Hugging Face's logo

mtzig
/
eo_train1-20_eval1-10_lr1e-5

Transformers
Safetensors
nanogpt
Generated from Trainer
Model card Files
xet
Community
mtzig commited on
Commit
ca8cf0d
·
verified ·
1 Parent(s): f37fe4b

Model save

Browse files
Files changed (5) hide show
  1. README.md +120 -0
  2. config.json +19 -0
  3. generation_config.json +4 -0
  4. model.safetensors +3 -0
  5. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - accuracy
7
+ model-index:
8
+ - name: eo_train1-20_eval1-10_lr1e-5
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # eo_train1-20_eval1-10_lr1e-5
16
+
17
+ This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.5281
20
+ - Accuracy: 0.65
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 1e-05
40
+ - train_batch_size: 128
41
+ - eval_batch_size: 128
42
+ - seed: 7658372
43
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
44
+ - lr_scheduler_type: cosine
45
+ - lr_scheduler_warmup_ratio: 0.1
46
+ - training_steps: 6000
47
+
48
+ ### Training results
49
+
50
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
51
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
52
+ | No log | 0 | 0 | 2.8464 | 0.025 |
53
+ | 2.0752 | 100.0 | 100 | 2.0603 | 0.5 |
54
+ | 0.8893 | 200.0 | 200 | 0.8853 | 0.5 |
55
+ | 0.7677 | 300.0 | 300 | 0.7673 | 0.5 |
56
+ | 0.7337 | 400.0 | 400 | 0.7334 | 0.525 |
57
+ | 0.7085 | 500.0 | 500 | 0.7082 | 0.5 |
58
+ | 0.6828 | 600.0 | 600 | 0.6824 | 0.55 |
59
+ | 0.66 | 700.0 | 700 | 0.6603 | 0.55 |
60
+ | 0.6514 | 800.0 | 800 | 0.6516 | 0.575 |
61
+ | 0.643 | 900.0 | 900 | 0.6428 | 0.55 |
62
+ | 0.6375 | 1000.0 | 1000 | 0.6371 | 0.575 |
63
+ | 0.6321 | 1100.0 | 1100 | 0.6325 | 0.575 |
64
+ | 0.6276 | 1200.0 | 1200 | 0.6281 | 0.575 |
65
+ | 0.624 | 1300.0 | 1300 | 0.6233 | 0.6 |
66
+ | 0.6196 | 1400.0 | 1400 | 0.6196 | 0.6 |
67
+ | 0.6156 | 1500.0 | 1500 | 0.6153 | 0.575 |
68
+ | 0.6108 | 1600.0 | 1600 | 0.6110 | 0.625 |
69
+ | 0.6067 | 1700.0 | 1700 | 0.6064 | 0.625 |
70
+ | 0.6022 | 1800.0 | 1800 | 0.6031 | 0.625 |
71
+ | 0.5993 | 1900.0 | 1900 | 0.5991 | 0.625 |
72
+ | 0.5961 | 2000.0 | 2000 | 0.5961 | 0.6 |
73
+ | 0.5925 | 2100.0 | 2100 | 0.5939 | 0.6 |
74
+ | 0.5895 | 2200.0 | 2200 | 0.5905 | 0.6 |
75
+ | 0.5866 | 2300.0 | 2300 | 0.5870 | 0.6 |
76
+ | 0.584 | 2400.0 | 2400 | 0.5840 | 0.6 |
77
+ | 0.5812 | 2500.0 | 2500 | 0.5813 | 0.6 |
78
+ | 0.5787 | 2600.0 | 2600 | 0.5785 | 0.6 |
79
+ | 0.5758 | 2700.0 | 2700 | 0.5756 | 0.65 |
80
+ | 0.5706 | 2800.0 | 2800 | 0.5705 | 0.675 |
