Hugging Face's logo

ninagroot
/
Baby-Llama-58M-RUN1

Text Generation
Transformers
Safetensors
llama
Generated from Trainer
text-generation-inference
Model card Files
xet
Community
ninagroot commited on
Commit
4a22763
·
verified ·
1 Parent(s): 6534de0

ninagroot/babyllamatest

Browse files
Files changed (3) hide show
  1. README.md +82 -42
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
- - Loss: 12.7407
17
 
18
  ## Model description
19
 
@@ -39,53 +39,93 @@ The following hyperparameters were used during training:
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: cosine
41
  - lr_scheduler_warmup_steps: 50
42
- - num_epochs: 40
43
  - mixed_precision_training: Native AMP
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:-----:|:----:|:---------------:|
49
- | 81.6068 | 1.0 | 2 | 75.8719 |
50
- | 81.1403 | 2.0 | 4 | 74.6429 |
51
- | 78.4746 | 3.0 | 6 | 72.5604 |
52
- | 78.6147 | 4.0 | 8 | 69.6859 |
53
- | 75.1485 | 5.0 | 10 | 67.9944 |
54
- | 73.5182 | 6.0 | 12 | 64.5075 |
55
- | 69.6393 | 7.0 | 14 | 61.3852 |
56
- | 66.9895 | 8.0 | 16 | 58.5262 |
57
- | 64.4746 | 9.0 | 18 | 55.6940 |
58
- | 60.8097 | 10.0 | 20 | 52.6993 |
59
- | 57.1714 | 11.0 | 22 | 49.5786 |
60
- | 53.8474 | 12.0 | 24 | 46.5081 |
61
- | 49.9873 | 13.0 | 26 | 43.6358 |
62
- | 48.7366 | 14.0 | 28 | 41.0406 |
63
- | 45.0539 | 15.0 | 30 | 38.7263 |
64
- | 44.0504 | 16.0 | 32 | 36.6352 |
65
- | 40.9533 | 17.0 | 34 | 34.6685 |
66
- | 39.9931 | 18.0 | 36 | 32.7875 |
67
- | 38.116 | 19.0 | 38 | 30.8567 |
68
- | 35.4181 | 20.0 | 40 | 28.9705 |
69
- | 34.0383 | 21.0 | 42 | 27.4282 |
70
- | 30.7991 | 22.0 | 44 | 26.4171 |
71
- | 29.8348 | 23.0 | 46 | 24.9225 |
72
- | 27.9282 | 24.0 | 48 | 23.9103 |
73
- | 25.8511 | 25.0 | 50 | 22.9495 |
74
- | 25.1711 | 26.0 | 52 | 21.5530 |
75
- | 24.2361 | 27.0 | 54 | 20.5871 |
76
- | 21.9294 | 28.0 | 56 | 19.0727 |
77
- | 20.435 | 29.0 | 58 | 18.0482 |
78
- | 18.682 | 30.0 | 60 | 17.0037 |
79
- | 17.4144 | 31.0 | 62 | 16.0468 |
80
- | 16.4872 | 32.0 | 64 | 15.2828 |
81
- | 16.2417 | 33.0 | 66 | 14.6359 |
82
- | 15.1244 | 34.0 | 68 | 14.1234 |
83
- | 14.0602 | 35.0 | 70 | 13.5799 |
84
- | 13.7722 | 36.0 | 72 | 13.3509 |
85
- | 13.377 | 37.0 | 74 | 12.9960 |
86
- | 13.4091 | 38.0 | 76 | 12.8183 |
87
- | 13.1398 | 39.0 | 78 | 12.7614 |
88
- | 13.1002 | 40.0 | 80 | 12.7407 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
89
 
90
 
91
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Loss: 6.1610
17
 
18
  ## Model description
19
 
 
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: cosine
41
  - lr_scheduler_warmup_steps: 50
42
+ - num_epochs: 80
43
  - mixed_precision_training: Native AMP
44
 
