Toflamus/GPT-2_para3M
Browse files- README.md +110 -6
- merges.txt +1 -1
- pytorch_model.bin +1 -1
- special_tokens_map.json +3 -21
- tokenizer.json +1 -2
- tokenizer_config.json +4 -28
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 2.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -42,20 +42,124 @@ The following hyperparameters were used during training:
|
|
| 42 |
- total_train_batch_size: 256
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: cosine
|
| 45 |
-
- lr_scheduler_warmup_steps:
|
| 46 |
- num_epochs: 1
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
|:-------------:|:-----:|:-----:|:---------------:|
|
| 52 |
-
|
|
| 53 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
|
| 55 |
|
| 56 |
### Framework versions
|
| 57 |
|
| 58 |
- Transformers 4.32.0
|
| 59 |
-
- Pytorch 2.0.1+
|
| 60 |
- Datasets 2.14.4
|
| 61 |
-
- Tokenizers 0.13.
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 2.3207
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 42 |
- total_train_batch_size: 256
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: cosine
|
| 45 |
+
- lr_scheduler_warmup_steps: 100
|
| 46 |
- num_epochs: 1
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
|:-------------:|:-----:|:-----:|:---------------:|
|
| 52 |
+
| 9.6976 | 0.01 | 100 | 7.7754 |
|
| 53 |
+
| 6.488 | 0.02 | 200 | 5.7795 |
|
| 54 |
+
| 5.3705 | 0.03 | 300 | 4.8609 |
|
| 55 |
+
| 4.5632 | 0.04 | 400 | 4.2544 |
|
| 56 |
+
| 4.141 | 0.05 | 500 | 3.9425 |
|
| 57 |
+
| 3.902 | 0.06 | 600 | 3.7189 |
|
| 58 |
+
| 3.7074 | 0.07 | 700 | 3.5514 |
|
| 59 |
+
| 3.5716 | 0.08 | 800 | 3.4291 |
|
| 60 |
+
| 3.4695 | 0.08 | 900 | 3.3253 |
|
| 61 |
+
| 3.3847 | 0.09 | 1000 | 3.2311 |
|
| 62 |
+
| 3.2974 | 0.1 | 1100 | 3.1595 |
|
| 63 |
+
| 3.2318 | 0.11 | 1200 | 3.0909 |
|
| 64 |
+
| 3.1698 | 0.12 | 1300 | 3.0329 |
|
| 65 |
+
| 3.1258 | 0.13 | 1400 | 2.9879 |
|
| 66 |
+
| 3.0802 | 0.14 | 1500 | 2.9396 |
|
| 67 |
+
| 3.046 | 0.15 | 1600 | 2.9017 |
|
| 68 |
+
| 3.0047 | 0.16 | 1700 | 2.8652 |
|
| 69 |
+
| 2.9701 | 0.17 | 1800 | 2.8320 |
|
| 70 |
+
| 2.9425 | 0.18 | 1900 | 2.8048 |
|
| 71 |
+
| 2.9141 | 0.19 | 2000 | 2.7757 |
|
| 72 |
+
| 2.8896 | 0.2 | 2100 | 2.7515 |
|
| 73 |
+
| 2.8667 | 0.21 | 2200 | 2.7263 |
|
| 74 |
+
| 2.8443 | 0.22 | 2300 | 2.7066 |
|
| 75 |
+
| 2.8288 | 0.23 | 2400 | 2.6815 |
|
| 76 |
+
| 2.8044 | 0.24 | 2500 | 2.6620 |
|
| 77 |
+
| 2.7886 | 0.25 | 2600 | 2.6471 |
