3450 | 0.0638 |
124
- | 0.0054 | 5.3430 | 3500 | 0.0624 |
125
- | 0.0069 | 5.4193 | 3550 | 0.0626 |
126
- | 0.0031 | 5.4957 | 3600 | 0.0694 |
127
- | 0.0109 | 5.5720 | 3650 | 0.0659 |
128
- | 0.0015 | 5.6483 | 3700 | 0.0763 |
129
- | 0.0034 | 5.7246 | 3750 | 0.0662 |
130
- | 0.0094 | 5.8010 | 3800 | 0.0618 |
131
- | 0.0027 | 5.8773 | 3850 | 0.0643 |
132
- | 0.0038 | 5.9536 | 3900 | 0.0737 |
133
- | 0.001 | 6.0300 | 3950 | 0.0605 |
134
- | 0.0022 | 6.1063 | 4000 | 0.0703 |
135
- | 0.0015 | 6.1826 | 4050 | 0.0744 |
136
- | 0.003 | 6.2589 | 4100 | 0.0747 |
137
- | 0.0022 | 6.3353 | 4150 | 0.0690 |
138
- | 0.0055 | 6.4116 | 4200 | 0.0651 |
139
- | 0.0039 | 6.4879 | 4250 | 0.0651 |
140
- | 0.0038 | 6.5643 | 4300 | 0.0638 |
141
- | 0.0009 | 6.6406 | 4350 | 0.0696 |
142
- | 0.0032 | 6.7169 | 4400 | 0.0711 |
143
- | 0.001 | 6.7932 | 4450 | 0.0718 |
144
- | 0.0024 | 6.8696 | 4500 | 0.0740 |
145
- | 0.0014 | 6.9459 | 4550 | 0.0733 |
146
- | 0.0017 | 7.0222 | 4600 | 0.0682 |
147
- | 0.0009 | 7.0986 | 4650 | 0.0757 |
148
- | 0.0016 | 7.1749 | 4700 | 0.0777 |
149
- | 0.0004 | 7.2512 | 4750 | 0.0799 |
150
- | 0.0019 | 7.3275 | 4800 | 0.0839 |
151
- | 0.0003 | 7.4039 | 4850 | 0.0895 |
152
- | 0.0001 | 7.4802 | 4900 | 0.0925 |
153
- | 0.0004 | 7.5565 | 4950 | 0.0872 |
154
- | 0.0005 | 7.6329 | 5000 | 0.0858 |
155
- | 0.0007 | 7.7092 | 5050 | 0.0887 |
156
- | 0.0002 | 7.7855 | 5100 | 0.0849 |
157
- | 0.0007 | 7.8618 | 5150 | 0.0830 |
158
- | 0.0016 | 7.9382 | 5200 | 0.0849 |
159
- | 0.0003 | 8.0145 | 5250 | 0.0859 |
160
- | 0.0001 | 8.0908 | 5300 | 0.0888 |
161
- | 0.0007 | 8.1672 | 5350 | 0.0886 |
162
- | 0.0003 | 8.2435 | 5400 | 0.0913 |
163
- | 0.0001 | 8.3198 | 5450 | 0.0924 |
164
- | 0.0008 | 8.3961 | 5500 | 0.0937 |
165
- | 0.0001 | 8.4725 | 5550 | 0.0948 |
166
- | 0.0001 | 8.5488 | 5600 | 0.0949 |
167
- | 0.0005 | 8.6251 | 5650 | 0.0965 |
168
- | 0.0009 | 8.7015 | 5700 | 0.0980 |
169
- | 0.001 | 8.7778 | 5750 | 0.0979 |
170
- | 0.0002 | 8.8541 | 5800 | 0.0976 |
171
- | 0.0004 | 8.9304 | 5850 | 0.0987 |
172
- | 0.0008 | 9.0068 | 5900 | 0.0994 |
173
- | 0.0001 | 9.0831 | 5950 | 0.1002 |
174
- | 0.0001 | 9.1594 | 6000 | 0.1007 |
175
- | 0.0001 | 9.2358 | 6050 | 0.1010 |
176
- | 0.0001 | 9.3121 | 6100 | 0.1017 |
177
- | 0.0003 | 9.3884 | 6150 | 0.1019 |
178
- | 0.0001 | 9.4647 | 6200 | 0.1018 |
179
- | 0.0006 | 9.5411 | 6250 | 0.1023 |
180
- | 0.0001 | 9.6174 | 6300 | 0.1022 |
181
- | 0.0006 | 9.6937 | 6350 | 0.1026 |
182
- | 0.0001 | 9.7701 | 6400 | 0.1025 |
183
- | 0.0 | 9.8464 | 6450 | 0.1024 |
184
- | 0.0 | 9.9227 | 6500 | 0.1023 |
185
- | 0.0002 | 9.9990 | 6550 | 0.1026 |
186
 