45
  ### Training results
46
 
47
  | Training Loss | Epoch | Step | Validation Loss |
48
  |:-------------:|:-----:|:----:|:---------------:|
49
+ | 81.7512 | 1.0 | 2 | 74.4291 |
50
+ | 81.3083 | 2.0 | 4 | 73.3596 |
51
+ | 78.6216 | 3.0 | 6 | 71.5365 |
52
+ | 80.396 | 4.0 | 8 | 70.3538 |
53
+ | 75.3713 | 5.0 | 10 | 67.4044 |
54
+ | 74.0418 | 6.0 | 12 | 64.0233 |
55
+ | 70.1637 | 7.0 | 14 | 60.8437 |
56
+ | 67.5864 | 8.0 | 16 | 57.9300 |
57
+ | 64.8984 | 9.0 | 18 | 55.0383 |
58
+ | 61.2535 | 10.0 | 20 | 52.0253 |
59
+ | 57.6171 | 11.0 | 22 | 48.9365 |
60
+ | 54.2922 | 12.0 | 24 | 45.8747 |
61
+ | 50.3849 | 13.0 | 26 | 43.0132 |
62
+ | 49.0703 | 14.0 | 28 | 40.4715 |
63
+ | 45.5158 | 15.0 | 30 | 38.1415 |
64
+ | 44.3002 | 16.0 | 32 | 35.9572 |
65
+ | 41.2208 | 17.0 | 34 | 33.8684 |
66
+ | 39.8837 | 18.0 | 36 | 31.8991 |
67
+ | 38.1152 | 19.0 | 38 | 29.8574 |
68
+ | 35.239 | 20.0 | 40 | 28.0249 |
69
+ | 33.6748 | 21.0 | 42 | 26.4792 |
70
+ | 30.4729 | 22.0 | 44 | 25.4216 |
71
+ | 29.436 | 23.0 | 46 | 24.1119 |
72
+ | 27.72 | 24.0 | 48 | 22.8196 |
73
+ | 25.5231 | 25.0 | 50 | 21.7862 |
74
+ | 24.8119 | 26.0 | 52 | 20.4891 |
75
+ | 23.3658 | 27.0 | 54 | 19.3795 |
76
+ | 21.4143 | 28.0 | 56 | 18.1634 |
77
+ | 20.032 | 29.0 | 58 | 17.0348 |
78
+ | 18.43 | 30.0 | 60 | 16.1163 |
79
+ | 16.897 | 31.0 | 62 | 15.2508 |
80
+ | 15.7483 | 32.0 | 64 | 14.3147 |
81
+ | 15.1794 | 33.0 | 66 | 13.5753 |
82
+ | 13.7129 | 34.0 | 68 | 12.8868 |
83
+ | 12.6031 | 35.0 | 70 | 12.6810 |
84
+ | 11.8192 | 36.0 | 72 | 11.9060 |
85
+ | 11.6487 | 37.0 | 74 | 11.3454 |
86
+ | 10.9525 | 38.0 | 76 | 10.8465 |
87
+ | 10.2164 | 39.0 | 78 | 10.1026 |
88
+ | 9.5492 | 40.0 | 80 | 9.6511 |
89
+ | 9.0438 | 41.0 | 82 | 9.2800 |
90
+ | 8.6141 | 42.0 | 84 | 8.8036 |
91
+ | 7.9373 | 43.0 | 86 | 8.6612 |
92
+ | 7.5371 | 44.0 | 88 | 8.1757 |
93
+ | 7.3186 | 45.0 | 90 | 8.1665 |
94
+ | 7.033 | 46.0 | 92 | 7.7424 |
95
+ | 6.7923 | 47.0 | 94 | 7.6650 |
96
+ | 6.4384 | 48.0 | 96 | 7.4306 |
97
+ | 6.2449 | 49.0 | 98 | 7.4175 |
98
+ | 6.1012 | 50.0 | 100 | 7.1466 |
99
+ | 6.0502 | 51.0 | 102 | 7.1740 |
100
+ | 5.7839 | 52.0 | 104 | 6.9619 |
101
+ | 5.6905 | 53.0 | 106 | 6.9416 |
102
+ | 5.665 | 54.0 | 108 | 6.7945 |
103
+ | 5.5401 | 55.0 | 110 | 6.7485 |
104
+ | 5.4773 | 56.0 | 112 | 6.6674 |
105
+ | 5.4169 | 57.0 | 114 | 6.6132 |
106
+ | 5.3628 | 58.0 | 116 | 6.5787 |
107
+ | 5.2021 | 59.0 | 118 | 6.4972 |
108
+ | 5.2817 | 60.0 | 120 | 6.4866 |
109
+ | 5.1901 | 61.0 | 122 | 6.4256 |
110
+ | 5.1268 | 62.0 | 124 | 6.3659 |
111
+ | 5.1105 | 63.0 | 126 | 6.3563 |
112
+ | 5.0539 | 64.0 | 128 | 6.3159 |
113
+ | 4.9715 | 65.0 | 130 | 6.3178 |
114
+ | 4.872 | 66.0 | 132 | 6.2741 |
115
+ | 4.9422 | 67.0 | 134 | 6.2699 |
116
+ | 4.944 | 68.0 | 136 | 6.2551 |
117
+ | 4.9487 | 69.0 | 138 | 6.2148 |
118
+ | 4.8968 | 70.0 | 140 | 6.2089 |
119
+ | 4.822 | 71.0 | 142 | 6.2093 |
120
+ | 4.965 | 72.0 | 144 | 6.1853 |
121
+ | 4.8401 | 73.0 | 146 | 6.1747 |
122
+ | 4.8539 | 74.0 | 148 | 6.1738 |
123
+ | 4.7751 | 75.0 | 150 | 6.1674 |
124
+ | 4.8871 | 76.0 | 152 | 6.1644 |
125
+ | 4.9347 | 77.0 | 154 | 6.1618 |
126
+ | 4.8009 | 78.0 | 156 | 6.1613 |
127
+ | 4.8121 | 79.0 | 158 | 6.1610 |
128
+ | 4.8048 | 80.0 | 160 | 6.1610 |
129
 
130
 
131
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b95b266c906f8a10c9a06b7ef00f57ba328d35b02576511cc7d51b8ef129717f
3
  size 217819016
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53a8b889e06337de6b9af958e4990a6791a3350eaf51ebd772cab14ea3a2988a
3
  size 217819016
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6d49873d6da6e46a5804e20834fe83e4d20fbe84a83bb526ea7410b0ff377414
3
  size 4984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a0208eef6e56a2ded47925bdfa0c4203cf9442e7d67cc216a9b300f64cce77e
3
  size 4984