Hugging Face's logo

jackshannon
/
phi-1_5-finetuned-question-generation

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files Community
jackshannon commited on
Commit
5b4d104
·
verified ·
1 Parent(s): 5331fe2

End of training

Browse files
Files changed (4) hide show
  1. README.md +59 -12
  2. adapter_config.json +2 -2
  3. adapter_model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 1.9079
22
 
23
  ## Model description
24
 
@@ -46,22 +46,69 @@ The following hyperparameters were used during training:
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 100
49
- - training_steps: 1000
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:-----:|:----:|:---------------:|
55
- | 3.0206 | 0.28 | 100 | 2.8708 |
56
- | 2.8007 | 0.56 | 200 | 2.6294 |
57
- | 2.5551 | 0.85 | 300 | 2.4552 |
58
- | 2.3554 | 1.13 | 400 | 2.2903 |
59
- | 2.2233 | 1.41 | 500 | 2.1771 |
60
- | 2.1114 | 1.69 | 600 | 2.0850 |
61
- | 2.0211 | 1.97 | 700 | 2.0163 |
62
- | 1.8995 | 2.25 | 800 | 1.9635 |
63
- | 1.835 | 2.54 | 900 | 1.9226 |
64
- | 1.81 | 2.82 | 1000 | 1.9079 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
65
 
66
 
67
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.9170
22
 
23
  ## Model description
24
 
 
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 100
49
+ - num_epochs: 5
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss |
54
  |:-------------:|:-----:|:----:|:---------------:|
55
+ | 2.9836 | 0.09 | 100 | 2.8641 |
56
+ | 2.8536 | 0.17 | 200 | 2.7929 |
57
+ | 2.8051 | 0.26 | 300 | 2.7567 |
58
+ | 2.7782 | 0.35 | 400 | 2.7092 |
59
+ | 2.7542 | 0.44 | 500 | 2.6946 |
60
+ | 2.6978 | 0.52 | 600 | 2.6719 |
61
+ | 2.6833 | 0.61 | 700 | 2.6497 |
62
+ | 2.6504 | 0.7 | 800 | 2.6172 |
63
+ | 2.6228 | 0.78 | 900 | 2.6008 |
64
+ | 2.6219 | 0.87 | 1000 | 2.5802 |
65
+ | 2.5629 | 0.96 | 1100 | 2.5519 |
66
+ | 2.5315 | 1.05 | 1200 | 2.5255 |
67
+ | 2.4813 | 1.13 | 1300 | 2.5156 |
68
+ | 2.4539 | 1.22 | 1400 | 2.4884 |
69
+ | 2.4466 | 1.31 | 1500 | 2.4660 |
70
+ | 2.4205 | 1.39 | 1600 | 2.4431 |
71
+ | 2.3937 | 1.48 | 1700 | 2.4238 |
72
+ | 2.3686 | 1.57 | 1800 | 2.4069 |
73
+ | 2.3209 | 1.66 | 1900 | 2.3826 |
74
+ | 2.3409 | 1.74 | 2000 | 2.3606 |
75
+ | 2.2874 | 1.83 | 2100 | 2.3453 |
76
+ | 2.309 | 1.92 | 2200 | 2.3222 |
77
+ | 2.2676 | 2.01 | 2300 | 2.2981 |
78
+ | 2.1734 | 2.09 | 2400 | 2.2892 |
79
+ | 2.1495 | 2.18 | 2500 | 2.2549 |
80
+ | 2.1163 | 2.27 | 2600 | 2.2401 |
81
+ | 2.1 | 2.35 | 2700 | 2.2317 |
82
+ | 2.1046 | 2.44 | 2800 | 2.2153 |
83
+ | 2.1138 | 2.53 | 2900 | 2.1938 |
84
+ | 2.0691 | 2.62 | 3000 | 2.1775 |
85
+ | 2.0945 | 2.7 | 3100 | 2.1563 |
86
+ | 2.045 | 2.79 | 3200 | 2.1408 |
87
+ | 2.0212 | 2.88 | 3300 | 2.1229 |
88
+ | 2.0011 | 2.96 | 3400 | 2.1156 |
89
+ | 1.983 | 3.05 | 3500 | 2.0942 |
90
+ | 1.9309 | 3.14 | 3600 | 2.0769 |
91
+ | 1.8844 | 3.23 | 3700 | 2.0709 |
92
+ | 1.9085 | 3.31 | 3800 | 2.0589 |
93
+ | 1.8827 | 3.4 | 3900 | 2.0405 |
94
+ | 1.8511 | 3.49 | 4000 | 2.0310 |
95
+ | 1.8807 | 3.57 | 4100 | 2.0170 |
96
+ | 1.8437 | 3.66 | 4200 | 2.0045 |
97
+ | 1.8667 | 3.75 | 4300 | 2.0036 |
98
+ | 1.8081 | 3.84 | 4400 | 1.9886 |
99
+ | 1.8688 | 3.92 | 4500 | 1.9767 |
100
+ | 1.8187 | 4.01 | 4600 | 1.9652 |
101
+ | 1.7511 | 4.1 | 4700 | 1.9592 |
102
+ | 1.7384 | 4.18 | 4800 | 1.9558 |
103
+ | 1.7843 | 4.27 | 4900 | 1.9474 |
104
+ | 1.7389 | 4.36 | 5000 | 1.9412 |
105
+ | 1.7465 | 4.45 | 5100 | 1.9346 |
106
+ | 1.7483 | 4.53 | 5200 | 1.9290 |
107
+ | 1.7149 | 4.62 | 5300 | 1.9246 |
108
+ | 1.7154 | 4.71 | 5400 | 1.9211 |
109
+ | 1.7637 | 4.8 | 5500 | 1.9188 |
110
+ | 1.7559 | 4.88 | 5600 | 1.9181 |
111
+ | 1.7204 | 4.97 | 5700 | 1.9170 |
112
 
113
 
114
  ### Framework versions
adapter_config.json CHANGED
@@ -20,10 +20,10 @@
20
  "revision": null,
21
  "target_modules": [
22
  "v_proj",
 
23
  "dense",
24
- "fc1",
25
  "q_proj",
26
- "k_proj",
27
  "fc2"
28
  ],
29
  "task_type": "CAUSAL_LM",
 
20
  "revision": null,
21
  "target_modules": [
22
  "v_proj",
23
+ "k_proj",
24
  "dense",
 
25
  "q_proj",
26
+ "fc1",
27
  "fc2"
28
  ],
29
  "task_type": "CAUSAL_LM",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5cd45520bede814b83497b26a30c5678f13df6261fea94d453a5587fc543f176
3
  size 56660888
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32d22604e71c0ed518393d6d1488b9015450e26a985ed7c58708f173bff03707
3
  size 56660888
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5298e0f030dae0ac073a6d5fa9075cb92719fbbc01f98c6f8eead4f7a77bce57
3
  size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d8a6346625b0d0cf6bb61dc0c804b72803ad60c8336003827e98a9d80f15e4f
3
  size 4920