81
+ | 0.5668 | 2900.0 | 2900 | 0.5666 | 0.65 |
82
+ | 0.5629 | 3000.0 | 3000 | 0.5627 | 0.575 |
83
+ | 0.5609 | 3100.0 | 3100 | 0.5606 | 0.65 |
84
+ | 0.5594 | 3200.0 | 3200 | 0.5591 | 0.65 |
85
+ | 0.5576 | 3300.0 | 3300 | 0.5574 | 0.65 |
86
+ | 0.5562 | 3400.0 | 3400 | 0.5560 | 0.65 |
87
+ | 0.5543 | 3500.0 | 3500 | 0.5544 | 0.675 |
88
+ | 0.553 | 3600.0 | 3600 | 0.5529 | 0.7 |
89
+ | 0.5512 | 3700.0 | 3700 | 0.5512 | 0.65 |
90
+ | 0.5495 | 3800.0 | 3800 | 0.5493 | 0.65 |
91
+ | 0.5468 | 3900.0 | 3900 | 0.5467 | 0.6 |
92
+ | 0.545 | 4000.0 | 4000 | 0.5448 | 0.625 |
93
+ | 0.5424 | 4100.0 | 4100 | 0.5424 | 0.65 |
94
+ | 0.5407 | 4200.0 | 4200 | 0.5408 | 0.65 |
95
+ | 0.5387 | 4300.0 | 4300 | 0.5387 | 0.65 |
96
+ | 0.5371 | 4400.0 | 4400 | 0.5371 | 0.65 |
97
+ | 0.5358 | 4500.0 | 4500 | 0.5358 | 0.65 |
98
+ | 0.5346 | 4600.0 | 4600 | 0.5345 | 0.675 |
99
+ | 0.5335 | 4700.0 | 4700 | 0.5334 | 0.65 |
100
+ | 0.5325 | 4800.0 | 4800 | 0.5326 | 0.7 |
101
+ | 0.5316 | 4900.0 | 4900 | 0.5315 | 0.675 |
102
+ | 0.5307 | 5000.0 | 5000 | 0.5308 | 0.65 |
103
+ | 0.5301 | 5100.0 | 5100 | 0.5301 | 0.65 |
104
+ | 0.5295 | 5200.0 | 5200 | 0.5295 | 0.65 |
105
+ | 0.5291 | 5300.0 | 5300 | 0.5291 | 0.65 |
106
+ | 0.5288 | 5400.0 | 5400 | 0.5288 | 0.65 |
107
+ | 0.5285 | 5500.0 | 5500 | 0.5285 | 0.65 |
108
+ | 0.5283 | 5600.0 | 5600 | 0.5283 | 0.65 |
109
+ | 0.5282 | 5700.0 | 5700 | 0.5282 | 0.65 |
110
+ | 0.5281 | 5800.0 | 5800 | 0.5281 | 0.65 |
111
+ | 0.5281 | 5900.0 | 5900 | 0.5281 | 0.65 |
112
+ | 0.5281 | 6000.0 | 6000 | 0.5281 | 0.65 |
113
+
114
+
115
+ ### Framework versions
116
+
117
+ - Transformers 4.46.0
118
+ - Pytorch 2.5.1
119
+ - Datasets 3.1.0
120
+ - Tokenizers 0.20.1
config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "NanoGPT"
4
+ ],
5
+ "bias": true,
6
+ "block_size": 256,
7
+ "dropout": 0.0,
8
+ "mlp_dim": 4,
9
+ "model_type": "nanogpt",
10
+ "n_embd": 256,
11
+ "n_head": 2,
12
+ "n_layer": 1,
13
+ "nonlinearity": "RELU",
14
+ "torch_dtype": "float32",
15
+ "transformers_version": "4.46.0",
16
+ "use_NoPE": true,
17
+ "use_layernorm": true,
18
+ "vocab_size": 14
19
+ }
generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.46.0"
4
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c811e0da51de04bcad857c635b13c08e09748a338efd1f646ebd7a4d7735098
3
+ size 3191304
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9e0e7c8f7dba1698df41a20d39893814c33030511a5859d3d30335215530ff0
3
+ size 5240