Hugging Face's logo

chchen
/
Llama-3.1-8B-Instruct-PsyCourse-fold5

PEFT
Safetensors
llama-factory
lora
Generated from Trainer
Model card Files Community
chchen commited on
Commit
4c79fd5
·
verified ·
1 Parent(s): eb4cac2

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-fold5
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-fold5 dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.0334
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
- | 0.9962 | 0.0758 | 50 | 0.7209 |
56
- | 0.1767 | 0.1517 | 100 | 0.1586 |
57
- | 0.0837 | 0.2275 | 150 | 0.0798 |
58
- | 0.0672 | 0.3033 | 200 | 0.0662 |
59
- | 0.0585 | 0.3791 | 250 | 0.0565 |
60
- | 0.0501 | 0.4550 | 300 | 0.0505 |
61
- | 0.0571 | 0.5308 | 350 | 0.0490 |
62
- | 0.0694 | 0.6066 | 400 | 0.0491 |
63
- | 0.0582 | 0.6825 | 450 | 0.0486 |
64
- | 0.0363 | 0.7583 | 500 | 0.0432 |
65
- | 0.0349 | 0.8341 | 550 | 0.0466 |
66
- | 0.0473 | 0.9100 | 600 | 0.0458 |
67
- | 0.0493 | 0.9858 | 650 | 0.0445 |
68
- | 0.0371 | 1.0616 | 700 | 0.0471 |
69
- | 0.0413 | 1.1374 | 750 | 0.0468 |
70
- | 0.0379 | 1.2133 | 800 | 0.0397 |
71
- | 0.0351 | 1.2891 | 850 | 0.0412 |
72
- | 0.03 | 1.3649 | 900 | 0.0419 |
73
- | 0.0321 | 1.4408 | 950 | 0.0431 |
74
- | 0.0448 | 1.5166 | 1000 | 0.0386 |
75
- | 0.0301 | 1.5924 | 1050 | 0.0367 |
76
- | 0.0248 | 1.6682 | 1100 | 0.0376 |
77
- | 0.0315 | 1.7441 | 1150 | 0.0390 |
78
- | 0.0357 | 1.8199 | 1200 | 0.0369 |
79
- | 0.0357 | 1.8957 | 1250 | 0.0344 |
80
- | 0.0434 | 1.9716 | 1300 | 0.0384 |
81
- | 0.0275 | 2.0474 | 1350 | 0.0352 |
82
- | 0.0204 | 2.1232 | 1400 | 0.0347 |
83
- | 0.022 | 2.1991 | 1450 | 0.0377 |
84
- | 0.0157 | 2.2749 | 1500 | 0.0375 |
85
- | 0.0287 | 2.3507 | 1550 | 0.0348 |
86
- | 0.0279 | 2.4265 | 1600 | 0.0358 |
87
- | 0.0164 | 2.5024 | 1650 | 0.0384 |
88
- | 0.03 | 2.5782 | 1700 | 0.0367 |
89
- | 0.0278 | 2.6540 | 1750 | 0.0399 |
90
- | 0.0305 | 2.7299 | 1800 | 0.0355 |
91
- | 0.0329 | 2.8057 | 1850 | 0.0376 |
92