|
| 78 |
+
| 2.7732 | 0.25 | 2700 | 2.6283 |
|
| 79 |
+
| 2.7576 | 0.26 | 2800 | 2.6101 |
|
| 80 |
+
| 2.7479 | 0.27 | 2900 | 2.5978 |
|
| 81 |
+
| 2.7256 | 0.28 | 3000 | 2.5819 |
|
| 82 |
+
| 2.7179 | 0.29 | 3100 | 2.5688 |
|
| 83 |
+
| 2.707 | 0.3 | 3200 | 2.5595 |
|
| 84 |
+
| 2.6921 | 0.31 | 3300 | 2.5471 |
|
| 85 |
+
| 2.6809 | 0.32 | 3400 | 2.5329 |
|
| 86 |
+
| 2.6779 | 0.33 | 3500 | 2.5232 |
|
| 87 |
+
| 2.663 | 0.34 | 3600 | 2.5154 |
|
| 88 |
+
| 2.6554 | 0.35 | 3700 | 2.5030 |
|
| 89 |
+
| 2.6437 | 0.36 | 3800 | 2.4967 |
|
| 90 |
+
| 2.6346 | 0.37 | 3900 | 2.4859 |
|
| 91 |
+
| 2.6293 | 0.38 | 4000 | 2.4768 |
|
| 92 |
+
| 2.6221 | 0.39 | 4100 | 2.4709 |
|
| 93 |
+
| 2.6178 | 0.4 | 4200 | 2.4623 |
|
| 94 |
+
| 2.6076 | 0.41 | 4300 | 2.4586 |
|
| 95 |
+
| 2.6025 | 0.41 | 4400 | 2.4492 |
|
| 96 |
+
| 2.5907 | 0.42 | 4500 | 2.4409 |
|
| 97 |
+
| 2.5896 | 0.43 | 4600 | 2.4369 |
|
| 98 |
+
| 2.5816 | 0.44 | 4700 | 2.4316 |
|
| 99 |
+
| 2.5783 | 0.45 | 4800 | 2.4256 |
|
| 100 |
+
| 2.577 | 0.46 | 4900 | 2.4204 |
|
| 101 |
+
| 2.5685 | 0.47 | 5000 | 2.4150 |
|
| 102 |
+
| 2.567 | 0.48 | 5100 | 2.4093 |
|
| 103 |
+
| 2.5564 | 0.49 | 5200 | 2.4059 |
|
| 104 |
+
| 2.5556 | 0.5 | 5300 | 2.4012 |
|
| 105 |
+
| 2.5496 | 0.51 | 5400 | 2.3997 |
|
| 106 |
+
| 2.545 | 0.52 | 5500 | 2.3956 |
|
| 107 |
+
| 2.5473 | 0.53 | 5600 | 2.3905 |
|
| 108 |
+
| 2.5389 | 0.54 | 5700 | 2.3856 |
|
| 109 |
+
| 2.5373 | 0.55 | 5800 | 2.3818 |
|
| 110 |
+
| 2.5318 | 0.56 | 5900 | 2.3787 |
|
| 111 |
+
| 2.5313 | 0.57 | 6000 | 2.3751 |
|
| 112 |
+
| 2.5285 | 0.58 | 6100 | 2.3722 |
|
| 113 |
+
| 2.5318 | 0.58 | 6200 | 2.3687 |
|
| 114 |
+
| 2.5229 | 0.59 | 6300 | 2.3666 |
|
| 115 |
+
| 2.5194 | 0.6 | 6400 | 2.3632 |
|
| 116 |
+
| 2.5174 | 0.61 | 6500 | 2.3598 |
|
| 117 |
+
| 2.5169 | 0.62 | 6600 | 2.3567 |
|
| 118 |
+
| 2.511 | 0.63 | 6700 | 2.3552 |
|
| 119 |
+
| 2.5093 | 0.64 | 6800 | 2.3546 |
|
| 120 |
+
| 2.5114 | 0.65 | 6900 | 2.3528 |
|
| 121 |
+
| 2.5064 | 0.66 | 7000 | 2.3492 |
|
| 122 |
+
| 2.507 | 0.67 | 7100 | 2.3483 |
|
| 123 |
+
| 2.502 | 0.68 | 7200 | 2.3445 |
|
| 124 |
+
| 2.4964 | 0.69 | 7300 | 2.3448 |
|
| 125 |
+
| 2.4999 | 0.7 | 7400 | 2.3423 |
|
| 126 |
+
| 2.4961 | 0.71 | 7500 | 2.3407 |
|
| 127 |
+
| 2.489 | 0.72 | 7600 | 2.3386 |
|
| 128 |
+
| 2.4926 | 0.73 | 7700 | 2.3384 |
|
| 129 |
+
| 2.4919 | 0.74 | 7800 | 2.3365 |
|
| 130 |
+
| 2.491 | 0.74 | 7900 | 2.3349 |