187
 
188
  ### Framework versions
 
16
 
17
  # Llama-3.1-8B-Instruct-PsyCourse-fold4
18
 
19
+ This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.0622
22
 
23
  ## Model description
24
 
 
46
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
  - lr_scheduler_type: cosine
48
  - lr_scheduler_warmup_ratio: 0.1
49
+ - num_epochs: 5.0
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:------:|:----:|:---------------:|
55
+ | 0.5681 | 0.0763 | 50 | 0.3814 |
56
+ | 0.122 | 0.1527 | 100 | 0.0796 |
57
+ | 0.0541 | 0.2290 | 150 | 0.0592 |
58
+ | 0.0502 | 0.3053 | 200 | 0.0522 |
59
+ | 0.0666 | 0.3816 | 250 | 0.0539 |
60
+ | 0.046 | 0.4580 | 300 | 0.0493 |
61
+ | 0.0458 | 0.5343 | 350 | 0.0527 |
62
+ | 0.0448 | 0.6106 | 400 | 0.0488 |
63
+ | 0.0567 | 0.6870 | 450 | 0.0462 |
64
+ | 0.0358 | 0.7633 | 500 | 0.0410 |
65
+ | 0.0445 | 0.8396 | 550 | 0.0407 |
66
+ | 0.0462 | 0.9159 | 600 | 0.0407 |
67
+ | 0.0363 | 0.9923 | 650 | 0.0410 |
68
+ | 0.0343 | 1.0686 | 700 | 0.0370 |
69
+ | 0.0413 | 1.1449 | 750 | 0.0378 |
70
+ | 0.0322 | 1.2213 | 800 | 0.0398 |
71
+ | 0.0342 | 1.2976 | 850 | 0.0385 |
72
+ | 0.0337 | 1.3739 | 900 | 0.0436 |
73
+ | 0.0295 | 1.4502 | 950 | 0.0373 |
74
+ | 0.0267 | 1.5266 | 1000 | 0.0386 |
75
+ | 0.0287 | 1.6029 | 1050 | 0.0380 |
76
+ | 0.0504 | 1.6792 | 1100 | 0.0388 |
77
+ | 0.0317 | 1.7556 | 1150 | 0.0391 |
78
+ | 0.0448 | 1.8319 | 1200 | 0.0366 |
79
+ | 0.0278 | 1.9082 | 1250 | 0.0362 |
80
+ | 0.0347 | 1.9845 | 1300 | 0.0344 |
81
+ | 0.0201 | 2.0609 | 1350 | 0.0355 |
82
+ | 0.0238 | 2.1372 | 1400 | 0.0357 |
83
+ | 0.0299 | 2.2135 | 1450 | 0.0371 |
84
+ | 0.0155 | 2.2899 | 1500 | 0.0384 |
85
+ | 0.0157 | 2.3662 | 1550 | 0.0391 |
86
+ | 0.0222 | 2.4425 | 1600 | 0.0370 |
87
+ | 0.0245 | 2.5188 | 1650 | 0.0360 |
88
+ | 0.0206 | 2.5952 | 1700 | 0.0376 |
89
+ | 0.0198 | 2.6715 | 1750 | 0.0363 |
90
+ | 0.0209 | 2.7478 | 1800 | 0.0370 |
91
+ | 0.026 | 2.8242 | 1850 | 0.0362 |
92
+ | 0.0197 | 2.9005 | 1900 | 0.0358 |
93
+ | 0.0291 | 2.9768 | 1950 | 0.0355 |
94
+ | 0.0091 | 3.0531 | 2000 | 0.0416 |
95
+ | 0.0132 | 3.1295 | 2050 | 0.0421 |
96
+ | 0.0115 | 3.2058 | 2100 | 0.0443 |
97
+ | 0.0131 | 3.2821 | 2150 | 0.0459 |
98
+ | 0.0132 | 3.3585 | 2200 | 0.0409 |
99
+ | 0.0077 | 3.4348 | 2250 | 0.0445 |
100
+ | 0.0156 | 3.5111 | 2300 | 0.0444 |
101
+ | 0.0125 | 3.5874 | 2350 | 0.0480 |
102
+ | 0.0089 | 3.6638 | 2400 | 0.0499 |
103
+ | 0.0125 | 3.7401 | 2450 | 0.0467 |
104
+ | 0.0115 | 3.8164 | 2500 | 0.0447 |
105
+ | 0.0062 | 3.8928 | 2550 | 0.0449 |
106
+ | 0.0112 | 3.9691 | 2600 | 0.0462 |
107
+ | 0.005 | 4.0454 | 2650 | 0.0465 |
108
+ | 0.0065 | 4.1217 | 2700 | 0.0502 |
109
+ | 0.0021 | 4.1981 | 2750 | 0.0543 |
110
+ | 0.0033 | 4.2744 | 2800 | 0.0556 |
111
+ | 0.0068 | 4.3507 | 2850 | 0.0572 |
112
+ | 0.0015 | 4.4271 | 2900 | 0.0599 |
113
+ | 0.0036 | 4.5034 | 2950 | 0.0602 |
114
+ | 0.0027 | 4.5797 | 3000 | 0.0615 |
115
+ | 0.0013 | 4.6560 | 3050 | 0.0615 |
116
+ | 0.0056 | 4.7324 | 3100 | 0.0618 |
117
+ | 0.0028 | 4.8087 | 3150 | 0.0618 |
118
+ | 0.0044 | 4.8850 | 3200 | 0.0620 |
119
+ | 0.0061 | 4.9614 | 3250 | 0.0622 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
120
 
121
 
122
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:688b6a85c26dd39605b5263e3604f04d6a4536fceb0beabd38ab0f8429d5f1cb
3
  size 83945296
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67e0e2fde6615cf54e629ab0899c6d35da63b9351774dd522e95b2dacd0c21e4
3
  size 83945296