- | 0.0248 | 2.8815 | 1900 | 0.0360 |
93
- | 0.0329 | 2.9573 | 1950 | 0.0334 |
94
- | 0.0129 | 3.0332 | 2000 | 0.0392 |
95
- | 0.01 | 3.1090 | 2050 | 0.0398 |
96
- | 0.0156 | 3.1848 | 2100 | 0.0395 |
97
- | 0.0151 | 3.2607 | 2150 | 0.0446 |
98
- | 0.0181 | 3.3365 | 2200 | 0.0362 |
99
- | 0.0114 | 3.4123 | 2250 | 0.0377 |
100
- | 0.0178 | 3.4882 | 2300 | 0.0382 |
101
- | 0.0131 | 3.5640 | 2350 | 0.0382 |
102
- | 0.0207 | 3.6398 | 2400 | 0.0354 |
103
- | 0.0211 | 3.7156 | 2450 | 0.0391 |
104
- | 0.0254 | 3.7915 | 2500 | 0.0401 |
105
- | 0.0171 | 3.8673 | 2550 | 0.0389 |
106
- | 0.0203 | 3.9431 | 2600 | 0.0386 |
107
- | 0.0128 | 4.0190 | 2650 | 0.0386 |
108
- | 0.0072 | 4.0948 | 2700 | 0.0465 |
109
- | 0.01 | 4.1706 | 2750 | 0.0525 |
110
- | 0.015 | 4.2464 | 2800 | 0.0420 |
111
- | 0.0122 | 4.3223 | 2850 | 0.0437 |
112
- | 0.0147 | 4.3981 | 2900 | 0.0449 |
113
- | 0.0139 | 4.4739 | 2950 | 0.0425 |
114
- | 0.011 | 4.5498 | 3000 | 0.0443 |
115
- | 0.0061 | 4.6256 | 3050 | 0.0445 |
116
- | 0.0159 | 4.7014 | 3100 | 0.0451 |
117
- | 0.0111 | 4.7773 | 3150 | 0.0408 |
118
- | 0.0058 | 4.8531 | 3200 | 0.0434 |
119
- | 0.0106 | 4.9289 | 3250 | 0.0438 |
120
- | 0.0123 | 5.0047 | 3300 | 0.0401 |
121
- | 0.0043 | 5.0806 | 3350 | 0.0508 |
122
- | 0.0051 | 5.1564 | 3400 | 0.0526 |
123
- | 0.0043 | 5.2322 | 3450 | 0.0509 |
124
- | 0.0049 | 5.3081 | 3500 | 0.0480 |
125
- | 0.0043 | 5.3839 | 3550 | 0.0463 |
126
- | 0.0052 | 5.4597 | 3600 | 0.0507 |
127
- | 0.0086 | 5.5355 | 3650 | 0.0571 |
128
- | 0.0026 | 5.6114 | 3700 | 0.0495 |
129
- | 0.0032 | 5.6872 | 3750 | 0.0489 |
130
- | 0.0034 | 5.7630 | 3800 | 0.0507 |
131
- | 0.0019 | 5.8389 | 3850 | 0.0481 |
132
- | 0.0024 | 5.9147 | 3900 | 0.0502 |
133
- | 0.0089 | 5.9905 | 3950 | 0.0476 |
134
- | 0.0029 | 6.0664 | 4000 | 0.0478 |
135
- | 0.004 | 6.1422 | 4050 | 0.0524 |
136
- | 0.0007 | 6.2180 | 4100 | 0.0559 |
137
- | 0.0038 | 6.2938 | 4150 | 0.0563 |
138
- | 0.0012 | 6.3697 | 4200 | 0.0545 |
139