|
| 131 |
+
| 2.4893 | 0.75 | 8000 | 2.3333 |
|
| 132 |
+
| 2.4909 | 0.76 | 8100 | 2.3318 |
|
| 133 |
+
| 2.4862 | 0.77 | 8200 | 2.3305 |
|
| 134 |
+
| 2.4884 | 0.78 | 8300 | 2.3299 |
|
| 135 |
+
| 2.49 | 0.79 | 8400 | 2.3280 |
|
| 136 |
+
| 2.4788 | 0.8 | 8500 | 2.3286 |
|
| 137 |
+
| 2.4865 | 0.81 | 8600 | 2.3272 |
|
| 138 |
+
| 2.4823 | 0.82 | 8700 | 2.3263 |
|
| 139 |
+
| 2.4844 | 0.83 | 8800 | 2.3255 |
|
| 140 |
+
| 2.4826 | 0.84 | 8900 | 2.3251 |
|
| 141 |
+
| 2.4844 | 0.85 | 9000 | 2.3243 |
|
| 142 |
+
| 2.4798 | 0.86 | 9100 | 2.3231 |
|
| 143 |
+
| 2.4864 | 0.87 | 9200 | 2.3231 |
|
| 144 |
+
| 2.4755 | 0.88 | 9300 | 2.3228 |
|
| 145 |
+
| 2.4735 | 0.89 | 9400 | 2.3228 |
|
| 146 |
+
| 2.4786 | 0.9 | 9500 | 2.3224 |
|
| 147 |
+
| 2.4791 | 0.91 | 9600 | 2.3222 |
|
| 148 |
+
| 2.4809 | 0.91 | 9700 | 2.3214 |
|
| 149 |
+
| 2.4778 | 0.92 | 9800 | 2.3213 |
|
| 150 |
+
| 2.4777 | 0.93 | 9900 | 2.3211 |
|
| 151 |
+
| 2.4798 | 0.94 | 10000 | 2.3209 |
|
| 152 |
+
| 2.4768 | 0.95 | 10100 | 2.3212 |
|
| 153 |
+
| 2.4808 | 0.96 | 10200 | 2.3209 |
|
| 154 |
+
| 2.4762 | 0.97 | 10300 | 2.3208 |
|
| 155 |
+
| 2.4778 | 0.98 | 10400 | 2.3208 |
|
| 156 |
+
| 2.4816 | 0.99 | 10500 | 2.3207 |
|
| 157 |
+
| 2.4728 | 1.0 | 10600 | 2.3207 |
|
| 158 |
|
| 159 |
|
| 160 |
### Framework versions
|
| 161 |
|
| 162 |
- Transformers 4.32.0
|
| 163 |
+
- Pytorch 2.0.1+cu117
|
| 164 |
- Datasets 2.14.4
|
| 165 |
+
- Tokenizers 0.13.2
|
merges.txt
CHANGED
|
@@ -1,4 +1,4 @@
|
|
| 1 |
-
#version: 0.2
|
| 2 |
Ġ t
|
| 3 |
Ġ a
|
| 4 |
h e
|
|
|
|
| 1 |
+
#version: 0.2 - Trained by `huggingface/tokenizers`
|
| 2 |
Ġ t
|
| 3 |
Ġ a
|
| 4 |
h e
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14352341
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14fff0e75573c30941f285deaa9e55953fbe83007725d41a1aee6a2b96c9a7fb
|
| 3 |
size 14352341
|
special_tokens_map.json
CHANGED
|
@@ -1,24 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"bos_token":
|
| 3 |
-
|
| 4 |
-
"lstrip": false,
|
| 5 |
-
"normalized": true,
|
| 6 |
-
"rstrip": false,
|
| 7 |
-
"single_word": false
|
| 8 |
-
},
|
| 9 |
-
"eos_token": {
|
| 10 |
-
"content": "<|endoftext|>",
|
| 11 |
-
"lstrip": false,
|
| 12 |
-
"normalized": true,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false
|
| 15 |
-
},
|
| 16 |
"pad_token": "<|endoftext|>",
|
| 17 |
-
"unk_token":
|
| 18 |
-
"content": "<|endoftext|>",
|
| 19 |
-
"lstrip": false,
