Hugging Face's logo

achisingh06
/
druk-ai-20250628_0745

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
achisingh06 commited on
Commit
2ad68ee
·
verified ·
1 Parent(s): 6ef0674

Model save

Browse files
Files changed (1) hide show
  1. README.md +105 -0
README.md ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: cc-by-nc-4.0
4
+ base_model: facebook/nllb-200-distilled-600M
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: druk-ai-20250628_0745
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
+ # druk-ai-20250628_0745
16
+
17
+ This model is a fine-tuned version of [facebook/nllb-200-distilled-600M](https://huggingface.co/facebook/nllb-200-distilled-600M) on the None dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 2.2690
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 0.0005
39
+ - train_batch_size: 4
40
+ - eval_batch_size: 4
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 2
43
+ - total_train_batch_size: 8
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 100
47
+ - num_epochs: 3
48
+ - mixed_precision_training: Native AMP
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss |
53
+ |:-------------:|:------:|:----:|:---------------:|
54
+ | 4.6718 | 0.0684 | 50 | 3.9966 |
55
+ | 3.3909 | 0.1367 | 100 | 3.2014 |
56
+ | 3.2175 | 0.2051 | 150 | 2.9763 |
57
+ | 3.1066 | 0.2734 | 200 | 2.9230 |
58
+ | 3.058 | 0.3418 | 250 | 2.8082 |
59
+ | 2.9733 | 0.4101 | 300 | 2.7560 |
60
+ | 2.9797 | 0.4785 | 350 | 2.7420 |
61
+ | 2.714 | 0.5468 | 400 | 2.6686 |
62
+ | 2.8964 | 0.6152 | 450 | 2.6501 |
63
+ | 2.7973 | 0.6835 | 500 | 2.6197 |
64
+ | 2.7552 | 0.7519 | 550 | 2.5710 |
65
+ | 2.7453 | 0.8202 | 600 | 2.5410 |
66
+ | 2.9687 | 0.8886 | 650 | 2.5268 |
67
+ | 2.7995 | 0.9569 | 700 | 2.5237 |
68
+ | 2.5497 | 1.0253 | 750 | 2.5099 |
69
+ | 2.6585 | 1.0936 | 800 | 2.4769 |
70
+ | 2.7442 | 1.1620 | 850 | 2.4660 |
71
+ | 2.7224 | 1.2303 | 900 | 2.4511 |
72
+ | 2.704 | 1.2987 | 950 | 2.4375 |
73
+ | 2.5466 | 1.3671 | 1000 | 2.4223 |
74
+ | 2.3552 | 1.4354 | 1050 | 2.4044 |
75
+ | 2.6877 | 1.5038 | 1100 | 2.4021 |
76
+ | 2.2772 | 1.5721 | 1150 | 2.3974 |
77
+ | 2.5707 | 1.6405 | 1200 | 2.3753 |
78
+ | 2.5388 | 1.7088 | 1250 | 2.3624 |
79
+ | 2.4451 | 1.7772 | 1300 | 2.3741 |
80
+ | 2.6623 | 1.8455 | 1350 | 2.3595 |
81
+ | 2.2503 | 1.9139 | 1400 | 2.3445 |
82
+ | 2.4205 | 1.9822 | 1450 | 2.3315 |
83
+ | 2.2562 | 2.0506 | 1500 | 2.3277 |
84
+ | 2.2127 | 2.1189 | 1550 | 2.3287 |
85
+ | 2.4043 | 2.1873 | 1600 | 2.3091 |
86
+ | 2.3461 | 2.2556 | 1650 | 2.3168 |
87
+ | 2.5133 | 2.3240 | 1700 | 2.2984 |
88
+ | 2.4444 | 2.3923 | 1750 | 2.2961 |
89
+ | 2.3056 | 2.4607 | 1800 | 2.2970 |
90
+ | 2.4537 | 2.5290 | 1850 | 2.2844 |
91
+ | 2.3241 | 2.5974 | 1900 | 2.2835 |
92
+ | 2.2608 | 2.6658 | 1950 | 2.2756 |
93
+ | 2.3779 | 2.7341 | 2000 | 2.2758 |
94
+ | 2.3757 | 2.8025 | 2050 | 2.2691 |
95
+ | 2.2582 | 2.8708 | 2100 | 2.2710 |
96
+ | 2.3975 | 2.9392 | 2150 | 2.2690 |
97
+
98
+
99
+ ### Framework versions
100
+
101
+ - PEFT 0.13.2
102
+ - Transformers 4.45.0
103
+ - Pytorch 2.6.0+cu124
104
+ - Datasets 2.21.0
105
+ - Tokenizers 0.20.3