Hugging Face's logo

Vitallyy
/
phi_1.5_mfr_instruct

PEFT
TensorBoard
Safetensors
phi
Generated from Trainer
Model card Files
xet
Metrics Community
Vitallyy commited on
Commit
bd29bf2
·
verified ·
1 Parent(s): 4ee9528

End of training

Browse files
Files changed (1) hide show
  1. README.md +17 -17
README.md CHANGED
@@ -1,7 +1,7 @@
1
  ---
2
- base_model: microsoft/phi-1_5
3
  library_name: peft
4
  license: mit
 
5
  tags:
6
  - generated_from_trainer
7
  model-index:
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 1.4898
20
 
21
  ## Model description
22
 
@@ -35,13 +35,13 @@ More information needed
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
- - learning_rate: 2e-05
39
  - train_batch_size: 2
40
  - eval_batch_size: 2
41
  - seed: 42
42
  - gradient_accumulation_steps: 4
43
  - total_train_batch_size: 8
44
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: cosine
46
  - training_steps: 2000
47
 
@@ -49,20 +49,20 @@ The following hyperparameters were used during training:
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
- | 1.4571 | 40.0 | 250 | 1.1032 |
53
- | 0.2487 | 80.0 | 500 | 1.3577 |
54
- | 0.1679 | 120.0 | 750 | 1.4287 |
55
- | 0.1586 | 160.0 | 1000 | 1.4562 |
56
- | 0.1544 | 200.0 | 1250 | 1.4729 |
57
- | 0.1521 | 240.0 | 1500 | 1.4843 |
58
- | 0.151 | 280.0 | 1750 | 1.4893 |
59
- | 0.1504 | 320.0 | 2000 | 1.4898 |
60
 
61
 
62
  ### Framework versions
63
 
64
- - PEFT 0.11.1
65
- - Transformers 4.41.2
66
- - Pytorch 2.3.0+cu121
67
- - Datasets 2.20.0
68
- - Tokenizers 0.19.1
 
1
  ---
 
2
  library_name: peft
3
  license: mit
4
+ base_model: microsoft/phi-1_5
5
  tags:
6
  - generated_from_trainer
7
  model-index:
 
16
 
17
  This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.4011
20
 
21
  ## Model description
22
 
 
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
+ - learning_rate: 1e-05
39
  - train_batch_size: 2
40
  - eval_batch_size: 2
41
  - seed: 42
42
  - gradient_accumulation_steps: 4
43
  - total_train_batch_size: 8
44
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
  - lr_scheduler_type: cosine
46
  - training_steps: 2000
47
 
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:----:|:---------------:|
52
+ | 3.31 | 50.0 | 250 | 1.2487 |
53
+ | 0.5629 | 100.0 | 500 | 0.4522 |
54
+ | 0.2176 | 150.0 | 750 | 0.4012 |
55
+ | 0.1899 | 200.0 | 1000 | 0.3989 |
56
+ | 0.1832 | 250.0 | 1250 | 0.3996 |
57
+ | 0.1801 | 300.0 | 1500 | 0.4006 |
58
+ | 0.1788 | 350.0 | 1750 | 0.4011 |
59
+ | 0.1784 | 400.0 | 2000 | 0.4011 |
60
 
61
 
62
  ### Framework versions
63
 
64
+ - PEFT 0.14.0
65
+ - Transformers 4.48.3
66
+ - Pytorch 2.5.1+cu124
67
+ - Datasets 3.3.2
68
+ - Tokenizers 0.21.0