|
| 20 |
-
"normalized": true,
|
| 21 |
-
"rstrip": false,
|
| 22 |
-
"single_word": false
|
| 23 |
-
}
|
| 24 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
}
|
tokenizer.json
CHANGED
|
@@ -9,7 +9,7 @@
|
|
| 9 |
"single_word": false,
|
| 10 |
"lstrip": false,
|
| 11 |
"rstrip": false,
|
| 12 |
-
"normalized":
|
| 13 |
"special": true
|
| 14 |
}
|
| 15 |
],
|
|
@@ -39,7 +39,6 @@
|
|
| 39 |
"continuing_subword_prefix": "",
|
| 40 |
"end_of_word_suffix": "",
|
| 41 |
"fuse_unk": false,
|
| 42 |
-
"byte_fallback": false,
|
| 43 |
"vocab": {
|
| 44 |
"!": 0,
|
| 45 |
"\"": 1,
|
|
|
|
| 9 |
"single_word": false,
|
| 10 |
"lstrip": false,
|
| 11 |
"rstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
"special": true
|
| 14 |
}
|
| 15 |
],
|
|
|
|
| 39 |
"continuing_subword_prefix": "",
|
| 40 |
"end_of_word_suffix": "",
|
| 41 |
"fuse_unk": false,
|
|
|
|
| 42 |
"vocab": {
|
| 43 |
"!": 0,
|
| 44 |
"\"": 1,
|
tokenizer_config.json
CHANGED
|
@@ -1,33 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"add_bos_token": false,
|
| 3 |
"add_prefix_space": false,
|
| 4 |
-
"bos_token":
|
| 5 |
-
"__type": "AddedToken",
|
| 6 |
-
"content": "<|endoftext|>",
|
| 7 |
-
"lstrip": false,
|
| 8 |
-
"normalized": true,
|
| 9 |
-
"rstrip": false,
|
| 10 |
-
"single_word": false
|
| 11 |
-
},
|
| 12 |
"clean_up_tokenization_spaces": true,
|
| 13 |
-
"eos_token":
|
| 14 |
-
|
| 15 |
-
"content": "<|endoftext|>",
|
| 16 |
-
"lstrip": false,
|
| 17 |
-
"normalized": true,
|
| 18 |
-
"rstrip": false,
|
| 19 |
-
"single_word": false
|
| 20 |
-
},
|
| 21 |
-
"errors": "replace",
|
| 22 |
-
"model_max_length": 2048,
|
| 23 |
-
"pad_token": null,
|
| 24 |
"tokenizer_class": "GPT2Tokenizer",
|
| 25 |
-
"unk_token":
|
| 26 |
-
"__type": "AddedToken",
|
| 27 |
-
"content": "<|endoftext|>",
|
| 28 |
-
"lstrip": false,
|
| 29 |
-
"normalized": true,
|
| 30 |
-
"rstrip": false,
|
| 31 |
-
"single_word": false
|
| 32 |
-
}
|
| 33 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<|endoftext|>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"model_max_length": 1024,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
"tokenizer_class": "GPT2Tokenizer",
|
| 8 |
+
"unk_token": "<|endoftext|>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4027
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c363c8942fdca6aa050728393ab72a9533fe8c19f2604ac1b1f8330fce72295
|
| 3 |
size 4027
|