- | 0.0021 | 6.4455 | 4250 | 0.0590 |
140
- | 0.0027 | 6.5213 | 4300 | 0.0562 |
141
- | 0.0045 | 6.5972 | 4350 | 0.0500 |
142
- | 0.0024 | 6.6730 | 4400 | 0.0543 |
143
- | 0.0025 | 6.7488 | 4450 | 0.0604 |
144
- | 0.0015 | 6.8246 | 4500 | 0.0572 |
145
- | 0.0043 | 6.9005 | 4550 | 0.0528 |
146
- | 0.0064 | 6.9763 | 4600 | 0.0526 |
147
- | 0.0009 | 7.0521 | 4650 | 0.0575 |
148
- | 0.0006 | 7.1280 | 4700 | 0.0605 |
149
- | 0.0007 | 7.2038 | 4750 | 0.0589 |
150
- | 0.0005 | 7.2796 | 4800 | 0.0606 |
151
- | 0.0002 | 7.3555 | 4850 | 0.0630 |
152
- | 0.0003 | 7.4313 | 4900 | 0.0659 |
153
- | 0.0017 | 7.5071 | 4950 | 0.0695 |
154
- | 0.0011 | 7.5829 | 5000 | 0.0662 |
155
- | 0.0017 | 7.6588 | 5050 | 0.0590 |
156
- | 0.0005 | 7.7346 | 5100 | 0.0615 |
157
- | 0.0019 | 7.8104 | 5150 | 0.0583 |
158
- | 0.0009 | 7.8863 | 5200 | 0.0603 |
159
- | 0.0015 | 7.9621 | 5250 | 0.0600 |
160
- | 0.0012 | 8.0379 | 5300 | 0.0613 |
161
- | 0.0003 | 8.1137 | 5350 | 0.0645 |
162
- | 0.0004 | 8.1896 | 5400 | 0.0653 |
163
- | 0.0002 | 8.2654 | 5450 | 0.0663 |
164
- | 0.0006 | 8.3412 | 5500 | 0.0675 |
165
- | 0.0004 | 8.4171 | 5550 | 0.0670 |
166
- | 0.0005 | 8.4929 | 5600 | 0.0671 |
167
- | 0.0018 | 8.5687 | 5650 | 0.0666 |
168
- | 0.0003 | 8.6445 | 5700 | 0.0645 |
169
- | 0.0008 | 8.7204 | 5750 | 0.0656 |
170
- | 0.0009 | 8.7962 | 5800 | 0.0665 |
171
- | 0.0005 | 8.8720 | 5850 | 0.0677 |
172
- | 0.0008 | 8.9479 | 5900 | 0.0680 |
173
- | 0.0007 | 9.0237 | 5950 | 0.0683 |
174
- | 0.0002 | 9.0995 | 6000 | 0.0691 |
175
- | 0.0009 | 9.1754 | 6050 | 0.0693 |
176
- | 0.0007 | 9.2512 | 6100 | 0.0696 |
177
- | 0.0006 | 9.3270 | 6150 | 0.0702 |
178
- | 0.0002 | 9.4028 | 6200 | 0.0703 |
179
- | 0.0001 | 9.4787 | 6250 | 0.0706 |
180
- | 0.0006 | 9.5545 | 6300 | 0.0706 |
181
- | 0.0004 | 9.6303 | 6350 | 0.0705 |
182
- | 0.0008 | 9.7062 | 6400 | 0.0709 |
183
- | 0.0004 | 9.7820 | 6450 | 0.0709 |
184
- | 0.0002 | 9.8578 | 6500 | 0.0709 |
185
- | 0.0001 | 9.9336 | 6550 | 0.0708 |
186
 
