Hugging Face's logo

G-ML-Hyly
/
stg-cli13b-t7-cdp-ca.dt.hlms.cln.inter-b4s1e1-20240102-0727

PEFT
Safetensors
Generated from Trainer
Model card Files Community
G-ML-Hyly commited on
Commit
b62bc5c
•
1 Parent(s): 89a3586

End of training

Browse files
Files changed (2) hide show
  1. README.md +132 -0
  2. adapter_model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,132 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: llama2
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: codellama/CodeLlama-13b-Instruct-hf
7
+ model-index:
8
+ - name: stg-cli13b-t7-cdp-ca.dt.hlms.cln.inter-b4s1e1-20240102-0727
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
+ # stg-cli13b-t7-cdp-ca.dt.hlms.cln.inter-b4s1e1-20240102-0727
16
+
17
+ This model is a fine-tuned version of [codellama/CodeLlama-13b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-13b-Instruct-hf) on the None dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.0656
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: 0.0002
39
+ - train_batch_size: 4
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
+ - lr_scheduler_type: cosine
44
+ - num_epochs: 1
45
+ - mixed_precision_training: Native AMP
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss |
50
+ |:-------------:|:-----:|:----:|:---------------:|
51
+ | 0.4007 | 0.02 | 100 | 0.0985 |
52
+ | 0.0888 | 0.04 | 200 | 0.0823 |
53
+ | 0.0834 | 0.05 | 300 | 0.0800 |
54
+ | 0.0783 | 0.07 | 400 | 0.0799 |
55
+ | 0.0782 | 0.09 | 500 | 0.0755 |
56
+ | 0.0798 | 0.11 | 600 | 0.0771 |
57
+ | 0.077 | 0.13 | 700 | 0.0734 |
58
+ | 0.0747 | 0.14 | 800 | 0.0745 |
59
+ | 0.076 | 0.16 | 900 | 0.0727 |
60
+ | 0.0791 | 0.18 | 1000 | 0.0775 |
61
+ | 0.0752 | 0.2 | 1100 | 0.0717 |
62
+ | 0.0721 | 0.22 | 1200 | 0.0729 |
63
+ | 0.0731 | 0.23 | 1300 | 0.0710 |
64
+ | 0.0832 | 0.25 | 1400 | 0.0727 |
65
+ | 0.0722 | 0.27 | 1500 | 0.0715 |
66
+ | 0.0738 | 0.29 | 1600 | 0.0715 |
67
+ | 0.071 | 0.31 | 1700 | 0.0705 |
68
+ | 0.0738 | 0.32 | 1800 | 0.0713 |
69
+ | 0.075 | 0.34 | 1900 | 0.0710 |
70
+ | 0.0732 | 0.36 | 2000 | 0.0703 |
71
+ | 0.0712 | 0.38 | 2100 | 0.0701 |
72
+ | 0.0702 | 0.4 | 2200 | 0.0699 |
73
+ | 0.0733 | 0.41 | 2300 | 0.0697 |
74
+ | 0.0739 | 0.43 | 2400 | 0.0691 |
75
+ | 0.0688 | 0.45 | 2500 | 0.0684 |
76
+ | 0.0692 | 0.47 | 2600 | 0.0689 |
77
+ | 0.0727 | 0.49 | 2700 | 0.0690 |
78
+ | 0.073 | 0.5 | 2800 | 0.0685 |
79
+ | 0.0752 | 0.52 | 2900 | 0.0691 |
80
+ | 0.0696 | 0.54 | 3000 | 0.0681 |
81
+ | 0.0708 | 0.56 | 3100 | 0.0684 |
82
+ | 0.072 | 0.58 | 3200 | 0.0681 |
83
+ | 0.0716 | 0.59 | 3300 | 0.0689 |
84
+ | 0.0723 | 0.61 | 3400 | 0.0678 |
85
+ | 0.0678 | 0.63 | 3500 | 0.0676 |
86
+ | 0.0695 | 0.65 | 3600 | 0.0672 |
87
+ | 0.0689 | 0.67 | 3700 | 0.0676 |
88
+ | 0.0716 | 0.68 | 3800 | 0.0671 |
89
+ | 0.07 | 0.7 | 3900 | 0.0667 |
90
+ | 0.0683 | 0.72 | 4000 | 0.0665 |
91
+ | 0.0704 | 0.74 | 4100 | 0.0664 |
92
+ | 0.0702 | 0.76 | 4200 | 0.0665 |
93
+ | 0.0678 | 0.77 | 4300 | 0.0662 |
94
+ | 0.0679 | 0.79 | 4400 | 0.0661 |
95
+ | 0.069 | 0.81 | 4500 | 0.0660 |
96
+ | 0.0675 | 0.83 | 4600 | 0.0661 |
97
+ | 0.0682 | 0.85 | 4700 | 0.0660 |
98
+ | 0.0697 | 0.86 | 4800 | 0.0659 |
99
+ | 0.0689 | 0.88 | 4900 | 0.0658 |
100
+ | 0.0665 | 0.9 | 5000 | 0.0658 |
101
+ | 0.067 | 0.92 | 5100 | 0.0657 |
102
+ | 0.0666 | 0.94 | 5200 | 0.0657 |
103
+ | 0.0704 | 0.95 | 5300 | 0.0656 |
104
+ | 0.0682 | 0.97 | 5400 | 0.0656 |
105
+ | 0.0663 | 0.99 | 5500 | 0.0656 |
106
+
107
+
108
+ ### Framework versions
109
+
110
+ - Transformers 4.37.0.dev0
111
+ - Pytorch 2.1.2+cu121
112
+ - Datasets 2.15.0
113
+ - Tokenizers 0.15.0
114
+ ## Training procedure
115
+
116
+
117
+ The following `bitsandbytes` quantization config was used during training:
118
+ - quant_method: QuantizationMethod.BITS_AND_BYTES
119
+ - load_in_8bit: False
120
+ - load_in_4bit: True
121
+ - llm_int8_threshold: 6.0
122
+ - llm_int8_skip_modules: None
123
+ - llm_int8_enable_fp32_cpu_offload: False
124
+ - llm_int8_has_fp16_weight: False
125
+ - bnb_4bit_quant_type: nf4
126
+ - bnb_4bit_use_double_quant: True
127
+ - bnb_4bit_compute_dtype: bfloat16
128
+
129
+ ### Framework versions
130
+
131
+
132
+ - PEFT 0.6.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:39cddcf40c168b4002df32aae3940301b5d282e51807fed954d91c252f0474cd
3
  size 125248064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5dc377564c2bce602ff1fc39ef8c570b230246b432cf2641f841cf14bf35ccb
3
  size 125248064