187
 
188
  ### Framework versions
 
16
 
17
  # Llama-3.1-8B-Instruct-PsyCourse-fold5
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: 0.0499
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
+ | 0.6474 | 0.0758 | 50 | 0.4173 |
56
+ | 0.0891 | 0.1517 | 100 | 0.0932 |
57
+ | 0.0787 | 0.2275 | 150 | 0.0860 |
58
+ | 0.059 | 0.3033 | 200 | 0.0593 |
59
+ | 0.053 | 0.3791 | 250 | 0.0517 |
60
+ | 0.0489 | 0.4550 | 300 | 0.0463 |
61
+ | 0.0525 | 0.5308 | 350 | 0.0471 |
62
+ | 0.0651 | 0.6066 | 400 | 0.0472 |
63
+ | 0.0552 | 0.6825 | 450 | 0.0467 |
64
+ | 0.0317 | 0.7583 | 500 | 0.0416 |
65
+ | 0.0313 | 0.8341 | 550 | 0.0426 |
66
+ | 0.0441 | 0.9100 | 600 | 0.0429 |
67
+ | 0.0436 | 0.9858 | 650 | 0.0390 |
68
+ | 0.0324 | 1.0616 | 700 | 0.0417 |
69
+ | 0.0356 | 1.1374 | 750 | 0.0421 |
70
+ | 0.0368 | 1.2133 | 800 | 0.0378 |
71
+ | 0.0322 | 1.2891 | 850 | 0.0418 |
72
+ | 0.0242 | 1.3649 | 900 | 0.0404 |
73
+ | 0.0278 | 1.4408 | 950 | 0.0395 |
74
+ | 0.0391 | 1.5166 | 1000 | 0.0356 |
75
+ | 0.0294 | 1.5924 | 1050 | 0.0360 |
76
+ | 0.023 | 1.6682 | 1100 | 0.0359 |
77
+ | 0.0282 | 1.7441 | 1150 | 0.0372 |
78
+ | 0.0343 | 1.8199 | 1200 | 0.0359 |
79
+ | 0.0335 | 1.8957 | 1250 | 0.0339 |
80
+ | 0.0389 | 1.9716 | 1300 | 0.0357 |
81
+ | 0.0277 | 2.0474 | 1350 | 0.0351 |
82
+ | 0.0193 | 2.1232 | 1400 | 0.0343 |
83
+ | 0.0211 | 2.1991 | 1450 | 0.0354 |
84
+ | 0.0149 | 2.2749 | 1500 | 0.0352 |
85
+ | 0.0258 | 2.3507 | 1550 | 0.0337 |
86
+ | 0.0255 | 2.4265 | 1600 | 0.0359 |
87
+ | 0.014 | 2.5024 | 1650 | 0.0377 |
88
+ | 0.0265 | 2.5782 | 1700 | 0.0336 |
89
+ | 0.0211 | 2.6540 | 1750 | 0.0344 |
90
+ | 0.0278 | 2.7299 | 1800 | 0.0355 |
91
+ | 0.0253 | 2.8057 | 1850 | 0.0363 |
92
+ | 0.0178 | 2.8815 | 1900 | 0.0345 |
93
+ | 0.0302 | 2.9573 | 1950 | 0.0340 |
94
+ | 0.0091 | 3.0332 | 2000 | 0.0358 |
95
+ | 0.0072 | 3.1090 | 2050 | 0.0411 |
96
+ | 0.0126 | 3.1848 | 2100 | 0.0394 |
97
+ | 0.0108 | 3.2607 | 2150 | 0.0404 |
98
+ | 0.0101 | 3.3365 | 2200 | 0.0381 |
99
+ | 0.0077 | 3.4123 | 2250 | 0.0382 |
100
+ | 0.0096 | 3.4882 | 2300 | 0.0379 |
101
+ | 0.0077 | 3.5640 | 2350 | 0.0392 |
102
+ | 0.015 | 3.6398 | 2400 | 0.0381 |
103
+ | 0.0095 | 3.7156 | 2450 | 0.0401 |
104
+ | 0.0174 | 3.7915 | 2500 | 0.0395 |
105
+ | 0.0105 | 3.8673 | 2550 | 0.0393 |
106
+ | 0.014 | 3.9431 | 2600 | 0.0385 |
107
+ | 0.0086 | 4.0190 | 2650 | 0.0402 |
108
+ | 0.005 | 4.0948 | 2700 | 0.0439 |
109
+ | 0.0044 | 4.1706 | 2750 | 0.0487 |
110
+ | 0.0032 | 4.2464 | 2800 | 0.0490 |
111
+ | 0.0052 | 4.3223 | 2850 | 0.0489 |
112
+ | 0.0089 | 4.3981 | 2900 | 0.0493 |
113
+ | 0.0095 | 4.4739 | 2950 | 0.0490 |
114
+ | 0.0049 | 4.5498 | 3000 | 0.0487 |
115
+ | 0.0025 | 4.6256 | 3050 | 0.0492 |
116
+ | 0.0066 | 4.7014 | 3100 | 0.0498 |
117
+ | 0.0089 | 4.7773 | 3150 | 0.0500 |
118
+ | 0.0021 | 4.8531 | 3200 | 0.0500 |
119
+ | 0.0056 | 4.9289 | 3250 | 0.0499 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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:c22dc6d916fade13ea37af66f27bb14316d56fb21e4cbfaa43a3c08b6fbdaf31
3
  size 83945296
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c48d19995e1cf8ea6de387d52f191534eeb36183579276c1713315e3ac323cf
3
  size 83945296